Duke University · Class of 2028
Nesim Benba.
ML research in healthcare, medical imaging, and modeling uncertainty.
I study Electrical & Computer Engineering, Mathematics, and Computer Science at Duke, with a focus on learning from complex data and reasoning under uncertainty.
01 / Research & projects
Learning, localization & uncertainty.
Healthcare ML · Siemens Healthineers
3D anatomical localization for surgical imaging
An automated detection pipeline for locating and orienting vertebrae in volumetric surgical scans, supporting preoperative anatomical localization.
Unsupervised learning · Modeling uncertainty
Volatility Regime Detection with Spectral Diffusion Clustering
Recovering low-dimensional market regime structure from noisy, high-dimensional financial signals using diffusion maps and spectral clustering.
Deep reinforcement learning · Duke Research
Deep RL for Portfolio Optimization
Learning adaptive decisions under latent volatility states, combining probabilistic inference, heavy-tailed return models, and reinforcement learning.
Personalized health · Applied ML & IoT
Somna — Sleep Condition Optimizer
Personalized sleep-condition predictions that connect physiological time-series with live environmental sensing and automated room control.
02 / A little about me
I’m interested in ML that can interpret medical images and make decisions when the world is uncertain.
My work spans 3D anatomical localization at Siemens Healthineers, probabilistic regime modeling and deep reinforcement learning at Duke, and unsupervised volatility detection with diffusion maps. Beyond the work: football, jazz, philosophy, poker, and travel.
More about my experience