Neurosurgeon.
Scientist.
Builder.

The nodose ganglion contains sensory neurons that carry information from our organs to the brain. During my PhD at the Feinstein Institutes, I have been working to understand how those cells and their fibers are organized in three dimensions.

Working with my co-PIs, Stavros Zanos and Eric H. Chang, I developed a workflow spanning human tissue preparation, clearing, immunostaining, whole-specimen light-sheet microscopy and computational reconstruction. I develop image-analysis tools to trace fibers and map cells, alongside innovative 3D-printing applications for light-sheet microscopy.

This is hands-on work across dissection, advanced imaging and computational anatomy. It has grown into studies of microanatomy and surgical access, with invited presentations at Rockefeller University and the Center for Pain Studies at UT Dallas.

Alongside the microscopic work, I wanted to map the vagus nerve at the scale a surgeon encounters it: where its branches leave the main trunk, which organs they reach, and how that anatomy varies between people.

With Stavros Zanos and our team, I turned thousands of anatomical branch observations into a standardized, surgically relevant map. The project brings together human dissection, quantitative anatomy and software in a Brain Stimulation paper, open data and analysis, and an interactive atlas that others can explore.

NIH SPARC invited a dedicated seminar on the atlas. Together with the nodose and fiber-reconstruction work, it forms the core of my PhD: understanding human vagus anatomy from organ-specific branches down to cells and fibers.

Human Vagus Nerve AtlasOpen full screen

Explore hereInteractive viewer

If the viewer does not load here, use “Open full screen.”

My neurosurgery thesis at Hacettepe developed a comprehensive volumetric analysis of the brain and its cerebrospinal-fluid spaces in idiopathic normal-pressure hydrocephalus (iNPH), combining quantitative MRI with detailed 3D anatomical models.

Together with Brazilian researchers, we later extended the dataset and analysis pipeline into three presentations at the Hydrocephalus Society conference, with one of our collaborative projects receiving an award. The methods also remain in use in Turkey.

I later built Cistern Explorer around those anatomical reconstructions, combining 3D anatomy and anatomical sections with an exploratory CSF-flow model. Place a catheter, introduce a tracer and inspect its spread through the modeled spaces. The fluid simulation is a later extension of the original thesis work.

This is a research demonstration for exploring model behavior, rather than a validated prediction of an individual patient’s CSF physiology.

Cistern ExplorerOpen full screen

Explore hereInteractive viewer

If the viewer does not load here, use “Open full screen.”

With Netanel Ben-Shalom, I developed an automated 3D measurement and topographic analysis method to quantify cranial deformity and its correction after cranioplasty. The aim was to give surgeons a reproducible way to assess changes in shape alongside their visual judgment.

Our paper, Automated 3D Measurement and Topographic Representation for Quantitative Assessment of Deformity Correction in Cranioplasty, is accepted in Operative Neurosurgery. I presented the work orally at CNS and at the neuroplastic surgery meeting at Lenox Hill Hospital.

I built Neuroplastic Anatomy to explore the layered anatomy of the scalp and head, and turn a chosen anatomical view into a surgical illustration. It grew from my collaboration with the Society for Neuroplastic Surgery.

Rotate the model, reveal or hide structures, adjust labels and separate the layers to understand how they fit together. The drawing workspace uses the arranged 3D view as the starting point for an anatomical illustration.

Neuroplastic AnatomyOpen full screen

Explore hereInteractive viewer

If the viewer does not load here, use “Open full screen.”

With Netanel Ben-Shalom at Lenox Hill Hospital, whose work includes transcranioplasty ultrasound through sonolucent implants, I built a rapid ultrasound annotation and segmentation tool. It trains random forests from sparse annotations, designed around the limited input a clinician could realistically provide in the operating room.

My contribution was hands-on implementation: turning the annotation workflow and machine-learning approach into a usable research tool. The manuscript has been submitted to a journal and to medRxiv. I also contribute to the group’s broader transcranioplasty-ultrasound research.

I built an interactive imaging workspace that brings registered MRI and intraoperative ultrasound into a shared 3D scene, alongside reference white-matter pathways and cortical regions. Linked views let researchers inspect the ultrasound window in its wider anatomical context.

The example below combines RESECT MRI/ultrasound with mapped population-reference tractography and HCP-MMP cortical regions. The pathways and regions provide anatomical context; they are not structures directly measured by ultrasound.

MRI–Ultrasound ExplorerOpen full screen

Explore hereInteractive viewer

If the viewer does not load here, use “Open full screen.”

With A. Gabriella Wernicke and Netanel Ben-Shalom, I built BrachyAtlas to combine anatomy, imaging, implant geometry and radiation dose in one patient-specific intracranial brachytherapy workflow.

It integrates those inputs with an AI agent that helps the clinician work through complex anatomical and connectomic reports, bringing modeled radiation exposure and the surrounding brain anatomy into a shared view for research planning.

With Randy D’Amico’s brain-tumor group, I build tools to examine patient-specific connectivity anomalies and how they change across scans, surgery and treatment.

The Connectomic Anomaly Explorer brings connectivity matrices, regional and network-level measurements, and longitudinal comparisons into an interface the research team can explore directly.

In 2024, I co-authored a Clinical Neurology and Neurosurgery study evaluating ChatGPT on brain-tumor diagnosis and treatment planning. I built the API-based system that sent clinical questions to the model and captured its answers for independent assessment by practicing neurosurgeons.

With Randy D’Amico, I later moved from benchmarking performance to a harder question: what happens when AI becomes part of the clinical cognitive system? That became our 2026 Neurosurgery letter on the extended mind and the uneven frontier of clinical AI.

We have also submitted a review to Neurosurgery on how AI studies should handle data quality, leakage, validation, interpretability, reproducibility and clinical meaning.