Testing quantum methods for healthcare AI.

My quantum-machine-learning research asks where emerging quantum methods can be evaluated meaningfully in healthcare—and where conventional machine learning remains the stronger reference. The work includes comparative studies of chronic kidney disease and heart-disease prediction, non-invasive glucose and HbA1c estimation, retinal-image biomarkers, and camera-agnostic optical coherence tomography generation. Across these studies, the emphasis is comparative evidence: carefully prepare the data, select features, evaluate quantum and classical alternatives, and report the experimental boundary rather than assuming a quantum advantage.

Empirical studies examine quantum support vector machines, quantum neural networks, and classical baselines in hybrid computing environments. The glucose and HbA1c work connects quantum learning with smartphone-derived photoplethysmography features. Kidney- and heart-disease studies compare prediction performance using public clinical or vital-sign data. Imaging research broadens the program: a systematic review considers the eye as a biomarker for ocular and systemic conditions, while a causal-inference quantum transfer model investigates generation of camera-agnostic OCT images from fundus photographs. A book chapter explores longer-term possibilities for quantum sensing, encryption, and mental-health monitoring in smart healthcare.

These publications span experiments, reviews, and forward-looking concepts, so their claims require different levels of caution. Performance on a selected dataset does not establish clinical utility, hardware efficiency, or general advantage. The research program therefore treats reproducibility, classical benchmarks, dataset scope, computational cost, and external validation as central questions. Its contribution is a documented path for testing quantum healthcare AI against concrete tasks while keeping emerging potential separate from established practice.