The brevity of Narasimhan’s life, cut short at forty-six, invites reflection on how much of the early history of Indian neurology remains scattered across obituaries, institutional records, and the memories of successive generations of trainees. Historical scholarship in medicine depends on this kind of archival reconstruction, and the authors of the new profile situate their subject within the broader narrative of how neurological practice developed in mid-twentieth-century India, a period when the country had very few dedicated neurologists, no large neurological research infrastructure, and limited access to the diagnostic technologies that were transforming the field elsewhere.

Understanding why profiles like this matter requires some appreciation of what neurology looked like in India during Narasimhan’s working years. Electroencephalography, which records the brain’s electrical activity through electrodes placed on the scalp, was still a young technology globally, having been applied to human subjects in the late 1920s and 1930s. Its spread to Indian institutions came slowly, and for most of the period in question, diagnosis rested on meticulous clinical examination: the detailed mapping of weakness, reflexes, sensation, and cranial nerve function that remains the foundation of neurological training today. In that environment, the authority of a neurologist rested on clinical acumen and teaching rather than on laboratory resources, and the reputations of the pioneers were built case by case, ward round by ward round.

The institutional history of Indian neurology provides the backdrop against which the new profile should be read. The Neurological Society of India, the professional body that united the country’s neurologists, neurosurgeons, and allied specialists, was founded in 1951, a milestone that marked the transition from isolated individual practice to an organized specialty with meetings, journals, and shared standards. Madras, now Chennai, emerged as one of the crucial centers of this development. The Madras Institute of Neurology at Madras Medical College, whose history has been documented in earlier reviews, became a training hub that produced generations of neurologists who then spread across the country, establishing departments in medical colleges and teaching hospitals from the south of India to the north.

Chennai’s role in this story is one reason the new profile is notable: one of its co-authors is based at Kilpauk Medical College in that city, and the lineage connecting mid-century Madras neurology to the present-day departments of Tamil Nadu is a living professional inheritance. The historians of Indian neurology who have written on this period, including the authors of a widely cited 2013 review of the field’s historical development, emphasize that the specialty’s growth was driven by a handful of interconnected figures, many of whom trained together, corresponded across institutions, and collectively argued for the resources needed to build dedicated neurological units. Biographical profiles are the raw material of that argument, because they demonstrate concretely what individual physicians contributed in an era of scarcity.

Epilepsy offers a particularly clear example of how the foundations laid by early Indian neurologists grew into a modern subspecialty. Reviews of epilepsy surgery in India describe a trajectory that began with a small number of pioneering centers performing surgical procedures for drug-resistant epilepsy and expanded, over decades, into a network of comprehensive epilepsy programs using video-electroencephalographic monitoring, magnetic resonance imaging, and invasive electrode recordings to localize seizure origins in the brain. Epidemiological analyses of epilepsy surgery in India published in the last decade show both the growth of surgical capacity and the persistent treatment gap, with millions of people with epilepsy in the country lacking access to specialist care. The clinicians of Narasimhan’s generation, working without any of these technologies, established the clinical frameworks for seizure classification and medical management on which that later surgical enterprise depended.

The Pioneers in Neurology series, in which the new profile appears, serves a function that goes beyond commemoration. By publishing short biographical accounts in a major international neurology journal, the series makes the history of the discipline visible to a global readership that may know little about how neurology developed in different parts of the world. This matters for scientific reasons as well as cultural ones. The epidemiology of neurological disease in India differs in important ways from that of high-income countries, including a higher burden of infections affecting the nervous system, distinct patterns of stroke etiology, and genetic variations that influence conditions such as muscular dystrophy and movement disorders. Knowing who built the clinical and research base in India helps explain why certain research questions were prioritized there, and why contemporary Indian neuroscience looks the way it does.

There is also a methodological point worth making about historical profiles of this kind. Because the authors state that no datasets were generated or analyzed in preparing the manuscript, the profile is a work of synthesis, drawing on previously published literature and historical records rather than new data collection. The reference list of the article points to the key secondary sources on Indian neurology’s history, including the 2013 historical perspective in the Annals of the Indian Academy of Neurology, the account of the Madras Institute of Neurology published in Neurology India, the memoir of the celebrated neurosurgeon B. Ramamurthi, and reviews of epilepsy surgery and its epidemiology in India. This citation network shows how the history of a medical specialty is assembled: from autobiographies, institutional histories, society records, and earlier biographical sketches, each source preserving fragments that later historians piece together.

For readers encountering Narasimhan’s name for the first time, the profile is an invitation to think about the contingency of medical history. A physician who dies at forty-six, in 1959, just as organized neurology was taking institutional shape in India, leaves behind an incomplete record, and the survival of his reputation depends on whether later colleagues judge the record worth preserving. The decision by neurologists at NIMHANS, today one of India’s premier neuroscience institutions, to document his career signals a deliberate effort to keep the founding generation in view. NIMHANS itself grew out of the mid-century expansion of mental health and neurosciences in the country, and its faculty writing about a predecessor from the 1940s and 1950s illustrates the continuity between the pioneers and the modern specialty.

The broader lesson of the profile is that the history of neuroscience is global, uneven, and still being written. The technologies that dominate headlines today, from transcranial magnetic stimulation to brain-machine interfaces, rest on generations of clinical work that was often unglamorous, poorly funded, and carried out in conditions very different from those of well-equipped research hospitals. In India, that work was done by a small cohort of physicians who built departments, founded a society, trained students, and treated enormous numbers of patients with limited means. Recovering the individual stories within that cohort, as the Journal of Neurology profile of S.T. Narasimhan does, is how the specialty understands its own origins, and how contemporary researchers and clinicians measure how far the field has traveled in the seven decades since his death.

Subject of Research: Historical profile of early Indian neurologist S.T. Narasimhan (1913–1959) and the development of neurology in India

Article Title: S.T. Narasimhan (1913–1959)

Article References: Harish, S., Gowda, N., & Balachandar, N. (2026). S.T. Narasimhan (1913–1959). Journal of Neurology, 273(11), Article 655. https://doi.org/10.1007/s00415-026-14197-y

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14197-y

Keywords: S.T. Narasimhan, history of neurology, Indian neurology, Pioneers in Neurology, Journal of Neurology, NIMHANS, Madras Institute of Neurology, Neurological Society of India, epilepsy surgery, electroencephalography, medical history, Chennai

The research, led by Guixian Tang and Chunyuan Zeng of the First Affiliated Hospital of Jinan University together with colleagues at Guangdong 999 Brain Hospital, examined 123 patients with drug-resistant temporal lobe epilepsy and 48 healthy controls. Every participant underwent preoperative magnetic resonance imaging with diffusion tensor imaging, as well as fluorodeoxyglucose positron emission tomography, which maps how avidly different brain regions consume glucose. The team then tracked each patient’s postoperative seizure outcome using the standard Engel classification, dividing the cohort into 83 patients who became completely seizure-free and 40 who continued to experience seizures after surgery.

The glymphatic system, first described in the early 2010s, is a network of channels that run alongside blood vessels, the perivascular spaces, through which cerebrospinal fluid flushes metabolic waste out of brain tissue, most vigorously during sleep. Because these channels are far too small to image directly in living humans, researchers rely on indirect proxies. The most popular is the diffusion tensor imaging analysis along the perivascular space, or DTI-ALPS index. It exploits the fact that water molecules diffuse more freely along the perivascular channels than across them; when those channels become clogged or degraded, the directional asymmetry of water diffusion diminishes, and the ALPS index falls.

When the team compared ALPS values across the three groups, a clear gradient emerged. Both patient groups showed lower ALPS indexes on the side of the brain harboring the epileptic focus, and in overall mean values, than the healthy controls. More strikingly, the seizure-free patients sat in between: healthy controls scored highest, seizure-free patients intermediate, and non-seizure-free patients lowest. In other words, the worse the glymphatic clearance capacity before surgery, the bleaker the prognosis after it. The ipsilateral and mean ALPS indexes declined progressively across the groups in a statistically significant manner, hinting that glymphatic health is not merely a byproduct of having epilepsy but a marker of how severely the disease has disrupted the brain’s housekeeping machinery.

What makes the study genuinely novel, however, is not the ALPS comparison itself but the way the researchers coupled glymphatic measurements to brain metabolism. Glucose is the brain’s principal fuel, and FDG-PET reveals where that fuel is being burned, region by region, expressed as a standardized uptake value ratio. In patients destined for seizure freedom, the ipsilateral ALPS index correlated with the metabolic activity of multiple discrete brain regions, with the associated hypometabolic areas localized to the frontal lobe and cingulate cortex on the same side as the epileptic focus. These correlations survived stringent statistical correction for multiple comparisons, with voxel-level thresholds set at a corrected significance level below 0.005 and cluster-level significance below 0.05.

In the patients who failed to achieve seizure freedom, the picture changed dramatically. There, the ipsilateral ALPS index no longer tracked regional glucose uptake in any meaningful way. Instead, it correlated with parameters describing the architecture of the brain’s entire metabolic network: the characteristic path length, a measure of how many steps information or influence must traverse across the network; the global efficiency, which quantifies how well the network as a whole integrates information; and a related network measure denoted lambda. The correlations were modest in magnitude but statistically significant, with correlation coefficients ranging from about negative 0.37 to positive 0.38 and p-values between 0.016 and 0.023.

This divergence is the conceptual heart of the paper. The seizure-free patients displayed what the authors call preserved glymphatic-metabolic coupling: the waste-clearance system and the brain’s energy consumption remained locally synchronized, with cleaner-clearing brains showing coherent, regionally specific metabolic patterns. The non-seizure-free patients, by contrast, showed a breakdown of that local coupling and a shift toward network-level covariance, in which glymphatic function was entangled with the global organization of metabolism rather than with any particular region. The authors interpret this shift as a sign of deeper, system-wide disruption, one that a focal resection of the epileptic temporal lobe cannot fully repair.

Curiously, the raw metabolic measurements themselves told a subtler story. When the researchers compared standardized uptake value ratios and metabolic network parameters directly between the seizure-free and non-seizure-free groups, they found no significant differences. The prognostic information was not in any single number but in the relationship between two numbers, in how the glymphatic index and the metabolic data moved together or failed to. This is a recurring lesson in modern neuroimaging: the most informative features are often not the measurements themselves but their correlations, covariances, and network signatures, which capture how brain systems interact rather than how they behave in isolation.

The clinical implications are potentially significant. Epilepsy surgery is irreversible, removing or disconnecting brain tissue in the hope of ending seizures, and patients and families weigh the decision carefully. A preoperative imaging marker that stratifies surgical prognosis could refine patient counseling, sharpen the selection of surgical candidates, and perhaps motivate more extensive or alternative interventions in patients whose glymphatic-metabolic profiles look unfavorable. Because the DTI-ALPS index requires only a standard diffusion MRI sequence already acquired in most presurgical evaluations, adding it to the clinical workup would demand no new hardware, no radioactive tracers beyond those already used, and no additional scanning sessions.

As with any retrospective single-center study, caveats apply. The findings establish correlation, not causation; a low ALPS index does not itself cause poor surgical outcomes, and the mechanisms linking waste clearance to seizure recurrence remain speculative. The ALPS index is also sensitive to imaging parameters, head motion, and anatomical variability, and the field is still standardizing how it should be measured and interpreted. The authors themselves frame the result as evidence that glymphatic dysfunction carries preoperative prognostic value, a hypothesis that will need prospective validation in independent cohorts before it changes clinical practice. Still, the study opens an intriguing window: the brain’s nocturnal cleaning crew, long dismissed as plumbing trivia, may hold one of the keys to predicting which patients with drug-resistant temporal lobe epilepsy will finally live seizure-free.

Subject of Research: Glymphatic dysfunction and glucose metabolism as preoperative prognostic markers in drug-resistant temporal lobe epilepsy surgery

Article Title: Preoperative correlations between glymphatic function and glucose metabolism in drug-resistant temporal lobe epilepsy: divergent patterns and postoperative outcomes

Article References: Tang, G., Zeng, C., Cai, Q., Luo, J., Wu, B., Zhang, P., Tang, Y., Li, S., Gong, H., Jiang, Y., Wu, H., Chen, G., Guo, Q., Xu, H., & Ling, X. (2026). Preoperative correlations between glymphatic function and glucose metabolism in drug-resistant temporal lobe epilepsy: divergent patterns and postoperative outcomes. BMC Medicine. https://doi.org/10.1186/s12916-026-05243-7

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05243-7

Keywords: temporal lobe epilepsy, glymphatic system, DTI-ALPS, FDG-PET, glucose metabolism, epilepsy surgery, surgical outcome, diffusion tensor imaging, metabolic network, seizure freedom, neuroimaging, drug-resistant epilepsy

In a study published in the International Journal of Computer Assisted Radiology and Surgery, Jingjing Peng of King’s College London and colleagues, working with neurosurgeons at the National Hospital for Neurology and Neurosurgery and King’s College Hospital, introduced NeuralShift, a U-Net-based deep learning framework that predicts a dense brain displacement field using only preoperative MRI and the side of the planned temporal lobe resection. The work targets epilepsy surgery, in which patients with drug-resistant seizures undergo removal of a temporal lobe to eliminate the seizure focus. Because these procedures follow relatively standardised corridors and positioning, they provided an ideal proving ground for a data-driven model of intraoperative deformation.

The clinical stakes are considerable. Neuronavigation aligns preoperative images with the patient’s anatomy in the operating theatre, guiding the surgeon toward the resection target while avoiding critical structures such as the optic radiation, the fibre bundle that carries visual information and sits dangerously close to the surgical corridor. When the brain shifts, the map becomes wrong, and the surgeon may be navigating a phantom. Intraoperative MRI can capture the deformation directly, but it is expensive, time-consuming, disruptive to the surgical workflow, and available in only a minority of operating rooms worldwide. Intraoperative ultrasound offers a faster and cheaper alternative but produces lower-contrast, operator-dependent images that provide limited objective information about deeper structures.

NeuralShift takes a deliberately different route. Rather than attempting to image the brain during surgery, the model learns a cohort-level prior: the statistical pattern of how temporal lobes deform after resection in this patient population. The architecture is built on U-Net, a convolutional network originally designed for biomedical image segmentation, and it produces three outputs simultaneously: a dense three-dimensional displacement field that warps the preoperative scan toward the intraoperative configuration, a prediction of the deformed brain mask, and the corresponding signed distance function, a continuous encoding of the distance to the brain boundary that supplies strong geometric constraints on global shape.

One of the study’s most technically interesting choices lies in how the network was trained. Voxel-wise physical measurements of brain deformation are essentially impossible to obtain, so the researchers used a registration-based surrogate supervision strategy. Each preoperative MRI was registered to its paired post-resection intraoperative MRI using the Fast Free-Form Deformation algorithm implemented in NiftyReg, and the resulting deformation field served as the regression target. The loss function was correspondingly sophisticated, combining a Cartesian mean squared error on displacement vectors with an auxiliary spherical-coordinate loss that penalises errors in the direction and magnitude of each displacement independently, alongside Dice and edge losses for the mask and a mean squared error for the signed distance function. This multi-task objective forces the network to respect both local deformation physics and global brain geometry at the same time.

Careful preprocessing was required to make the paired scans comparable. Intraoperative images contain a surgical cavity and intensity heterogeneity that break standard registration tools, so the team designed a bespoke pipeline. Anatomists manually identified three landmarks on each scan, the anterior commissure, posterior commissure, and intercomissural height, defining a subject-specific coordinate system that standardised orientation before both images were rigidly and affinely registered into the Montreal Neurological Institute template space. Critically, the model was given only a binary hemisphere indicator encoding resection laterality, deliberately excluding any information about the resection cavity itself, because using the cavity segmented from the intraoperative scan would constitute target leakage, revealing an outcome that cannot be known before surgery.

The evaluation was conducted with notable methodological rigour for a feasibility study. From 98 paired preoperative and intraoperative MRI cases, eight were reserved as a fixed validation set for checkpoint selection, and the remaining 90 were divided into nine disjoint test folds of ten cases each. Nine independently initialised models were trained, each evaluated on a fold it had never seen, yielding one out-of-fold prediction for every subject. The results showed consistent improvement: the predicted brain masks achieved an unweighted mean Dice score of 0.97 across folds, up from 0.93 for the undeformed preoperative anatomy, while registration-referenced target registration error at clinically chosen landmarks, including the frontal operculum, the insula, the aqueduct of Sylvius, the corpus callosum, and the optic tract, fell from a range of 1.46 to 4.76 millimetres down to 1.12 to 3.05 millimetres after applying the predicted field.

The authors are unusually candid about the limits of these numbers. Because the landmark reference positions were generated by the same registration algorithm used as the training target, the error reduction measures how faithfully the network reproduces the registration surrogate, not independently verified physical accuracy. The Dice scores describe overlap with masks produced within the study’s own preprocessing workflow rather than manually delineated reference standards. No patient-specific biomechanical model was run on the same cohort, so no claims of equivalence or superiority to physics-based approaches are made. The study is also confined to a single centre with a homogeneous surgical population, meaning the learned deformation prior may not transfer to tumour resections, where oedema, craniotomy geometry, and lesion diversity introduce far greater variability.

Even so, the conceptual contribution is significant. Biomechanical models of brain shift, typically built on finite element methods, require assumptions about tissue elasticity, boundary conditions, loading, and resection geometry, along with a numerical solve that can burden the surgical workflow. NeuralShift, once trained, produces a deformation prior through a single feed-forward pass, without patient-specific meshing or parameter assignment. The trade-off is the absence of explicit physical guarantees and dependence on the deformation patterns represented in the training cohort. The authors position the two paradigms as complementary rather than competing, and they outline a compelling hybrid future in which the learned prior is updated in real time with sparse intraoperative observations such as ultrasound or tracked cortical surface points, preserving rapid inference while adapting to the actual surgical state.

The road ahead includes prospectively recorded resection plans that could localise expected tissue removal more precisely than a hemisphere indicator, explicit separation of true mechanical deformation from the volume change of the resection cavity itself, uncertainty quantification, and multi-centre validation with independently delineated intraoperative surfaces. The code will be released publicly after acceptance. If those steps succeed, the idea that a neural network can anticipate the brain’s silent drift before surgery begins could mark a turning point in image-guided neurosurgery, turning one of the operating room’s oldest frustrations into a solved prediction problem.

Subject of Research: Deep learning prediction of intraoperative brain shift in temporal lobe resection for epilepsy surgery

Article Title: From pre- to intra-operative MRI: predicting brain shift in temporal lobe resection for epilepsy surgery

Article References: Peng, J., Fiore, G., Liu, Y., Ellum, K., Dasgupta, D., Ashkan, K., McEvoy, A., Miserocchi, A., Ourselin, S., Duncan, J., & Granados, A. (2026). From pre- to intra-operative MRI: predicting brain shift in temporal lobe resection for epilepsy surgery. International Journal of Computer Assisted Radiology and Surgery. https://doi.org/10.1007/s11548-026-03794-x

Image Credits: AI Generated

DOI: 10.1007/s11548-026-03794-x

Keywords: brain shift, NeuralShift, U-Net, intraoperative MRI, preoperative MRI, epilepsy surgery, temporal lobe resection, neuronavigation, deep learning, displacement field, target registration error, computer-assisted surgery