The macular hole is a full-thickness defect in the macula, the pinprick-sized patch of retinal tissue responsible for sharp central vision. When it forms, patients typically notice distortion, a dark or missing spot in the middle of their visual field, and progressive loss of reading ability. Optical coherence tomography, or OCT, is the diagnostic workhorse: it uses low-coherence light to produce cross-sectional images of the retina with micrometer-scale resolution, letting clinicians see the hole’s edges, the lifted cuff of surrounding tissue, and the state of the underlying choroid. From these scans, ophthalmologists extract a handful of measurements — minimum hole diameter, base diameter, hole height on the left and right sides, and central choroidal thickness — that together inform staging and surgical strategy.

The problem is that these measurements are slow, subjective, and vulnerable to inter-observer variability. Two experienced graders can produce different numbers from the same scan, and in a busy clinic the differences can matter: hole size and configuration influence whether a surgeon expects a high closure rate with a simple gas tamponade or needs more extensive vitreoretinal surgery. Automating the extraction of these parameters has therefore been a long-standing goal, but medical images pose distinctive challenges for deep learning. OCT scans are grayscale, low in contrast compared with natural photographs, and dominated by layered textures that differ radically from the colorful, object-rich images that standard detection networks were designed to handle.

The research team, led by Zhiyuan Zhao, Xinqi Yu, Xiaochun Wang, Bin Wu, and Sheng Zhou from the Institute of Biomedical Engineering of the Chinese Academy of Medical Sciences and Peking Union Medical College together with Tianjin Eye Hospital, tackled these challenges with a fusion architecture. LMMR-YOLO — short for Lightweight Model of Mask R-CNN-YOLO — combines two of the most influential families in computer vision. Mask R-CNN, a two-stage detector prized for its precise instance segmentation, was used to expand the training dataset in a targeted way and to contribute fine-grained localization. YOLOv12-obb, a modern single-stage detector oriented around rotated bounding boxes, supplied the speed and efficiency backbone. The fusion means the system can both delineate the lesion accurately and detect it fast enough for routine use.

Several technical innovations underpin the model’s performance on grayscale medical data. The researchers integrated a grayscale image adaptation module with Efficient Channel Attention, a mechanism that lets the network learn which feature channels carry the most diagnostic information — essentially teaching it where to look in the layered structure of a retinal scan. They replaced heavy backbone components with C2f_Ghost and A2C2f_Lite modules, lightweight building blocks that preserve feature extraction power while slashing computational cost. An Oriented Bounding Box detection head allows the model to fit tightly around lesions that appear at arbitrary angles in the scan, and an Image Enhancement Algorithm sharpens contrast before detection, with a Fusion Detection Algorithm merging the outputs of the enhanced pipeline and the Mask R-CNN branch into a single verdict.

The study was retrospective, drawing on 606 OCT images from 61 patients, with ethics approval from Tianjin Eye Hospital and a waiver of individual informed consent because the data were anonymized. Performance was evaluated with the standard battery of detection metrics: precision, recall, mean average precision at an intersection-over-union threshold of 0.5, F1-score, and accuracy. On the internal test set, LMMR-YOLO achieved an mAP@0.5 of 92.69 percent and an overall accuracy of 94.78 percent for key-point detection of macular hole parameters. Precision reached 97.14 percent, meaning that when the model flags a measurement point, it is almost always correct, while recall of 87.88 percent indicates it catches the large majority of true points. The harmonic F1-score of 92.28 percent balances those two figures and sits comfortably in the range clinicians would expect from careful human graders.

Two secondary results are particularly telling about the model’s reliability. Normalized classification accuracy for central choroidal thickness — a parameter that reflects the health of the vascular layer beneath the retina and is increasingly used in staging — came out at 0.97. And the mirror misidentification rate between left hole height and right hole height, an error mode in which a system confuses the two sides of an asymmetric hole, was just 0.03. Those numbers matter because staging systems for macular holes depend on exactly such lateral distinctions; a model that routinely swapped left and right would be clinically useless no matter how impressive its headline accuracy.

Equally important is what the model does not cost. LMMR-YOLO carries only 3.34 million trainable parameters and requires 11.10 gigafloating-point operations per inference — figures that place it firmly in the lightweight category, small enough to contemplate running on standard clinic hardware or even portable devices rather than server farms. Inference speed measured 9.44 frames per second, approaching real-time performance, and the training time was short. In medical AI, where many celebrated models demand billions of parameters and hours of GPU time per scan batch, a system that balances accuracy against efficiency is arguably more valuable than a marginally more accurate but computationally prohibitive alternative. The authors frame this balance as a new paradigm for medical grayscale image analysis and model training.

The clinical implications extend beyond the numbers. Accurate, automated extraction of minimum hole diameter, base diameter, hole height, and choroidal thickness could standardize staging of full-thickness and lamellar macular holes, reduce grading variability between centers, and support personalized surgical planning — for example, helping surgeons anticipate closure likelihood and choose among operative techniques. Because the pipeline is fast and light, it could also be embedded in screening workflows, flagging urgent cases in high-volume imaging services where specialist reading time is scarce. The researchers suggest the approach supports ophthalmic diagnosis, staging, and personalized care, and its architecture — grayscale adaptation, channel attention, oriented detection, and fusion of segmentation and detection branches — is general enough to be adapted to other retinal and medical imaging tasks.

Caveats remain, as they do for any retrospective single-center study. The dataset of 606 images from 61 patients, while substantial for a rare lesion, will need validation on larger and more diverse cohorts before the model can be trusted across different scanner manufacturers, image qualities, and patient populations. The published version is an early-release, peer-reviewed accepted manuscript that will undergo further editorial refinement. Still, the trajectory is clear: the fusion of segmentation precision with detection speed, tuned specifically for the grayscale, low-contrast world of OCT, has produced a system that measures the anatomy of a sight-threatening lesion nearly as well as experts do — and does it in milliseconds. For the millions of people at risk of macular holes worldwide, that speed and consistency could eventually translate into earlier detection, better-planned surgery, and more preserved vision.

Subject of Research: Deep learning-based automatic detection and measurement of macular hole parameters in optical coherence tomography images

Article Title: A deep learning-based study on automatic extraction and measurement of macular hole parameters

Article References: Zhao, Z., Yu, X., Wang, X., Lu, K., Wu, B., & Zhou, S. (2026). A deep learning-based study on automatic extraction and measurement of macular hole parameters. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02899-8

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02899-8

Keywords: macular hole, deep learning, optical coherence tomography, LMMR-YOLO, Mask R-CNN, YOLOv12, medical imaging, ophthalmology, retinal disease, image segmentation, lightweight neural network, computer-aided diagnosis

The eye is, in many respects, an ideal organ for aging research. It is paired, which allows within-person comparisons; it can be imaged repeatedly and non-invasively over decades; it can be treated locally with injections, lasers, and surgery; and its function can be quantified with extraordinary precision, from visual acuity to contrast sensitivity to dark adaptation. This tractability has fueled an explosion of models that estimate biological age from retinal images and other ocular data. Deep learning applied to retinal photographs has been reported to predict biological age and stratify morbidity and mortality risk. Multimodal retinal aging clocks built on optical coherence tomography and fundus imaging have been proposed for systemic health assessment. Liquid-biopsy proteomics of ocular fluids, combined with machine learning, has been used to identify cellular drivers of eye aging in vivo. Each of these approaches produces a number expressed in years, but the review emphasizes that the familiar unit conceals fundamental differences in what was measured and what the model was built to predict.

To untangle these differences, Bair proposes an eye-specific framework organized around five declarations that should accompany any ocular age estimate. The first is the proximal source: which biological material or signal the estimate derives from, whether a two-dimensional fundus image, a three-dimensional OCT scan, tear fluid or aqueous humor proteomics, or systemic omics that only indirectly reflect the eye. The second is the development target: what the model was actually trained to predict, which may be chronological age, a clinical outcome, or some composite index. The third is the reported quantity: whether the output is an absolute age, a residual difference from chronological age, an age-adjusted deviation, or a risk score. The fourth is the evidence design: whether the supporting study was cross-sectional, longitudinal, or interventional. The fifth is the intended use: whether the estimate is meant for mechanistic research, risk stratification, monitoring of an intervention, or clinical decision-making. Only when all five are stated explicitly, the review argues, can readers compare estimates across studies or translate them into practice.

Beyond these five declarations, the framework identifies several annotations that remain essential for any meaningful interpretation. Laterality matters, because the two eyes of one person are correlated and statistical approaches that treat eyes as independent observations inflate certainty; ophthalmology has long wrestled with this people-and-eyes problem. The reference population matters, because a model trained on one demographic or ethnic group may not calibrate to another. Disease and treatment context matters, since conditions such as diabetes, glaucoma, and age-related macular degeneration, and treatments such as intravitreal injections, alter the very features the models read. The technical version of the model matters, because updated algorithms can shift outputs without any biological change in the patient. And uncertainty in the stated output matters, because a single point estimate in years carries no information about its own reliability.

Perhaps the most consequential distinction the review draws is between a cross-sectional residual and a longitudinal aging rate. When a model estimates that a 60-year-old’s retina looks like that of a typical 66-year-old, the resulting six-year gap is a cross-sectional residual: a snapshot comparison against a reference population at one moment in time. It does not follow that the person is aging six years faster than everyone else. Repeated application of a cross-sectional model to the same individual over time does not establish an individual aging rate, because the model was never designed or validated to track change within a person. Longitudinal claims require repeated measurements evaluated against a longitudinal reference, and the observed change must be distinguishable from measurement variability. The review points to reference change values, a concept from clinical chemistry, as the appropriate tool for deciding whether a difference between two measurements on the same person exceeds the noise inherent in the method.

This confusion between cross-sectional and longitudinal inference is not unique to the eye. The brain-age literature, which pioneered the estimation of biological age from magnetic resonance imaging, has documented similar pitfalls, including regression dilution and the statistical traps that arise when age gaps are used as predictors of outcomes. The review imports lessons from that field, along with normative-modeling approaches that conceptualize disease as a deviation from expected trajectories rather than a categorical state. It also draws on the formal biomarker validation literature, including the framework that distinguishes discovery, analytical validation, clinical validation, and clinical utility, and on criteria for surrogate endpoints that specify when a biomarker can stand in for a clinical outcome in trials. By these standards, most ocular age estimates today sit firmly at the discovery stage.

The review further insists that visual-aging claims need validation against function, not just against images. The visual system offers psychophysical and performance-based measures, such as contrast sensitivity, rod-mediated dark adaptation, and hazard detection in night-driving simulators, as well as patient-reported outcomes. Delayed dark adaptation, for example, has been established as a functional biomarker for incident early age-related macular degeneration, and natural-history studies are now charting how visual function changes over years in normal aging and in early disease. An ocular age estimate that correlates with none of these functional measures risks being a statistical artifact rather than a biological signal. Differences across measurement domains, whether between an imaging-based clock and a proteomic one, must be assessed both against measurement error and against the distribution expected in the reference population, so that a difference is not declared meaningful simply because it is statistically detectable.

Even when a persistent difference survives these scrutiny layers, the review counsels restraint in interpretation. A residual age gap may support a candidate biological pattern only after technical explanations, such as image quality, camera differences, and algorithm version, and clinical explanations, such as occult disease and medication effects, have been evaluated and excluded. Only then does the gap become a hypothesis about retinal aging that can be tested in appropriate designs. Prognostic claims, in turn, require separate prospective evidence demonstrating that the estimate predicts outcomes in new samples, and clinical claims require decision-level evidence showing that acting on the estimate improves patient-relevant results. Tools such as decision curve analysis, which quantifies the net benefit of acting on a prediction at various threshold probabilities, and reporting standards such as TRIPOD+AI and its risk-of-bias companion PROBAST+AI, provide the methodological scaffolding for this progression from prediction to practice.

The practical upshot for clinicians, researchers, and the growing consumer market for biological-age tests is a checklist of questions to ask before taking an ocular age number at face value. What tissue or signal was measured? What was the model trained to predict, and against what reference population? Is the reported quantity an absolute age, a residual, or a calibrated deviation, and what is its uncertainty? Was the supporting evidence cross-sectional or longitudinal, and if longitudinal, was the change larger than measurement noise? Has the estimate been validated against visual function or patient-reported outcomes? And has any study shown that using the estimate changes decisions in ways that help patients? Until these questions are answered, the review concludes, a cross-sectional age residual should not be mistaken for an aging rate, a prognostic marker, or a clinically actionable test. The eye may indeed hold a clock, but reading it responsibly requires knowing what kind of clock it is, what it was built to measure, and what evidence stands behind each tick.

Subject of Research: Interpretation and validation of biological age estimates derived from ocular imaging and biomarkers

Article Title: Interpreting ocular age estimates: source, output, evidence, and clinical meaning

Article References: Bair, H. (2026). Interpreting ocular age estimates: source, output, evidence, and clinical meaning. Biogerontology, 27(5), Article 175. https://doi.org/10.1007/s10522-026-10525-x

Image Credits: AI Generated

DOI: 10.1007/s10522-026-10525-x

Keywords: biological age, ocular aging, retinal imaging, fundus photography, optical coherence tomography, biomarkers, deep learning, prediction models, longitudinal studies, visual function, biogerontology, clinical validation

The research, led by Luca Bollo and Angela Vidal-Jordana at the Multiple Sclerosis Centre of Catalonia in collaboration with Vall d’Hebron University Hospital and published in the Journal of Neurology, prospectively followed 46 patients experiencing their first episode of unilateral subacute optic neuritis. Each participant underwent a comprehensive workup: detailed clinical examination, orbital and brain magnetic resonance imaging, optical coherence tomography of the retinal nerve fiber layer, laboratory testing including serum antibodies against aquaporin-4 and myelin oligodendrocyte glycoprotein, and visual evoked potentials, a neurophysiological test that measures how efficiently the visual pathway conducts signals from the retina to the visual cortex. The investigators then applied the original ICON 2022 criteria and compared the results with two modified versions of the framework.

The headline finding is stark. Under the original ICON 2022 criteria, only 20 of the 46 patients, or 43.5 percent, reached the category of definite optic neuritis. When the researchers loosened the clinical requirements, allowing a diagnosis with incomplete clinical features, and required two abnormal paraclinical tests among MRI, optical coherence tomography, and laboratory findings, the proportion of definite cases rose to 32 patients, or 69.6 percent, a difference that was highly statistically significant. When visual evoked potentials were added to the paraclinical toolkit alongside the less stringent clinical definitions, the figure climbed further still, to 39 patients, or 84.8 percent, nearly double the yield of the original criteria.

To understand why this matters, it helps to unpack what each test actually measures. Magnetic resonance imaging with orbital sequences can reveal T2-hyperintense signal and gadolinium enhancement along the optic nerve, direct evidence of inflammation and blood-nerve barrier disruption. Optical coherence tomography quantifies the thickness of the retinal nerve fiber layer and the ganglion cell complex, detecting the axonal loss that follows an inflammatory attack. Visual evoked potentials, by contrast, probe function rather than structure: a flashing or pattern-reversing checkerboard stimulus is presented to each eye while electrodes over the occipital cortex record the timing and amplitude of the brain’s response. Demyelination delays the P100 latency, the characteristic peak of the response, often before structural thinning becomes measurable on scans.

This functional sensitivity is precisely why the addition of VEPs proved so powerful in the study. Patients with incomplete clinical presentations, for example those without the classic combination of pain, visual acuity loss, color vision impairment, and a relative afferent pupillary defect, could still be classified as definite optic neuritis when their paraclinical workup was typical. The effect was most pronounced in exactly this subgroup: individuals whose symptoms did not tick every clinical box but whose MRI, OCT, and electrophysiology told an unambiguous story of optic nerve inflammation. In clinical terms, the modified criteria rescued a substantial fraction of patients who would otherwise have been left in diagnostic limbo, a category with real consequences for treatment decisions, follow-up intensity, and counseling about the risk of future demyelinating events such as multiple sclerosis.

The Barcelona group’s work does not stand in isolation. Since the ICON criteria appeared in The Lancet Neurology in 2022, several groups have tested them against independent cohorts and reported similar concerns about stringency. Correspondence in the same journal questioned how the criteria performed in atypical presentations, and the Acute Optic Neuritis Network published an application of the criteria in 2024 that highlighted classification gaps. Meanwhile, the field of demyelinating disease diagnostics has been moving rapidly toward incorporating the optic nerve more broadly. The 2024 revisions of the McDonald criteria for multiple sclerosis, the most consequential diagnostic framework in the field, now allow optic nerve involvement, demonstrated by MRI or by visual evoked potentials, to contribute to the demonstration of dissemination in space, a change that formally elevates VEPs from a niche neurophysiological tool to a first-line diagnostic instrument.

Against that backdrop, the new study offers prospective, patient-level evidence that the optic neuritis criteria should evolve in the same direction. The authors’ proposed modification is conceptually simple: rather than demanding a complete clinical syndrome, the framework would accept a less strict clinical category and then require corroboration from two abnormal paraclinical tests, with VEPs counted alongside MRI, OCT, and serology. This modular architecture mirrors how modern neurology increasingly thinks about diagnosis, as a Bayesian exercise in which each test contributes independent information about the probability of disease. A delayed P100 latency on VEPs carries information that is partly independent of a normal-appearing retinal nerve fiber layer on OCT, and both can be informative when the MRI shows only subtle optic nerve signal change.

There are important caveats. The cohort of 46 patients is modest, and all participants were recruited at specialized centers in Catalonia with access to expert neuroradiology, neuro-ophthalmology, and neurophysiology, so the modified criteria will need validation in larger and more diverse populations, including settings where antibody testing and VEP recording are less readily available. The study also focused on first-episode unilateral optic neuritis, leaving open the question of how the modified framework performs in bilateral, recurrent, or pediatric disease, where differential diagnoses such as neuromyelitis optica spectrum disorder and MOG antibody-associated disease loom larger and where serological markers carry decisive weight. The researchers themselves frame their findings as supporting revisions aimed at broader clinical applicability rather than as a finished replacement for the 2022 consensus.

Even so, the implications for patients are tangible. A person who wakes with a sore, blurry eye but whose examination findings are incomplete currently risks being labeled with uncertain optic neuropathy, delaying disease-modifying therapy if multiple sclerosis is the underlying cause, or delaying plasma exchange and immunosuppression if aquaporin-4 antibody disease is suspected. By formally recognizing that paraclinical evidence, especially electrophysiological evidence from visual evoked potentials, can compensate for an atypical clinical picture, the modified criteria promise faster, more confident classification and, downstream, better-targeted treatment. The study was funded by the Instituto de Salud Carlos III, and its datasets are available from the corresponding authors, an invitation for other centers to stress-test the approach.

The broader lesson reaches beyond one disease. Diagnostic criteria in neurology have historically leaned heavily on clinical acumen, with imaging and laboratory tests playing supporting roles. The trajectory now visible across multiple sclerosis, neuromyelitis optica, MOG antibody disease, and optic neuritis points toward criteria that fuse clinical features with structural imaging, retinal layer quantification, serology, and neurophysiology in a single probabilistic framework. The Barcelona study demonstrates concretely what that fusion can achieve: the proportion of patients confidently classified as having definite optic neuritis nearly doubled, from 43.5 to 84.8 percent, simply by trusting the objective signals that modern technology can already deliver. If subsequent multicenter validation confirms these results, the humble visual evoked potential, a test first described decades ago, may finally take its place at the center of optic nerve diagnostics, and fewer patients will be left waiting in the gray zone between suspicion and certainty.

Subject of Research: Prospective evaluation of the 2022 international consensus criteria for optic neuritis and the diagnostic value of visual evoked potentials

Article Title: Evaluation of the 2022 international consensus criteria for optic neuritis: the role of clinical features and visual evoked potentials

Article References: Bollo, L., Vidal-Jordana, A., Rodríguez-Acevedo, B., Sceppacuercia, S., Mongay-Ochoa, N., Zancan, V., Ajdinaj, P., Braga, N., Cabello, S., Alberich, M., Corral, J., Auger, C., Tintoré, M., Montalban, X., Pareto, D., Moncho, D., Rovira, À., & Sastre-Garriga, J. (2026). Evaluation of the 2022 international consensus criteria for optic neuritis: the role of clinical features and visual evoked potentials. Journal of Neurology, 273(10), Article 561. https://doi.org/10.1007/s00415-026-13972-1

Image Credits: AI Generated

DOI: 10.1007/s00415-026-13972-1

Keywords: optic neuritis, diagnostic criteria, visual evoked potentials, ICON criteria, MRI, optical coherence tomography, multiple sclerosis, demyelination, neurology, optic nerve, MOG antibodies, prospective study

Obsessive-compulsive disorder affects roughly one to two percent of the population and is characterized by intrusive, distressing thoughts and repetitive behaviors performed to neutralize the anxiety they provoke. For decades the condition was understood primarily through psychological and neural circuit models, but a growing body of research has implicated inflammatory and immunological mechanisms as well. Studies have reported altered levels of circulating cytokines, markers of oxidative stress, and immune cell activity in OCD, and some patients show worsening of symptoms after streptococcal infections, hinting that the immune system may participate in the disorder in ways that standard brain imaging cannot easily capture. The problem has always been measurement: directly sampling neuroinflammation in a living brain requires invasive procedures that are impossible to justify in routine research.

The cornea offers an elegant workaround. It is the most densely innervated tissue in the human body, supplied by thousands of slender nerve fibers that branch into a delicate mesh just beneath its surface, known as the subbasal nerve plexus. Critically, this living nerve network can be visualized noninvasively with corneal confocal microscopy, a technique that uses a low-power laser to capture high-resolution images of the cornea at the level of single cells. The method has already proven its worth in other fields: it is widely used to detect the small-fiber nerve damage of diabetic neuropathy, often before symptoms appear, and it has revealed corneal nerve loss in conditions ranging from multiple sclerosis to Parkinson’s disease and dementia. Alongside the nerves, the cornea also hosts resident immune sentinels called dendritic cells, which normally sit scattered and quiet but rapidly increase in number and sprout long processes when inflammation or nerve injury is present. Because the small fibers of the subbasal plexus share structural and functional similarities with the small nerve fibers found throughout the body, changes seen in the cornea are thought to mirror what is happening in the wider peripheral and possibly central nervous system.

In the new study, the researchers recruited twenty-nine patients with OCD who had never received psychiatric medication, deliberately excluding drug effects that can themselves alter immune and neural measures, and twenty-eight age- and sex-matched healthy controls. Each participant underwent a comprehensive ophthalmic examination, and the team applied two complementary imaging approaches. The first was enhanced-depth imaging optical coherence tomography, which uses light waves to measure structures at the back of the eye, specifically the choroid, the vascular layer beneath the retina that nourishes it. The researchers measured subfoveal choroidal thickness and calculated the choroidal vascularity index, a ratio that distinguishes the vessel-filled portion of the choroid from its supporting stromal tissue and has emerged in other studies as a marker of ocular and systemic inflammation. The second approach was corneal confocal microscopy, used to quantify the density of dendritic cells and a battery of subbasal nerve plexus parameters, including corneal nerve fiber density, branch density, trunk density, fiber length, branch length, and trunk length, together with measures of branching and tortuosity.

The choroidal results were largely unremarkable. Subfoveal choroidal thickness, which some earlier inflammation studies had flagged as potentially informative, did not differ between the OCD patients and controls, a null result the authors report with a p-value of 0.787. The choroidal vascularity index was statistically higher in the OCD group, but the absolute difference was small, suggesting that the choroid, at least as measured here, carries limited information about the disorder. A modest positive correlation between the index and obsession scores on the Yale-Brown Obsessive Compulsive Scale was observed, hinting that vascular changes might track one dimension of symptomatology, but the signal was weak.

The corneal findings were another matter entirely. Dendritic cell density in the cornea of OCD patients averaged 106.25 cells per square millimeter, compared with just 32.29 cells per square millimeter in the healthy controls, a more than threefold increase that was highly significant, with a p-value below 0.001. In parallel, every parameter of the subbasal nerve plexus was significantly reduced in the patients: fiber density, fiber length, and branching all fell well below control levels, each with p-values below 0.001. In other words, the eyes of medication-naive OCD patients showed the twin signatures of an activated immune system and diminished small-fiber nerve integrity, a pattern that in other diseases is interpreted as neuroinflammation or a neuroimmune response to nerve stress. Notably, because none of the patients had ever taken psychotropic medication, the alterations cannot be attributed to treatment, one of the most common confounds in biological psychiatry research.

Intriguingly, the corneal measures did not correlate with overall symptom severity on the Yale-Brown scale. That absence of a dose-response relationship cuts both ways. It suggests that the corneal changes are unlikely to serve as a simple severity gauge that rises and falls with how badly a patient feels at a given moment. But it also leaves open the possibility that the alterations mark a stable trait of the disorder, a biological vulnerability present regardless of current symptom load, rather than a state marker that fluctuates with episodes. Distinguishing trait from state will require exactly what the authors call for next: larger samples followed over time, ideally with scans repeated before and after treatment to see whether the nerve and immune signatures respond to clinical improvement.

Several cautions temper the excitement. The study is a case-control comparison of a modest number of participants, so confounders such as sleep quality, stress, body mass index, smoking, and subclinical dry eye, which the team partially addressed using ocular surface measures like the Ocular Surface Disease Index, tear breakup time, and exclusion criteria, could still influence corneal findings. Dendritic cells in the cornea are exquisitely sensitive to environmental factors, and anxiety itself, which is elevated in OCD, might modulate immune activity through stress hormone pathways. The authors themselves are explicit that the choroidal data argue against overinterpreting systemic inflammation and that the corneal results, however striking, require validation. There is also no evidence yet that the corneal nerve changes reflect anything happening inside the brain, only that the peripheral neuroimmune state of OCD patients differs measurably from that of healthy people.

Even with those caveats, the study opens a genuinely novel line of investigation. If corneal confocal microscopy can be validated in larger and longitudinal cohorts, it would give psychiatry something it sorely lacks: a rapid, inexpensive, radiation-free biomarker assay that requires only a desktop imaging device already found in most ophthalmology clinics. Such a tool could help stratify patients by underlying biology rather than symptom checklist, identify inflammatory subtypes of OCD that might respond to immunomodulatory approaches, and provide an objective readout in treatment trials. It would also reinforce a broader and increasingly persuasive idea in neuroscience, that the immune and nervous systems are intertwined partners in mental illness, and that the evidence of that partnership may be visible not deep inside the skull but in the transparent, exquisitely innervated window of the eye.

Subject of Research: Neuroimmune alterations in obsessive-compulsive disorder detected by corneal confocal microscopy

Article Title: Corneal confocal microscopy reveals neuroimmune alterations in drug-naïve patients with obsessive-compulsive disorder: a case-control study

Article References: Selvi, H. C., Atay, B., Arslanyürek, İ., Özmen, M. C., & Ekmekci, İ. (2026). Corneal confocal microscopy reveals neuroimmune alterations in drug-naïve patients with obsessive-compulsive disorder: a case-control study. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08757-9

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08757-9

Keywords: obsessive-compulsive disorder, corneal confocal microscopy, neuroinflammation, dendritic cells, corneal nerve fibers, optical coherence tomography, choroidal vascularity index, biomarker, psychiatry, small-fiber neuropathy, neuroimmune system, ophthalmic imaging

The brain care score, developed at the McCance Center for Brain Health, condenses twelve modifiable factors into a single number: four physical measures (blood pressure, hemoglobin A1c, cholesterol, and body mass index), five lifestyle behaviors (nutrition, alcohol intake, smoking, physical activity, and sleep), and three social-emotional dimensions (stress, relationships, and meaning in life). In the UK Biobank adaptation used here, the score ranges up to 19 points, with higher values indicating better brain care. Previous studies have linked higher scores to favorable neuroimaging markers and reduced risks of dementia, stroke, and late-life depression. What remained unknown was whether the score also tracks with the fine-grained architecture of the retina, an outgrowth of the central nervous system that shares vascular, metabolic, and inflammatory biology with the brain.

To find out, the team led by investigators at The Chinese University of Hong Kong analyzed optical coherence tomography (OCT) scans from 28,657 UK Biobank participants aged 40 to 69. OCT is a non-invasive imaging technique that uses light interference to map retinal layers at micrometer resolution, and the researchers examined eight distinct layers plus overall macular thickness. After rigorous quality control and statistical adjustment for age, sex, ethnicity, education, deprivation, cardiovascular disease history, refractive error, and intraocular pressure, a clear pattern emerged: each 5-point increase in the brain care score was associated with thicker retinal nerve fiber layer, thicker ganglion cell–inner plexiform layer, thicker photoreceptor segments, and greater average macular thickness.

The effect sizes are small in absolute terms—a few tenths of a micrometer per 5-point score increase, translating to roughly 1.2 percent difference in nerve fiber layer thickness and about half a percent for the ganglion cell layer relative to cohort means. The researchers are careful to stress that these are population-level structural differences, not differences an ophthalmologist could detect in a single patient or act on clinically. Yet the pattern was remarkably consistent. Participants in the lowest score quartile had significantly thinner layers across the board compared with those in the highest quartile, all trend tests were highly significant, and the findings held up in sensitivity analyses that accounted for nonlinear aging effects, restricted the sample to European participants, and replicated the associations using repeat OCT visits years later.

Perhaps the most provocative result concerns retinal biological age. The team trained a support vector machine on 60 OCT-derived structural metrics from thousands of healthy participants to predict chronological age from retinal structure alone, then defined an OCT age gap as the bias-corrected predicted age minus actual age. Positive values indicate a retina that looks older than it should. After full adjustment, each 5-point increase in brain care score was associated with a 0.326-year reduction in this retinal age gap, with the mean gap falling progressively from 0.39 years in the lowest score quartile to −0.08 years in the highest. In other words, people who take better care of their brains carry retinas that read as younger on a machine-learning clock.

Dissecting the score into its components revealed where the signal lives. The physical and lifestyle domains drove the associations, while the social-emotional domain showed no independent link to retinal structure. Favorable blood pressure, hemoglobin A1c, and body mass index were each tied to thicker specific layers and a smaller retinal age gap, with effects on the age gap ranging from about a quarter to nearly half a year. Non-smoking, moderate alcohol intake, regular aerobic exercise, and at least seven hours of nightly sleep each contributed modest reductions in retinal biological age. Interestingly, total cholesterol below 190 milligrams per deciliter was actually associated with a slightly larger retinal age gap, echoing a growing literature on the complex, sometimes U-shaped relationships between lipids and neurological health.

To probe the biology behind these correlations, the researchers turned to metabolomics. For a subset of 14,656 participants with nuclear magnetic resonance profiling of 249 plasma biomarkers, they ran exploratory mediation analyses asking whether circulating metabolites statistically carry part of the association between brain care score and retinal thickness. The candidate pathways that emerged centered on lipid metabolism, unsaturated fatty acids, and branched-chain amino acids. One principal component reflecting HDL-enriched lipid profiles mediated a small fraction of the association with nerve fiber and ganglion cell layer thickness, while another dominated by omega-3 polyunsaturated fatty acids showed positive indirect effects on photoreceptor layer thickness—a plausible finding given that docosahexaenoic acid, the dominant long-chain fatty acid in photoreceptor outer-segment membranes, is essential for photoreceptor function.

The authors are appropriately cautious about these mediation results. Because the brain care score, the metabolites, and the retinal measurements were all captured at the same point in time, no temporal ordering can be established, and the indirect-effect estimates cannot confirm genuine biological causation. The principal components themselves may be sample-dependent, and the proportion of the total effect explained by any single metabolic pattern was modest—ranging from under 3 percent to about 11 percent. Still, the convergence on lipid and amino acid pathways is biologically coherent: excess branched-chain amino acids have been implicated in oxidative stress and inflammation in animal models of diabetic retinopathy, and lipoprotein subclass composition has repeatedly surfaced in studies of age-related macular degeneration.

The broader significance of the study lies in what it says about the retina as a sentinel organ. Thinner retinal nerve fiber and ganglion cell layers have previously been linked to cognitive decline, reduced brain volumes, Alzheimer’s disease, and incident dementia, and photoreceptor thinning has been associated with morbidity and mortality. By showing that a practical, modifiable brain health index tracks with both retinal structure and a machine-learned retinal age, the findings reinforce the idea that the same vascular and metabolic forces that shape brain aging also leave fingerprints in the eye—fingerprints that a routine OCT scan, already common in eye clinics, can capture.

Limitations remain, and the researchers enumerate them candidly. The observational design leaves room for residual confounding, the cohort’s healthier-than-average volunteers and predominantly European ancestry may limit generalizability, and the OCT-based age clock correlates only moderately with chronological age, meaning it captures just one facet of retinal aging. The brain care score itself is still a prototype awaiting systematic refinement. What the study delivers is a hypothesis-generating map: a demonstration that brain care and retinal health travel together, that specific metabolic pathways plausibly connect them, and that longitudinal studies with repeated measurements should now test whether improving one’s brain care score actually slows the thinning of the retina—and, by extension, perhaps the aging of the brain behind it.

Subject of Research: Associations between the brain care score and retinal layer thickness and retinal biological age in the UK Biobank

Article Title: Brain care score and retinal health: structural and metabolic insights from the UK Biobank

Article References: Yu, J., Zhang, Y., Gao, Y. L., Ho, M., Kam, K. W., Gong, B., Young, A. L., Pang, C. P., Tham, C. C., Yam, J. C., & Chen, L. J. (2026). Brain care score and retinal health: structural and metabolic insights from the UK Biobank. GeroScience. https://doi.org/10.1007/s11357-026-02579-z

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02579-z

Keywords: brain care score, retina, optical coherence tomography, UK Biobank, retinal biological age, metabolomics, macular thickness, neurodegeneration, lipid metabolism, omega-3 fatty acids, GeroScience, brain health

The team, led by ophthalmology researchers at the Eye Institute of Fudan University in Shanghai, quantified systemic aging using three complementary instruments. The first was PhenoAge acceleration, a measure derived from clinical biomarkers that estimates how much faster a person’s physiology is aging compared with their chronological peers. The second was age acceleration calculated by the Klemera-Doubal method, another biomarker-based biological age algorithm with a long track record in aging research. The third was a frailty index, a cumulative score built from deficits across health domains that captures the loss of physiological reserve characteristic of advanced aging. Each of these measures tells a slightly different story about aging, and the researchers wanted to know whether all of them converged on the same retinal signature.

That signature turned out to be strikingly consistent. Across all three aging metrics, higher biological age acceleration and greater frailty were associated with reduced macular thickness, with the effect spanning every inner retinal subfield measured by optical coherence tomography. The standardized effect sizes ranged from −1.386 to −0.421, and all associations remained highly significant after correction for multiple comparisons, with false discovery rate adjusted P values below 0.001. In practical terms, people whose bodies were aging faster than their birthdays implied tended to have measurably thinner retinas, and the relationship held whether aging was defined by molecular biomarkers, composite algorithms, or the clinical syndrome of frailty.

Optical coherence tomography, the imaging technology behind these measurements, deserves a moment of explanation. Originally described in Science in 1991, OCT uses low-coherence interferometry to generate cross-sectional images of biological tissue with micrometer-scale resolution. In the UK Biobank, tens of thousands of participants underwent retinal OCT scanning, producing an enormous, standardized dataset of retinal layer thicknesses. Because the retina is embryologically part of the central nervous system and shares vascular and metabolic characteristics with the brain, changes in its structure have been proposed as noninvasive biomarkers for neurological and systemic disease. Previous work has linked retinal thinning to cardiovascular risk, Alzheimer disease, and early age-related macular degeneration, but the broader question of how whole-body aging states map onto retinal structure had remained poorly characterized.

The most technically ambitious part of the new study involved the plasma metabolome. Blood samples collected from participants between 2006 and 2010 had been profiled using nuclear magnetic resonance spectroscopy, a platform that quantifies hundreds of circulating metabolites, including lipoprotein subclasses, fatty acids, amino acids, and glycolysis-related markers. From this high-dimensional data, the researchers derived metabolic signatures of aging using elastic net regression, a machine learning technique that selects sparse, predictive combinations of variables while guarding against overfitting. The resulting metabolomic aging signatures were then tested as statistical mediators of the relationship between biological aging and retinal thinning.

The mediation results were the study’s headline finding. Metabolic signatures accounted for 81.14 percent of the association between PhenoAge acceleration and macular thickness, 19.92 percent of the association for Klemera-Doubal method age acceleration, and 32.27 percent of the association for the frailty index, with all mediation effects significant after FDR correction. These proportions are remarkable, particularly for PhenoAge, and they suggest that circulating metabolites are not merely passive bystanders in the aging process but plausible conduits through which systemic senescence reaches the neurosensory retina. The retina is among the most metabolically demanding tissues in the body, with photoreceptors and the retinal pigment epithelium locked in a tightly coupled metabolic ecosystem that depends on glucose, lactate shuttling, and mitochondrial oxidative metabolism. Disruption of systemic metabolic homeostasis, the authors argue, is therefore well positioned to leave structural fingerprints in retinal tissue.

The study also incorporated the exposome, the concept introduced by cancer epidemiologist Christopher Wild in 2012 to describe the totality of environmental exposures an individual experiences across a lifetime. The researchers assembled exposome factors from phenotypic data covering lifestyle, diet, air pollution, and mental health. Their integrative analyses showed that adverse exposome profiles, including tobacco exposure, poor diet, air pollution, and negative psychosocial states such as loneliness and depression, were each correlated with reduced macular thickness. More importantly, systemic aging and metabolic dysregulation emerged as significant statistical intermediaries within these multidimensional pathways, meaning that environmental burdens appear to translate into retinal change at least partly by accelerating biological aging and reshaping the circulating metabolome.

This framing has implications that extend well beyond ophthalmology. If the retina reflects the convergence of environmental stress, metabolic dysregulation, and biological aging, then a routine retinal scan could in principle serve as a rapid, noninvasive readout of an individual’s cumulative aging trajectory. The findings align with a growing body of work on retinal oculomics, including phenome-wide analyses of UK Biobank OCT images that have linked ocular measurements to systemic health, and epidemiological studies showing that ambient air pollution is associated with retinal thinning and age-related macular degeneration. The new study unifies these threads by proposing an explicit causal architecture in which exposures act on aging biology, aging biology acts on metabolism, and metabolism acts on the retina.

The authors are careful about what their statistics can and cannot show. Mediation analysis in observational data identifies statistical intermediaries, not proven causal mechanisms, and the cross-sectional design of the UK Biobank baseline assessments means that temporal ordering cannot be fully established. The metabolomic platform used, while comprehensive for lipids and small molecules, does not capture every biologically relevant compound, and the elastic net signatures are predictive composites rather than single causal metabolites. Residual confounding by socioeconomic factors, which shape both exposome and health outcomes, remains a persistent challenge in cohort studies of this kind. Nevertheless, the sheer scale of the cohort, the consistency of the associations across three independent aging metrics, and the rigorous multiple-comparison correction lend considerable weight to the central conclusion.

For the aging research community, the study adds the retina to the growing list of organs whose structural integrity tracks systemic biological age, and it elevates plasma metabolism to the status of a key correlate of neurosensory retinal health. For clinicians, it hints at a future in which retinal imaging, already fast and inexpensive, might help identify people whose bodies are aging faster than their years, potentially guiding earlier interventions on smoking, diet, air quality, and psychosocial wellbeing. And for the public, the message is a vivid one: the same exposures that wear down the heart, the brain, and the metabolism may also be quietly etched into the tissue that lets you read this page. The eye, it seems, does not only take in the world; it keeps a record of what the world has done to us.

Subject of Research: Associations between systemic biological aging, frailty, exposome factors, plasma metabolomics, and retinal structural changes in the UK Biobank

Article Title: Association of systemic aging and frailty with retinal alterations: insights from an integrated exposome and metabolome framework

Article References: Chen, T., Wang, D., Ma, Y., Ye, Y., Wang, X., Lei, Y., Zhou, X., & Zhao, J. (2026). Association of systemic aging and frailty with retinal alterations: insights from an integrated exposome and metabolome framework. GeroScience. https://doi.org/10.1007/s11357-026-02572-6

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02572-6

Keywords: retinal oculomics, biological aging, frailty, plasma metabolomics, exposome, UK Biobank, macular thickness, optical coherence tomography, PhenoAge, mediation analysis, GeroScience, aging biomarkers

Soft tissue sarcomas are malignant, locally invasive tumors arising from mesenchymal cells, and they are among the most common cancers in dogs. Sarcomas account for roughly 10 to 15 percent of malignant tumors in dogs, and about 80 percent of those are soft tissue sarcomas rather than bone tumors. The primary treatment is surgical removal, and the success of local control hinges on whether histological assessment of the margins confirms complete excision. Positive margins leave cancerous cells behind, increasing the risk of local recurrence and the morbidity that comes with it. The stakes are therefore high for getting margin information quickly, ideally while the patient is still on the table.

Optical coherence tomography, or OCT, is the imaging technology at the heart of this effort. It uses near-infrared light to generate real-time, high-resolution images of tissue microstructure, in much the same way ultrasound uses sound waves but with micrometer-scale resolution. Traditional spectral-domain OCT builds images from the intensity of reflected light, revealing depth-resolved structural detail. The technology has been tested for margin assessment in human breast cancer, where it achieved sensitivities of 92 to 100 percent for detecting positive margins, and human clinical trials are underway. The Ohio State team, led by Yuanlong Wang, Laura E. Selmic, and Ping Zhang, has now extended this approach to companion animals with an important upgrade: polarization sensitivity.

Polarization-sensitive OCT, or PS-OCT, is a set of hardware and software extensions that track the polarization state of the light reflected from tissue. This adds contrast mechanisms that ordinary OCT cannot provide. The key property is birefringence, an optical signature that arises from the arrangement of subcellular collagen within tissue and reflects how organized that tissue is. When tissue is damaged, degenerated, or necrotic, its structure breaks down and its birefringence drops. Cancerous tissue, with its disrupted architecture, therefore looks measurably different from healthy fat or muscle under polarization contrast. In prior studies of human breast tissue, PS-OCT demonstrated both qualitative and quantitative differences between cancerous and normal tissue. The system used in the study, a Thorlabs Telesto PS-OCT with a 1300 nanometer central wavelength, 3.5 millimeter imaging depth, and 5.5 micrometer axial resolution in air, synchronously captures traditional OCT images and three polarization metrics: retardation, optic axis, and degree of polarization uniformity, known as DOPU.

The physics behind these metrics is elegant. Two cameras record the reflected light as complex numbers, from which total intensity is computed pixel-wise for the standard OCT image. Retardation measures the difference in optical path experienced by two orthogonal linearly polarized states, calculated as an angle that reflects the ratio of irradiances in each polarization channel. The optic axis describes the orientation of the tissue’s birefringent axis within the plane transverse to the beam, derived from the Stokes parameters that fully characterize the polarization of light. DOPU quantifies the uniformity of polarization within a pixel neighborhood, ranging from 0 to 1, and serves as a regularized measure of how orderly the tissue’s polarization response is. Together with the intensity image, these four channels provide a far richer description of tissue than intensity alone, and the study set out to prove that a neural network could exploit that richness.

The researchers enrolled 48 canine soft tissue sarcoma specimens under an Institutional Animal Care and Use Committee approved protocol, ultimately analyzing 40 after excluding eight tumors that turned out not to be sarcomas. Board-certified veterinary surgeons excised the tumors with margins chosen purely on clinical grounds, and each specimen was wrapped in saline-soaked gauze to prevent drying before imaging. The team scanned the entire surgical margin in B-mode, continuously sweeping the tissue, and captured paired OCT and PS-OCT frames. An expert reviewed image quality, and a pathologist’s evaluation of corresponding histopathology sections provided the gold standard tissue labels. From 140 image pairs, the team cropped 1,553 patches using a sliding window with a 50-pixel stride, resizing each to 224 by 224 pixels to fit the ResNet50 backbone, a convolutional neural network architecture widely used in medical imaging.

The heart of the technical contribution lies in how the four image channels are combined. The team tested two fusion strategies. Early fusion simply concatenates all four images along the channel dimension and feeds them into a single ResNet50 backbone. Joint fusion instead runs four separate backbones, one per metric, and merges their learned feature vectors through a learnable weighted sum before a linear classifier makes the final cancer-versus-normal call. The joint fusion model won decisively, though at the cost of four times the parameters, longer training, and greater vulnerability to overfitting, a trade-off the authors describe explicitly between performance and computational complexity. Training used a 70-15-15 split of train, validation, and test data, with dogs kept whole within a single split to prevent patient overlap, five-fold cross-validation for hyperparameter tuning, random horizontal flips for augmentation, and early stopping against the validation set. Performance was assessed with AUROC, area under the precision-recall curve, F1 score, precision, recall, and accuracy, with uncertainty estimated by bootstrapping the test set 1,000 times.

The results were unambiguous. Both fusion strategies substantially outperformed a baseline model trained solely on traditional OCT images, which the authors attribute to the complementary polarization information. Adding PS-OCT metrics improved AUROC by up to 0.15 for cancerous image classification, and the more polarization metrics included, the greater the gain. The final joint fusion model reached 0.989 AUROC and 91 percent accuracy in detecting positive margins. Notably, the OCT-only baseline achieved the highest recall but at a punishing cost to precision, illustrating why intensity alone is insufficient for reliable intraoperative decisions. Ablation experiments with partial inputs, pairing OCT with just one PS-OCT metric at a time, confirmed that each polarization channel contributes, and that the full multimodal model is the strongest performer on the threshold-agnostic metrics that best reflect underlying discriminative power.

Perhaps the most clinically compelling feature is the diagnostic curve. Rather than simply reporting whether an image contains cancer, the model slides a fine window with a 5-pixel stride across the original image, computes the cancer probability for each patch, and aggregates overlapping predictions into a one-dimensional curve showing the probability of cancer at every horizontal position. In case studies, the curve stayed flat for pure cancerous and pure normal images, and for a mixed image containing tumor in roughly one third of the frame, it rose sharply over the cancerous region and tapered gradually across the margin into fat. This effectively performs a one-dimensional segmentation of the tumor, giving surgeons a map of where cancerous tissue lies rather than a bare yes-or-no verdict, and the window size and aggregation method remain adjustable to clinical preference.

The authors are candid about limitations. The training set is relatively small, and convolutional networks depend heavily on the comprehensiveness of their data, so broader validation is needed before generalizability can be trusted. Raw OCT images carry optical artifacts, motion blur, and noise whose effects on the model remain unquantified, and the fusion strategies explored were deliberately simple, leaving higher-order interactions between polarization metrics unexploited. The model is also far from clinical deployment; explainability, integration into surgical workflows, and careful validation of its effect on surgical outcomes all remain ahead. Still, the trajectory is clear. By teaching a neural network to read the polarization fingerprints that cancer leaves in collagen, this work moves real-time, AI-assisted margin assessment from a promising concept toward a practical tool, one that could spare dogs a second surgery and, if the companion-animal findings translate, may one day help human surgeons answer that oldest of operating room questions with far greater confidence.

Subject of Research: Deep learning-based intraoperative surgical margin assessment for canine soft tissue sarcoma using polarization-sensitive optical coherence tomography

Article Title: Deep learning-assisted surgical margin assessment for canine soft tissue sarcoma based on polarization-sensitive optical coherence tomography

Article References: Wang, Y., Selmic, L. E., & Zhang, P. (2025). Deep learning-assisted surgical margin assessment for canine soft tissue sarcoma based on polarization-sensitive optical coherence tomography. Veterinary Oncology, 2(1), Article 17. https://doi.org/10.1186/s44356-025-00032-5

Image Credits: AI Generated

DOI: 10.1186/s44356-025-00032-5

Keywords: deep learning, polarization-sensitive OCT, optical coherence tomography, soft tissue sarcoma, surgical margins, veterinary oncology, canine cancer, convolutional neural networks, birefringence, cancer imaging, intraoperative diagnosis, ResNet50

To understand why this matters, it helps to consider what OCT actually does. The technique works a bit like ultrasound, except it uses near-infrared light instead of sound. A broadband or rapidly swept laser illuminates the sample, and an interferometer compares the weak light reflected from different depths within the sample against a reference path. By analyzing the interference pattern, the system reconstructs a depth profile with micrometer-scale axial resolution. Sweeping the beam across the surface builds up cross-sectional images, and stacking those slices yields full three-dimensional volumes. Because the method is entirely contactless and relies on low-power light rather than ionizing radiation, it can inspect delicate, soft, or transparent materials without deforming or damaging them, a property that becomes invaluable when the object under test is a photoresist layer that a physical probe would smear.

The catch has always been the hardware. A laboratory OCT system typically sprawls across an optical table: a bulky swept laser, discrete fiber couplers, a reference arm with moving parts or modulators, balanced photodetectors, and a maze of alignment-critical connections. That complexity translates into cost, fragility, and maintenance overhead, which explains why OCT has penetrated hospitals far more successfully than production lines. Photonic integrated circuits offer a way out, promising to fold the interferometer, the detectors, and eventually the light source onto chips that can be fabricated by the thousands in semiconductor foundries. But the field has struggled with two persistent problems: choosing a material platform that foundries can manufacture reliably at low optical loss, and designing a flexible sample arm so the chip-bound engine can still reach real-world objects.

The Shanghai Jiao Tong team, led by corresponding author Xingchen Ji, attacked both problems simultaneously. Their chip integrates low-loss silicon nitride interferometers with germanium photodiodes on a commercially fabricated silicon-based platform, combining the distinct strengths of multiple photonic integration technologies in a single device. Silicon nitride is prized in photonics for its extremely low propagation loss and broad spectral transparency, making it ideal for the interferometric heart of an OCT engine. Germanium, meanwhile, absorbs near-infrared light efficiently and serves as the natural photodetector material on silicon. The challenge is that these waveguide platforms have different geometries and optical modes, so the researchers developed an interlayer coupling structure to transfer light between the silicon nitride and silicon layers with a coupling loss of less than 0.15 decibels, a figure low enough to preserve the interference contrast on which OCT sensitivity depends.

The result is a remarkably dense consolidation of optical function. The interferometer and photodetectors occupy a footprint of just 0.9 by 0.4 millimeters, yet the design deliberately keeps the sample arm external, connected through fiber. That architectural choice preserves flexibility: the same chip-based engine can drive different probes depending on the inspection task. Ji emphasized the intent behind the design, saying the aim was to create a compact OCT module adaptable to practical industrial inspections. Integrating the interferometer and photodetectors on-chip reduces the number of discrete optical components, he noted, while the fiber-connected sample arm significantly expands the system’s operational range. In other words, the chip handles the precision optics, and the fiber handles the reach.

Performance figures from the experimental testing suggest the miniaturization did not come at the expense of capability. The system achieved 87 decibels of sensitivity, a measure of its ability to detect weakly reflected light from deep within a sample, along with a 3.42-millimeter sensitivity roll-off range, meaning usable signal persists over a respectable imaging depth. The team resolved internal structures within a smartphone camera module, demonstrating that the chip can inspect the kind of densely packaged, multi-layer consumer electronics that are notoriously difficult to verify non-destructively. Thickness measurements of a silica wafer, compared against established contact-based methods, showed absolute errors below 2 micrometers, while measurements of soft SU-8 photoresist blocks yielded relative errors under 0.12 percent. For semiconductor and microfabrication workflows, where film thicknesses are specified to sub-micrometer tolerances, that level of agreement with reference techniques is the difference between a laboratory curiosity and a production tool.

Perhaps the most striking demonstration involved a probe barely wider than a human hair is thick. By connecting an angle-polished ball-lens fiber probe with a diameter of just 450 micrometers, the researchers captured three-dimensional images of threads hidden inside a metal optical post and of internal layers within a roll of plastic tape. The ball lens focuses the emitted light to a tight spot while the angled polish suppresses back-reflections that would otherwise swamp the weak signal from the sample. This plug-and-play attachment effectively turns the chip-based engine into a miniature endoscope for machines, capable of inspecting bores, threads, and sealed assemblies that no conventional microscope objective can physically access.

Hang Su, the study’s first author and a doctoral student at Shanghai Jiao Tong University, highlighted why contactless operation matters so much in practice. Contactless measurement is invaluable when physical contact could deform a soft surface, Su explained, and the fiber probe allows the team to examine structures that are inaccessible to conventional microscopes. Describing the platform as a plug-and-play OCT module, Su said it paves the way for the next generation of chip-scale OCT engines. That framing captures the broader vision: rather than building a monolithic instrument for one task, the researchers envision a standardized photonic core to which different probes and sample arms can be attached as easily as swapping a lens on a camera.

The commercial implications extend beyond the demonstrations themselves. Because the chip was fabricated on a commercially available silicon photonics platform, the design is compatible with foundry-scale production, which is the prerequisite for driving down OCT system costs to levels that mid-sized manufacturers could justify. The researchers note that complementary technologies, including embedded III-V semiconductor lasers on silicon-on-insulator wafers and chip-based silicon nitride tunable delay lines, could be incorporated to realize a fully integrated, miniaturized OCT system in which even the swept light source lives on the chip. The system described in the paper already integrates a III-V semiconductor light source, silicon nitride interferometers, silicon-germanium photodetectors, and the ball-lens microprobe, drawing on the advantages of each material platform. A Chinese provisional patent application has been filed regarding the technology, signaling the team’s intent to move from demonstration toward deployment.

If that transition succeeds, the implications could reach well beyond metrology labs. Inline, non-destructive inspection of coatings, adhesives, semiconductor packages, medical device surfaces, and additive-manufactured parts could shift from sampling-based quality checks to continuous, three-dimensional verification. The same chip-scale engines could find uses in machine vision systems that need depth information at micrometer resolution, or in compact sensors embedded directly into production equipment. What the Shanghai Jiao Tong team has shown is that the optical complexity that once confined OCT to the optical table can be compressed onto a chip the size of a printed period, without sacrificing the sensitivity or accuracy that made the technique worth shrinking in the first place. The factory floor, long the harder market for optical coherence tomography, may finally be within reach.

Subject of Research: Chip-based photonic integrated circuit for swept-source optical coherence tomography in industrial inspection

Article Title: Compact photonic chip enables three-dimensional OCT inspection for industrial samples

Article References: Compact photonic chip enables three-dimensional OCT inspection for industrial samples. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: optical coherence tomography, photonic integrated circuits, silicon photonics, silicon nitride, industrial inspection, non-destructive testing, swept-source laser, ball-lens fiber probe, 3D imaging, metrology, germanium photodiodes, Shanghai Jiao Tong University

Retinal thickness is one of the most important structural biomarkers in ophthalmology. Swelling of the retina, known as edema, is a hallmark of diabetic macular edema, one of the leading causes of vision loss in working-age adults, and thickness changes also track the progression of glaucoma, age-related macular degeneration, and other conditions. Clinicians measure it using optical coherence tomography, or OCT, which bounces near-infrared light into the eye to build depth-resolved cross-sectional images of the retinal layers. From those scans, software computes a thickness map centered on the fovea, the pit of sharpest vision. The problem is that OCT machines are costly, require trained operators, and remain unavailable in many rural clinics, screening programs, and low-resource health systems. Color fundus photography, by contrast, is fast, cheap, and widely deployed — but it captures only a two-dimensional surface view, with no direct information about depth or thickness.

The research team, led by Maryam Yahyaie and Reza AghaeiZadeh Zoroofi of the University of Tehran’s School of Electrical and Computer Engineering, together with ophthalmologists Alireza Ramezani and Zhale Rajavi of Shahid Beheshti University of Medical Sciences, asked a deceptively simple question: if a machine learns the statistical relationship between the surface appearance of the retina and its underlying thickness, could it infer the missing dimension? Their answer is a generative artificial intelligence framework they call the Anatomically Conditional Latent Diffusion Model, or AC-LDM, trained on paired datasets in which each fundus photograph is matched to the thickness map derived from the same patient’s OCT scan.

The technical heart of the system is a diffusion model, the same family of generative architectures that has transformed image synthesis in recent years. Diffusion models learn to create data by reversing a gradual noising process: during training, the model observes images corrupted by increasing amounts of random noise and learns to denoise them step by step. At generation time, starting from pure noise, it iteratively refines a random field into a coherent image that matches the conditions it is given — in this case, the fundus photograph. Rather than operating on raw pixels, which is computationally expensive, the framework works in a compressed latent space. A variational autoencoder first learns to squeeze each retinal thickness map into a compact perceptual representation that preserves essential structure while discarding irrelevant detail, and the diffusion process then unfolds within that efficient latent space.

What distinguishes AC-LDM from a generic image-to-image translator is its anatomical guidance. The researchers observed that a standard cross-attention module — the mechanism that lets a diffusion model consult the fundus photograph while generating the thickness map — treats all spatial locations equally, which can blur or distort modality-specific retinal structure. Their solution is a spatially weighted cross-attention module, or SW-CA, which injects anatomically informed spatial weights into the attention computation. Regions of the fundus image that carry more reliable information about retinal topography are given greater influence during generation, while less informative areas are down-weighted. The design draws on attention-based conditioning strategies developed in modern vision architectures, but adapts them to the specific geometry and physiology of the retina, encouraging the model to respect the biological correspondence between what a fundus camera sees and what OCT measures.

Evaluating generative medical images is notoriously difficult, because a plausible-looking output is not necessarily an accurate one. The team therefore benchmarked AC-LDM against representative convolutional, generative adversarial, and transformer-based synthesis approaches using a battery of established metrics. Peak signal-to-noise ratio, or PSNR, quantifies pixel-level fidelity between the synthesized and true thickness maps; the structural similarity index, or SSIM, captures preservation of local structure such as the foveal depression and perifoveal rings. Two perceptual and distributional measures round out the picture: LPIPS, the learned perceptual image patch similarity, which correlates with human judgments of visual difference, and FID, the Fréchet Inception distance, which compares the statistical distribution of generated images against real ones.

The results were striking. AC-LDM achieved a PSNR of 30.88 decibels and an SSIM of 0.871, indicating both high pixel-level agreement and strong structural fidelity with the ground-truth thickness maps. On the perceptual side, it recorded an LPIPS of 0.105 and an FID of 29.5, both lower — and therefore better — than the baseline models it was compared against. Lower LPIPS means the synthesized maps look more like real ones to learned perceptual features, while the lower FID indicates that the overall population of generated maps is statistically closer to authentic retinal topography. Together, these numbers suggest that the anatomically guided diffusion approach does not merely produce attractive images but captures genuine structural information about the retina.

The implications extend well beyond a single laboratory benchmark. If thickness maps can be reliably synthesized from fundus photographs, screening programs for diabetic retinopathy — which already rely heavily on fundus photography in community and telemedicine settings — could gain a proxy measure of macular edema without deploying OCT hardware. The framework also opens a door for computational ophthalmology more broadly: large-scale fundus image archives, collected over decades and numbering in the millions, could potentially be retrofitted with synthesized thickness information, enabling retrospective studies of retinal structure at a scale OCT has never achieved. Cross-modal synthesis of this kind may likewise support data augmentation, privacy-preserving data sharing, and the training of downstream diagnostic models.

The authors are careful, however, to frame the work as a research advance rather than a clinical tool. Their own conclusion notes that clinical validation with thickness-specific error metrics — such as mean absolute error, root mean squared error, and agreement measures for central foveal thickness under the standard ETDRS grid — remains necessary before any translational use. Image-fidelity and perceptual scores, while encouraging, do not by themselves prove that synthesized thickness values are accurate enough to guide treatment decisions such as initiating anti-VEGF injections for macular edema. The study was approved by the Research Ethics Committee of Shahid Beheshti University of Medical Sciences, conducted under the Declaration of Helsinki, and used fully anonymized retrospective imaging data hashed to comply with HIPAA and GDPR requirements.

Still, the trajectory is clear and compelling. Diffusion models have already conquered natural image generation, and this study demonstrates how domain knowledge — in this case, the anatomy of the retina — can be woven into their attention mechanisms to solve a genuinely medical problem. The work, which received no specific funding and is published open access, points toward a future in which the humble fundus camera, already ubiquitous from Tehran to rural screening vans, becomes a window not just onto the surface of the retina but into its hidden third dimension. For the hundreds of millions of people at risk of retinal disease worldwide, many of whom live where OCT machines may never reach, that window could make the difference between blindness detected too late and sight preserved in time.

Subject of Research: Cross-modal synthesis of retinal thickness maps from color fundus photographs using an anatomically guided latent diffusion model

Article Title: Latent diffusion with anatomical guidance for retinal thickness map synthesis

Article References: Yahyaie, M., Zoroofi, R. A., Ramezani, A., & Rajavi, Z. (2026). Latent diffusion with anatomical guidance for retinal thickness map synthesis. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02792-4

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02792-4

Keywords: artificial intelligence, diffusion models, retinal thickness map, color fundus photography, optical coherence tomography, medical imaging, ophthalmology, deep learning, cross-modal synthesis, macular edema, diabetic retinopathy, computational ophthalmology

A new study published in the Journal of Neurology by Dejan Jakimovski of the University of Rochester and the University at Buffalo, together with an international team of collaborators, tackles that question head-on, and its answer is striking. The researchers found that prolonged P100 latency is directly and independently associated with hallmark MRI signatures of neurodegeneration in multiple sclerosis, including reduced whole brain volume, reduced gray matter volume, and a heavier burden of T2-FLAIR lesions. Critically, these associations persisted even after the team statistically accounted for retinal structural damage and for age, two factors that could otherwise explain the link. The work suggests that a humble electrophysiological measurement of visual processing speed may function as a window onto the diffuse neurodegenerative processes quietly reshaping the brain.

The central methodological challenge the investigators faced was one of disentanglement. Shifts in P100 latency could, in principle, arise from changes anywhere along the visual pathway. Upstream, at the level of the retina, loss of retinal ganglion cell axons, measurable as thinning of the peripapillary retinal nerve fiber layer, could plausibly slow the signal before it even reaches the optic nerve. Downstream, widespread damage within the brain, including demyelination of visual radiations and atrophy of cortical and subcortical structures, could equally delay the arrival of the evoked response. Without a way to separate these possibilities, the interpretation of a delayed P100 wave remained ambiguous. A delayed signal might simply be a retinal story told in milliseconds.

To resolve this ambiguity, the team enrolled 64 people with multiple sclerosis and subjected each participant to a trio of complementary assessments. Spectral domain optical coherence tomography, performed on Heidelberg Spectralis hardware, provided precise measurements of peripapillary retinal nerve fiber layer thickness, an established structural proxy for retinal axonal integrity. Standardized VEP testing yielded monocular P100 latency values for each eye, capturing the functional speed of signal conduction through the visual system. Finally, magnetic resonance imaging delivered quantitative measures of whole brain volume, gray matter volume, and T2-FLAIR lesion volume, the canonical MRI outcomes that neurologists use to gauge tissue destruction and disease progression.

The analytical centerpiece of the study was a series of mediation analyses, a statistical framework that allows researchers to ask whether the relationship between two variables persists after accounting for a third variable that might lie on the causal path between them. In this case, the investigators modeled whether the association between P100 latency and MRI outcomes was mediated by retinal nerve fiber layer thickness, while also adjusting for the well-known effect of age on brain volume. If retinal thinning were the true driver, then once it was held constant, the connection between VEP latency and brain atrophy should have evaporated. It did not.

The results were unambiguous. P100 latency was significantly associated with lower whole brain volume, with a p-value of 0.0037, and remarkably, 87.8 percent of that total effect proved to be direct and independent of both retinal nerve fiber layer measures and the significant age-related covariate effect on brain volume. In other words, nearly nine-tenths of the relationship between slowed visual conduction and brain shrinkage could not be explained by retinal damage or the passage of time. The pattern extended to the other MRI outcomes as well: greater P100 latency directly and independently predicted lower gray matter volume, with p equal to 0.015, and greater T2-FLAIR lesion volume, with p equal to 0.0036, after adjustment for the mediating effects of retinal thickness and age.

These findings carry weight because they reposition the VEP from a narrow diagnostic tool to something approaching a whole-nervous-system barometer. The P100 latency, long appreciated for its role in detecting subclinical optic nerve involvement and in supporting diagnosis under frameworks such as the McDonald criteria, now shows age- and retinal-independent associations with the very imaging outcomes that define neurodegeneration in multiple sclerosis. The authors suggest that, in addition to their established diagnostic utility, VEP measures may provide an objective neurophysiological metric for assessing neurodegeneration itself, a prospect with obvious appeal for clinical trials of remyelinating and neuroprotective agents, where sensitive, repeatable, and inexpensive outcome measures are perpetually in demand.

The study did not emerge from a vacuum. Prior work by overlapping teams had already hinted that VEP latency carries prognostic significance beyond the optic nerve. Earlier investigations linked prolonged latency to longitudinal worsening of fatigue in people with multiple sclerosis and to variance in cognitive performance, while other analyses associated VEP latency measures with brain tissue volume differences and with cortical thinning attributed to trans-synaptic degeneration following optic neuritis. Longitudinal cohort studies had likewise proposed VEP as a prognostic biomarker for neuroaxonal damage, and historical analyses had explored its potential as an outcome measure for remyelination trials, building on observations that latency can recover slowly over years as remyelination proceeds after optic neuritis. The new mediation analysis strengthens this literature by rigorously excluding retinal structural change as an alternative explanation for the latency-brain association.

The interplay between optical coherence tomography and VEP is itself a story of complementary technologies converging on the same biology. OCT offers a structural readout, quantifying the physical thinning of the nerve fiber layer as axons are lost, and has been incorporated into the 2024 McDonald diagnostic criteria for multiple sclerosis alongside VEP. VEP offers a functional readout, capturing conduction speed through myelinated pathways, which can be prolonged by demyelination even when axonal architecture remains partially intact. By combining both modalities in a single cohort and formally testing mediation, the researchers could demonstrate that the functional signal conveys information about global brain health that the structural measurement alone does not capture, a distinction with real consequences for how clinicians and trialists interpret these tests.

Certain caveats temper the enthusiasm. The study was a retrospective analysis of de-identified clinical data, approved under exemption criteria with a waiver of informed consent, and its sample of 64 participants, while adequate for the mediation analyses performed, is modest by the standards of biomarker validation. The datasets are not publicly available due to institutional restrictions, though de-identified data may be shared with qualified researchers upon reasonable request. The cross-sectional design, relating latency to MRI outcomes measured at the same time, cannot establish the temporal precedence needed to declare VEP latency a true predictive biomarker of future atrophy rather than a concurrent correlate. Longitudinal confirmation in larger, independent cohorts will be essential before P100 latency earns a place alongside MRI volumetry in monitoring algorithms. Nevertheless, the prospect is tantalizing: a noninvasive, relatively inexpensive electrophysiological test, already available in most neurology departments, may offer a real-time readout of neurodegenerative processes that MRI can only capture after tissue has been lost. For a disease in which the earliest and most consequential damage often unfolds invisibly, a millisecond-scale signal from the eye may prove to be one of the most informative messages the brain ever sends.

Subject of Research: The association between visual evoked potential latency delays and MRI-based neurodegeneration measures in multiple sclerosis.

Article Title: VEP latency delays predicts MRI neurodegenerative outcomes in multiple sclerosis

Article References: Jakimovski, D., Weller, J., Zivadinov, R., Weinstock-Guttman, B., Golan, D., Zarif, M., Costello, F., Sergott, R. C., Galetta, S. L., Kenney, R., Balcer, L. J., Van Hecke, W., Smeets, D., Dhakal, B., Morrow, S. A., Covey, T. J., & Gudesblatt, M. (2026). VEP latency delays predicts MRI neurodegenerative outcomes in multiple sclerosis. Journal of Neurology, 273(10), Article 621. https://doi.org/10.1007/s00415-026-14151-y

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14151-y

Keywords: multiple sclerosis, visual evoked potentials, P100 latency, optical coherence tomography, retinal nerve fiber layer, brain volume, gray matter atrophy, T2-FLAIR lesion volume, neurodegeneration, MRI biomarkers, optic nerve, neurophysiology