Osteoporosis is a silent disease by design. Bone mineral drains away for decades without pain or symptoms until a fragility fracture—often of the hip, spine, or wrist—suddenly changes a person’s life. The diagnosis matters enormously for prevention, because effective medications can slow or reverse bone loss and dramatically reduce fracture risk once the condition is identified. The problem is identification. The standard diagnostic tool, dual-energy X-ray absorptiometry, or DXA, is safe, inexpensive, and fast, but it is underused: many people who would qualify for testing never receive it, particularly older adults in primary care settings who fall through the cracks of screening guidelines.

The Shanghai team set out to answer a deceptively simple question: if an AI algorithm reads the lumbar spine on a routine non-contrast CT scan—something already acquired for countless other reasons—how closely does its verdict match what DXA would say? Their retrospective study enrolled 518 patients who had undergone both DXA and lumbar non-contrast CT, allowing direct comparison of the two modalities in the same individuals. Based on the DXA results, each patient’s bone quality was classified into one of three standard categories: normal, osteopenia (reduced bone mass not yet severe enough for an osteoporosis diagnosis), or osteoporosis.

The technology at the heart of the study is built on convolutional neural networks, the deep learning architecture that has transformed medical image analysis. The commercial AI software automatically segmented the vertebral bodies on the CT scans—drawing precise boundaries around each vertebra without human input—and then extracted volumetric bone mineral density, or vBMD, values from the thoracic vertebra T12 down through the lumbar vertebra L2. This is a key technical distinction from DXA. DXA measures area bone mineral density, a two-dimensional projection expressed in grams per square centimeter, whereas CT provides true three-dimensional volumetric density in milligrams per cubic centimeter, separating dense cortical shell from the trabecular interior where bone loss typically begins.

Because there is no single obvious way to summarize density across three vertebrae, the researchers devised four different classification strategies. Method 1 averaged the vBMD across all three vertebrae, T12 plus L1 plus L2. Method 2 averaged only T12 and L1. Method 3 averaged T12 and L2, and method 4 averaged L1 and L2. Each strategy produced its own bone-quality classification, allowing the team to determine which vertebral combination best captured the true skeletal status as defined by DXA. This kind of head-to-head comparison of measurement windows is practically important, because in real-world CT scans the field of view does not always include the same vertebrae from patient to patient.

To evaluate agreement, the researchers deployed a battery of statistical instruments. Intraclass correlation coefficients measured how consistently the AI-derived density values tracked with DXA across the cohort. Bland-Altman analysis examined whether the differences between the two methods were systematically biased or randomly scattered across the range of bone densities. Linear Cohen’s weighted kappa statistics quantified how often the categorical verdicts—normal, osteopenia, or osteoporosis—matched between modalities, with weighting that accounts for how far apart disagreements fall on the ordered scale. Multi-categorical logistic regression and receiver operating characteristic curves then estimated the diagnostic performance of each of the four AI-based classification methods, with a p-value below 0.05 serving as the threshold for statistical significance.

The results paint a picture of a technology that works well, though not perfectly. The four AI-based methods agreed remarkably with one another, posting an intraclass correlation coefficient of 0.909 with a 95 percent confidence interval of 0.893 to 0.923—meaning the choice of vertebral window matters far less than one might fear. Agreement between the AI methods and DXA was good but more modest: the ICCs ranged from 0.649 for method 2 (T12 plus L1) to 0.689 for method 1 (T12 plus L1 plus L2), with methods 3 and 4 falling in between at 0.666 and 0.680 respectively. These coefficients indicate substantial, clinically useful concordance, while also reflecting the inherent differences between two-dimensional areal and three-dimensional volumetric density measurements.

The most striking numbers came from the decision function that separates osteoporosis from the combined pool of normal and osteopenic patients. Here the AI achieved a precision of 0.834—meaning that when it flags a patient as osteoporotic, it is right about 83 percent of the time—and a recall of 0.735, meaning it catches roughly three-quarters of true osteoporosis cases. In screening terms, that combination is compelling. A missed case can still be caught at a later scan, while a false alarm simply triggers a confirmatory DXA, a cheap and harmless test. Among the four approaches, the three-vertebra average of T12, L1, and L2—the method that draws on the most bone data—showed the best diagnostic performance overall.

The implications reach well beyond a single hospital in Shanghai. Opportunistic screening of this kind requires no additional radiation, no additional scanner time, and no additional patient visit; the AI simply runs in the background on images that already exist. If validated across diverse populations and scanner types, such software could quietly build a bridge between the enormous volume of routine CT imaging and the stubbornly underdiagnosed population of people with brittle bones. Every abdominal scan performed for kidney stones or appendicitis could double as a skeletal health check, flagging patients who should be referred for confirmatory testing and treatment before their first fracture.

Cautions remain, as they must with any retrospective, single-center study. The cohort of 518 patients underwent both DXA and lumbar CT, a population that may not represent the general screening-age public, and the software was a specific commercial product whose performance could vary with different CT acquisition parameters or patient demographics. The study was also approved by the Medical Ethical Review Committee of Shanghai Changzheng Hospital with informed consent waived due to its retrospective design, and it was supported by the National Natural Science Foundation of China and other Chinese funding bodies, which had no role in the study design or analysis. Still, the core finding stands on its own: AI-driven vBMD extraction from routine non-contrast CT agrees well with DXA, and the simplest strategy—averaging density across T12 through L2—performs best. The era in which every CT scan quietly doubles as a bone-density test may be closer than anyone expected.

Subject of Research: AI-driven opportunistic osteoporosis screening using volumetric bone mineral density from routine non-contrast CT compared with DXA

Article Title: The diagnostic accuracy of AI-driven opportunistic osteoporosis screening based on routine non-contrast CT

Article References: Zhao, B., Sun, K., Shen, Q., Zhang, T., Xu, S., Qian, B., Ni, J., Duan, G., Wang, X., & Xiao, Y. (2026). The diagnostic accuracy of AI-driven opportunistic osteoporosis screening based on routine non-contrast CT. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02812-3

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02812-3

Keywords: osteoporosis, artificial intelligence, computed tomography, DXA, bone mineral density, opportunistic screening, radiology, machine learning, convolutional neural networks, volumetric BMD, osteopenia, medical imaging

The research team set out to answer a deceptively simple question: if Iran were to launch a fully organized, population-based breast cancer screening program tomorrow, what would it need to fix first? To do this rigorously, the investigators designed a three-stage convergent mixed-methods study. In the first stage, they conducted a systematic literature review, screening 953 records and retaining seven high-quality studies from which they extracted the building blocks of an assessment checklist. This checklist covered the domains that international guidelines consider essential for organized screening: target population definition, invitation and recall systems, diagnostic follow-up, workforce and equipment, information systems, financing, and performance monitoring.

Because a checklist built from a handful of studies could miss locally relevant criteria, the researchers then convened a two-round Delphi panel of six experts to refine and validate the instrument. The Delphi technique, a structured consensus method in which experts score and comment on items across successive rounds until agreement stabilizes, allowed the team to merge international evidence with the practical judgment of Iranian policy makers and clinicians. Only after the checklist had been finalized and validated did the team apply it, drawing on the data gathered in the literature review together with six semi-structured interviews with policy makers, oncologists, and frontline healthcare providers who run screening services on the ground.

The picture that emerged is one of a system with real assets but weak connective tissue. Iran does have a national guideline for breast cancer screening, and it has a broad network of health centers and screening facilities distributed across the country. Trained personnel exist, and public awareness campaigns about breast cancer have been running for years. Yet implementation of screening remains inconsistent from province to province, with significant disparities in access, resource allocation, and program reach. In practice, most screening is opportunistic: women are examined when they happen to visit a health facility for another reason, rather than being systematically identified, invited, and followed up by the health system.

This distinction between opportunistic and organized screening is more than bureaucratic hair-splitting. In organized programs, the health system maintains a registry of the eligible population, issues personal invitations at prescribed intervals, tracks every woman who fails to attend, and monitors outcomes such as cancer detection rates and interval cancers. Opportunistic screening, by contrast, systematically misses women who are less likely to seek care, and it produces uneven quality because there is no mechanism to verify that examinations and mammography readings meet standards. The Iranian study found that key performance indicators, including early detection rates and follow-up completion rates, are not systematically monitored anywhere in the program, which makes it nearly impossible to know whether screening is actually saving lives or merely generating activity.

Technical and administrative weaknesses compound the problem. The researchers documented manual data management across much of the screening pathway, meaning that records of examinations, referrals, and diagnostic outcomes are often kept on paper or in disconnected local systems rather than in an integrated electronic registry. Without such a registry, no invitation-based program can function, because the system cannot know who is due for screening, who has been screened, and who needs recall for suspicious findings. The study also identified insufficient dedicated funding for screening as a persistent constraint, with resources for promotion, equipment maintenance, and follow-up competing against other priorities in health budgets that are not ring-fenced for cancer control.

The interviews with frontline providers and specialists added texture to these structural findings. Providers described the practical consequences of limited proactive outreach: women in remote and underserved provinces may live far from mammography facilities, may not receive any reminder that screening is due, and may encounter the health system only when symptoms appear, often at later and harder-to-treat stages. Policy makers acknowledged that although Iran’s primary health care network, one of the most extensive in the region, reaches deep into rural areas, that network has not been fully harnessed for systematic cancer screening in the way it has for maternal and child health or for the management of other non-communicable diseases through initiatives such as IraPEN, the Iranian package of essential non-communicable disease services.

What makes the study notable for health policy observers is that it does not simply catalog failures. The authors emphasize that Iran’s established infrastructure, trained personnel, and ongoing awareness campaigns constitute a strong foundation on which an organized program could be built relatively quickly. The country’s primary health care system already assigns populations to specific health houses and health centers, which is precisely the catchment structure that invitation-based screening requires. The missing elements are the information systems to register and track eligible women, sustainable and dedicated financing, continued workforce development in mammography technique and reading, and a monitoring framework with defined indicators that are routinely reported and acted upon.

As a feasible pathway forward, the study proposes piloting an invitation-based screening model in selected provinces or defined catchment areas before attempting national scale-up. This incremental approach mirrors the trajectory followed by many countries that successfully transitioned from opportunistic to organized mammography screening: demonstrate the model in a manageable setting, measure coverage and quality indicators, fix the operational problems that only appear in practice, and then expand. The authors argue that such pilots would generate the local evidence on costs, acceptance, and logistics needed to convince decision makers to invest in the integrated electronic systems and monitoring frameworks that a national program demands.

The stakes are considerable. Because breast cancer is the top cause of cancer death among Iranian women, and because survival depends strongly on the stage at which the disease is detected, the difference between a functioning organized screening program and an opportunistic one is measured in lives. The study, conducted as part of the first author’s PhD thesis in health policy and approved by the ethics committee of Tehran University of Medical Sciences, received no specific external funding and involved no competing interests. Its message to Iran’s health authorities is direct: the foundations are in place, the checklist for success has now been written and validated, and the next step is a deliberate, monitored shift from waiting for women to arrive to actively inviting them in.

Subject of Research: Situation analysis of the breast cancer screening system in Iran using a mixed-methods design

Article Title: Situation analysis of breast cancer screening in Iran: a mixed-methods study

Article References: Dorri Kafrani, S., Akbari Sari, A., Yunesian, M., Moadabshoar, L., & Daroudi, R. (2026). Situation analysis of breast cancer screening in Iran: a mixed-methods study. International Journal for Equity in Health. https://doi.org/10.1186/s12939-026-02991-z

Image Credits: AI Generated

DOI: 10.1186/s12939-026-02991-z

Keywords: breast cancer screening, Iran, organized screening, opportunistic screening, health system assessment, Delphi method, mammography, health equity, non-communicable diseases, screening policy, primary health care, mixed methods

The study, led by Zhenzhen Guan, Yijin Wang, and colleagues, tackles a fundamental blind spot in musculoskeletal medicine: the intimate, two-way relationship between bone and muscle. These tissues are anatomical next-door neighbors, packed tightly along the spine, and they constantly exchange chemical and metabolic signals. When osteoporosis and sarcopenia, the age-related loss of muscle, strike together, patients face sharply higher rates of falls, fractures, and death. Yet most clinical assessments examine these tissues separately, and often only at a single slice of a single vertebra, a sampling strategy that misses the spatial patchiness of bone loss and the segment-by-segment variation in muscle health.

At the heart of the new work is a deep learning architecture the team calls UNet3D, a modified descendant of the celebrated U-Net convolutional network that has become a workhorse of medical image segmentation. The researchers confronted a classic engineering dilemma: true 3D processing of medical volumes is computationally punishing, while 2D slice-by-slice analysis loses the spatial context that matters for anatomy. Their solution was a dimensional transformation module that ingests sixteen consecutive 512-by-512 grayscale images and, through two dual-convolution operations applied before any downsampling, converts them into enriched 3D feature maps. This early-stage dimensionality reduction slashes computational cost while preserving the fine anatomical margins that a segmentation network must capture. The output is structured as a 2D categorical map with softmax activation, deliberately aligned with the DICOM format that hospitals already use, so the tool can slot into existing clinical workflows without friction.

The training data came from two independent centers and were filtered with unusual rigor. Of 2,700 osteoporosis patients screened at the first center, 2,320 were excluded for reasons ranging from prior anti-osteoporosis treatment to vertebral fractures, severe scoliosis, or inadequate image quality; a similar winnowing at the second center left just 30 external test patients. The final cohort comprised 410 patients with confirmed osteoporosis, defined by a DXA T-score of -2.5 or lower, aged 55 to 75, contributing 2,460 vertebral bodies across more than 76,000 CT slices and 2,050 intervertebral disc sections on MRI. Every image was annotated slice by slice by a trained annotator, then reviewed by a spine surgeon with 20 years of experience and finally by a chief physician with 30, a layered quality-control pipeline designed to give the neural network trustworthy ground truth.

The performance numbers are striking. On lumbar spine CT, UNet3D achieved a pixel accuracy of 0.984 and a mean intersection over union, the standard strictness metric for segmentation, of 0.760 in the internal test set, beating U-Net, U2-Net, and Mask R-CNN. On MRI, it reached a pixel accuracy of 0.973 and a mean intersection over union of 0.724 for the far harder task of delineating ten bilateral paraspinal muscles and intermuscular spaces at five disc levels, one hundred distinct labels in total. Critically, the model held up on the completely independent external test set from the second center, with mean intersection over union values of 0.741 on CT and 0.713 on MRI, evidence that the system generalizes beyond the data it was trained on rather than merely memorizing one hospital’s scanner quirks.

Once the vertebrae were segmented, the pipeline applied a gradient-adaptive algorithm to perform what the authors describe as oscillatory separation of cortical and cancellous bone. This matters because osteoporotic changes are concentrated in trabecular bone, the spongy interior lattice whose metabolic activity runs roughly eight times faster than that of the dense cortical shell. Traditional threshold-based methods struggle to find that boundary in osteoporotic patients, where differential bone loss blurs the distinction and partial volume effects corrupt the measurement. Instead of a small manual region of interest in one vertebra, the system computes Hounsfield unit values across the entire vertebral volume of interest for every level from L1 to S1, capturing the spatial heterogeneity of bone mineral density. The resulting values correlated strongly with DXA T-scores at L1 through L4, with correlation coefficients ranging from 0.85 to 0.93, validating the automated measurement against the clinical reference standard.

On the MRI side, the team faced a data limitation: their annotations covered only the intervertebral disc levels, not the full muscle continuum. Their workaround was elegant, parameterizing the 2D cross-sectional anatomy at each disc level and then spatially integrating those sections using a CT-MRI co-registration framework to reconstruct the complete L1-to-S1 continuum of muscles and the fat-filled gaps between them. The volumetric census that emerged was richly detailed. The psoas major claimed the largest total bilateral volume at roughly 173,000 cubic millimeters, while the quadratus lumborum was the smallest. The muscles and spaces showed distinct craniocaudal patterns: the longissimus peaked at L1-L2 and shrank toward the sacrum, the multifidus followed a biphasic rise-and-fall, and the psoas dominated the upper lumbar segments. Marked left-right asymmetries appeared throughout, with the most dramatic imbalances concentrated at the L4-L5 level, where some left-to-right volume ratios exceeded 25-fold.

The statistical payoff came when the team linked the two modalities. Using a multivariate linear regression model with Z-score standardization, they found that paraspinal muscle volumes were positively correlated with vertebral Hounsfield unit values, while intermuscular space volumes were inversely correlated, a pattern that held more strongly on the left side and in the upper lumbar segments. Among individual muscles, the multifidus, longissimus, and iliocostalis showed the strongest associations with bone density, and among the spaces, the gaps between the longissimus and multifidus, the iliocostalis and longissimus, and the multifidus and vertebral lamina carried the strongest signals. Enlarged intermuscular spaces, the authors note, likely reflect atrophy of the adjacent muscles, so the shrinking muscle and widening gap tell the same story from two directions. The team also quantified fatty infiltration using automated optical inspection and found a significant inverse relationship between fat in the paraspinal muscles and vertebral bone density, with the multifidus and psoas showing the strongest negative correlations.

That fat finding carries real biological weight. When adipose tissue infiltrates muscle, it directly replaces contractile fibers, degrading muscle quality and strength, which in turn weakens the mechanical stimulation that bones need to maintain their mass. Fat also secretes inflammatory cytokines that stoke chronic inflammation in the bone microenvironment, activating osteoclasts, the cells that resorb bone, and accelerating bone loss. The synergy between failing muscle and brittle bone elevates the risk and severity of vertebral fractures, and patients with the combined condition of osteosarcopenia face higher rates of frailty, disability, and falls. The segment-specific correlations observed here hint that different spinal levels experience unique mechanical loading from their adjacent muscles, suggesting that regional variation, not a single averaged measurement, should guide how clinicians interpret bone-muscle relationships in the lumbosacral spine.

The researchers are candid about the limitations. The study is retrospective, which invites selection bias; DXA, as a 2D projection technique, has inherent constraints as a reference standard even though quantitative CT remains too costly, radiologically heavy, and scarce to serve that role here; and the cohort consisted exclusively of osteoporosis patients in one spinal region, without accounting for confounders like nutrition or physical activity. Future work will incorporate quantitative CT calibration, Dixon water-fat sequences, more diverse populations, and extension to the thoracic and cervical spine through transfer learning. Still, the vision is compelling: an automated tool that transforms scans already sitting in hospital archives into objective, three-dimensional musculoskeletal metrics, enabling opportunistic screening of the many patients whose osteoporosis would otherwise go undetected until a bone breaks. In a rapidly aging world, that kind of quiet, early warning system could prove to be exactly what the silent disease has been waiting for.

Subject of Research: A deep learning model for automated 3D quantification of vertebral bone density and paraspinal muscle metrics from routine CT and MRI in osteoporosis patients

Article Title: Construction of musculoskeletal quantitative model based on deep learning and study of musculoskeletal relationship in patients with osteoporosis

Article References: Guan, Z., Wang, Y., Zhang, S., Zhang, Y., Wang, L., Chen, Y., & Lu, X. (2026). Construction of musculoskeletal quantitative model based on deep learning and study of musculoskeletal relationship in patients with osteoporosis. Journal of Advanced Research, 88, 899-909. https://doi.org/10.1016/j.jare.2026.01.037

Image Credits: AI Generated

DOI: 10.1016/j.jare.2026.01.037

Keywords: osteoporosis, deep learning, UNet3D, medical image segmentation, paraspinal muscles, bone mineral density, DXA, Hounsfield units, sarcopenia, fatty infiltration, spine imaging, opportunistic screening

The study, registered retrospectively on ClinicalTrials.gov as NCT07162168, enrolled 4,152 participants who had undergone noncontrast routine CT scans. The cohort included 1,947 men, with a median age of 59 years spanning a remarkable range from 18 to 99. Rather than relying on slow, labor-intensive manual tracing of the femur, the team trained a deep learning segmentation model based on the nnU-Net framework, a self-configuring architecture widely regarded as one of the most robust tools for biomedical image segmentation. A subset of proximal femurs was first annotated by hand to teach the model exactly where the bone begins and ends in each volumetric scan. Once trained, the automated system could carve out the femoral volume of interest on its own, achieving a Dice similarity coefficient of 0.963 on the test set, a score indicating near-perfect overlap with expert annotations.

With the femur reliably isolated in thousands of scans, the researchers turned to radiomics, the practice of converting medical images into hundreds of quantitative features that describe texture, shape, intensity and spatial patterns within a defined region. Using the open-source PyRadiomics package, they extracted features from each segmented femur, following standardized conventions designed to make radiomic measurements reproducible across scanners and institutions. The idea behind radiomics is deceptively simple: the microscopic architecture of bone, including trabecular thinning, cortical porosity and the redistribution of mineral, leaves fingerprints in the statistical texture of a CT image. Those fingerprints, aggregated across thousands of features, may encode a person’s skeletal age far more richly than a single density measurement.

To test whether these features actually carry information about aging and sex, the team built machine learning models in a development cohort of 2,969 participants and then evaluated them in a completely separate external validation cohort of 1,183 participants, a design that guards against the optimistic results that plague many single-center AI studies. For age, the best performer was a support vector regression model, which estimated chronological age from femoral radiomics with a coefficient of determination of 0.50 in external validation, a mean absolute error of 8.84 years and a root mean squared error of 11.32 years. In plain terms, the model could roughly halve the uncertainty of a blind guess about a person’s age using nothing but the texture of their hip bone.

Sex classification proved even more striking. Two algorithms, support vector classification and the gradient-boosting method XGBoost, both achieved an area under the receiver operating characteristic curve of 0.94 in external validation, with support vector classification showing slightly higher accuracy and F1-score. An AUC of 0.94 means the model distinguished male from female skeletons with high reliability across the entire cohort, capturing structural sexual dimorphism in the femur that goes well beyond simple size differences. Because the features were extracted automatically from routine scans, the result hints that sex-related skeletal architecture is legible in ordinary clinical imaging without any dedicated bone protocol.

Perhaps the most consequential finding concerns the relationship between the radiomics-derived age index and actual bone mineral biology. In the external validation cohort, the researchers compared their radiomic age index against volumetric bone mineral density of the femoral neck measured by quantitative computed tomography, the gold-standard three-dimensional density technique. The radiomics index tracked vBMD more strongly than chronological age did, with a Spearman correlation of minus 0.546 versus minus 0.443 for age alone, a difference the authors confirmed statistically using Steiger’s test (Z = 4.97, P < 0.001). This comparison matters because it suggests the radiomic signature is not merely a proxy for how many birthdays a person has had, but a reflection of genuine structural deterioration in the bone, the kind of change that underlies osteoporosis and hip fracture risk.

The clinical logic here is what researchers call opportunistic imaging. Dual-energy X-ray absorptiometry, the standard test for osteoporosis, is underused: many people at risk never get scanned, and fractures often arrive as the first sign of disease. But a large fraction of older adults undergo abdominal, chest or pelvic CT for unrelated reasons, and those scans routinely capture the proximal femur. If an automated pipeline can convert that incidental imaging into a skeletal age index or a bone quality flag, clinicians could identify patients with accelerated bone aging who would benefit from confirmatory density testing or preventive treatment, all without a single additional scan, extra radiation dose or added visit. The authors caution that their radiomic patterns still need validation against hard clinical outcomes such as incident fractures before such screening becomes practice, but the biological anchoring to vBMD is an encouraging first step.

The technical achievement should not be undersold either. Automated segmentation at a Dice coefficient above 0.96 removes the bottleneck that has historically limited radiomics to small, single-center studies requiring hours of expert contouring. By combining nnU-Net segmentation, standardized PyRadiomics extraction and rigorous external validation across nationwide cohorts, the study offers a template for how imaging biomarkers can be scaled to population level. The work was supported by the Beijing Natural Science Foundation-Fengtai Joint Innovation Project, the National Natural Science Foundation of China Science Center Program and the Shenzhen Medical Research Fund, and it was approved by the ethics committee of Peking University People’s Hospital with informed consent waived for the retrospective use of de-identified images.

There are, of course, caveats to keep in mind. The cohort was drawn from Chinese hospitals, and radiomic features can be sensitive to scanner vendor, reconstruction settings and dose, so multicenter generalization beyond this dataset remains to be demonstrated. The age model’s mean absolute error of nearly nine years is impressive for image-only prediction but far too coarse to serve as a precise biological clock for any individual. And the study measured associations with density, not with fractures, falls or mortality, so the ultimate clinical value hinges on longitudinal follow-up. Still, the direction of travel is clear: the femur that appears incidentally on a routine scan carries a quantifiable structural biography, written in texture patterns that machines can now read at scale.

If subsequent studies confirm that radiomics-derived bone age predicts fractures and responds to osteoporosis treatment, the implications could extend well beyond orthopedics. Bone is a living archive of hormonal change, nutrition, inflammation and mechanical loading, and a cheap, automated readout of its microarchitecture could inform research on aging itself, from sarcopenia to metabolic bone disease. For now, the study stands as a proof of concept that the skeleton’s quiet remodeling, normally invisible in the reading room, can be decoded from data that hospitals already possess, waiting in the archive.

Subject of Research: Automated CT radiomics of the proximal femur for quantifying age- and sex-related bone structural changes

Article Title: Automated proximal femur CT radiomics identifies age- and sex-related radiomic patterns of bone structure: a nationwide multicohort validation study

Article References: Cheng, H., Zhuang, L., Zhang, Y., Ran, Y., Jiang, B., & Kou, Y. (2026). Automated proximal femur CT radiomics identifies age- and sex-related radiomic patterns of bone structure: a nationwide multicohort validation study. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02836-9

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02836-9

Keywords: radiomics, proximal femur, CT imaging, deep learning, nnU-Net, bone aging, bone mineral density, osteoporosis, machine learning, opportunistic screening, quantitative CT, hip fracture risk

The research, led by Sanaa Amina Gourine of the Nara Institute of Science and Technology together with colleagues from Miyazaki University, Shiga University of Medical Science, Osaka University, and Ehime University, describes a pipeline for volumetric segmentation and quantitative analysis of 21 torso structures in CT images. Segmentation, in this context, means teaching an algorithm to trace the exact boundaries of each muscle and fat compartment voxel by voxel across a three-dimensional scan. Once those boundaries are known, software can compute each structure’s volume in cubic centimeters and its mean density in Hounsfield units, the standard radiological measure of tissue composition. Lower muscle density on CT reflects fatty infiltration of the muscle, a hallmark of aging and disuse that correlates with reduced strength and physical performance. Extracting these numbers automatically from thousands of existing scans would turn routine clinical imaging into an opportunistic screening resource for conditions that currently go undiagnosed.

The technical heart of the study is a comparison between two deep learning architectures, both augmented with a mechanism for estimating predictive uncertainty. The first is a two-dimensional Bayesian UNet, a convolutional neural network that processes the scan slice by slice and uses a technique called Monte Carlo dropout sampling. Instead of producing a single deterministic answer, the network runs its prediction multiple times with randomly deactivated internal units, and the variability across those runs serves as a proxy for how confident the model is at each location. The second, and ultimately the winning, approach is a fully three-dimensional, patch-based model built on the well-known nnU-Net framework, which the authors call the 3D MSKSegmenter. Rather than relying on repeated stochastic sampling, this model derives uncertainty directly from the entropy of its prediction probabilities: when the network’s output probabilities are spread evenly across competing classes, entropy is high and the prediction is flagged as unreliable.

That uncertainty estimate is not a mathematical flourish; it is the framework’s quality control system. In the validation experiments, the researchers found strong negative correlations between predictive uncertainty and segmentation accuracy, measured by the Dice coefficient, the standard overlap metric comparing automated contours against manual ground truth. The correlation reached minus 0.870 for the 3D MSKSegmenter and an even steeper minus 0.955 for the two-dimensional Bayesian UNet. In practical terms, when the model reports low uncertainty, its segmentation is almost always accurate, and when uncertainty climbs, accuracy tends to collapse. This relationship allows the framework to act as its own inspector: for large-scale database studies where no expert can review every output, the system can automatically filter out cases whose uncertainty exceeds a threshold, ensuring that only reliable measurements flow into downstream statistical analyses. The authors validated this quality control mechanism across a fivefold cross-validation experiment on 20 manually labeled CT images.

The 3D MSKSegmenter outperformed both its two-dimensional counterpart and a formidable external benchmark. The team compared their results against TotalSegmentator, a widely used, state-of-the-art open-source model capable of segmenting 104 anatomical structures across the whole body. Despite that model’s breadth, the torso-specialized 3D MSKSegmenter achieved a superior average Dice coefficient of 0.89 plus or minus 0.03 across the 21 target structures, a statistically significant advantage over both competitors at p less than 0.001. The result underscores a recurring lesson in medical image analysis: models trained specifically on the structures and imaging characteristics of a target anatomy, with careful handling of the enormous variability in body habitus, scan protocol, and contrast phase found in real hospital databases, can beat generalist systems on their home turf.

Robustness was then tested at scale. The researchers applied the framework to a single-institution database of approximately 1,200 patients, an external test set far larger than the labeled training and validation data. The framework proved stable across this heterogeneous clinical population, and the uncertainty-based filtering continued to identify the cases most likely to contain segmentation errors. This matters because the ultimate ambition of the project is not to segment 1,200 scans but to enable population-scale studies in which tens of thousands of torso CTs, already acquired for other reasons, are mined for musculoskeletal biomarkers without any additional imaging cost or radiation exposure. Opportunistic screening of this kind has been proposed for cardiometabolic risk and sarcopenia, but it has been limited by the absence of segmentation tools that are simultaneously accurate across many muscle groups and honest about their own failure modes.

The downstream analysis delivered the kind of biological signal that justifies the engineering effort. Measuring muscle volume and mean Hounsfield units across the large database, the researchers detected significant age- and sex-related trends in muscle density and volume, consistent with the known epidemiology of sarcopenia and age-related fatty muscle degeneration. Because the framework quantifies 19 individual muscles rather than a single aggregate muscle area at one reference slice, it opens the door to site-specific analyses: whether the gluteal muscles, the paraspinal stabilizers, or the abdominal wall degrade at different rates, and how those patterns relate to conditions such as hip osteoarthritis, knee osteoarthritis, or postoperative physical function. Prior work by overlapping research groups has linked gluteal muscle volume and quality to early recovery after total hip arthroplasty and lower limb muscle degeneration to knee osteoarthritis severity, and a scalable torso-wide tool provides the missing population-level instrument for such investigations.

The study also situates itself within a rapidly maturing field. Earlier approaches to automated muscle segmentation relied on random forest classifiers, atlas-based registration, or fuzzy c-means clustering, typically targeting one or a few muscle groups. The rise of deep learning brought UNet-style architectures and, more recently, the self-configuring nnU-Net, which has become a de facto standard in biomedical segmentation. Bayesian extensions of the UNet introduced calibrated confidence estimation to medical imaging, and previous validations demonstrated that uncertainty estimation works for bone and muscle assessment in hip-to-knee CT. The present work extends that lineage in two directions: it covers an unusually comprehensive set of torso musculature and fat compartments, and it explicitly integrates uncertainty as a gatekeeper for biomarker extraction in very large databases, rather than treating it as a laboratory curiosity.

Transparency is part of the package. The code for the 3D MSKSegmenter, built on the open nnU-Net repository, has been released on GitHub by the NAIST Bio-Image Intelligence group, and the trained models are available from the corresponding authors on reasonable request. The retrospective study was approved by local ethics committees and conducted in accordance with the Helsinki Declaration, with informed consent obtained through opt-out mechanisms. Funding came from the Japan Society for the Promotion of Science, the Japan Agency for Medical Research and Development, and the Cross-ministerial Strategic Innovation Promotion Program on integrated health care systems.

The broader implication is that the raw material for a new generation of musculoskeletal epidemiology is already sitting on hospital servers. As populations age, sarcopenia and related muscle disorders are projected to become a major driver of disability and health care cost, yet muscle status is rarely measured outside specialized research settings. A framework that can sweep through existing CT archives, quantify 21 body composition structures per patient, and flag its own uncertain outputs for review turns routine diagnostic imaging into a longitudinal window on the aging musculoskeletal system. If external validations at additional institutions replicate the accuracy and reliability reported here, the humble torso CT, acquired for everything from lung cancer screening to abdominal pain, could quietly become one of the most valuable population health instruments in modern medicine.

Subject of Research: Automated deep learning segmentation and uncertainty-based quality control for volumetric musculoskeletal analysis of torso CT images

Article Title: Automated framework for volumetric musculoskeletal segmentation and assessment of torso CT images in large-scale databases

Article References: Gourine, S. A., Soufi, M., Otake, Y., Masaki, Y., Murakami, Y., Nagatani, Y., Watanabe, Y., Uemura, K., Kono, K., Takao, M., Fujimori, T., Sugano, N., & Sato, Y. (2026). Automated framework for volumetric musculoskeletal segmentation and assessment of torso CT images in large-scale databases. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-026-03785-y

Image Credits: AI Generated

DOI: 10.1007/s11548-026-03785-y

Keywords: musculoskeletal segmentation, torso CT, deep learning, nnU-Net, predictive uncertainty, Bayesian UNet, sarcopenia, body composition, TotalSegmentator, Hounsfield units, large-scale databases, opportunistic screening