AI-powered chest X-ray pneumonia detection with 86% accuracy and 96.4% sensitivity, validated on an independent (cross-operator) cohort of 485 pediatric samples. Built with TensorFlow & FastAPI.

Systematic evaluation of hallucination risks in Large Language Models (GPT-4, Claude 3, Gemini Pro) for clinical proteomics and mass spectrometry interpretation. Production-ready detection framework with comprehensive benchmarks.

Ethnic bias analysis in medical imaging AI: Demonstrating that explainable-by-design models achieve 80% bias reduction across 5 ethnic groups (50k images)

In-hospital cardiac mortality predictor trained on 15,757 real Indian patient records. Achieves AUC 0.977 using only routine blood tests & echo. Nurse-ready Streamlit web dashboard works on any phone in <3 sec. Discovered alcohol J-curve automatically. Ready for real ward deployment to save lives today.

A clinical-grade malaria parasite detection system using YOLOv8 object detection with 99.14% mAP50 performance.

Production-grade clinical intelligence system speedrun: 95% accuracy, full explainability, zero pretrained models. Trained on class-weighted data, tested with real anxiety, deployed to impress recruiters.

This repository provides a compact RAG evaluation harness tailored for clinical + genomic use cases. It operates on de-identified synthetic notes and curated genomic snippets, measures retrieval quality and grounding/faithfulness, and produces JSONL/CSV/HTML reports.

Clinical + Genomic **RAG evaluation (pro)** with hybrid retrieval (BM25 + embeddings), stronger faithfulness, YAML configs, and HTML dashboards. Python **3.10+**.

Clinical decision support system for chest X-ray analysis with explainable AI. Deep learning + Grad-CAM for transparent medical diagnosis. PyTorch | Medical Imaging | DICOM

A multi-agent healthcare system that automates patient intake, triage, lab waiting, and clinician briefings using Google ADK.

Oloche's AI Oncologist is a diagnostic tool that detects breast cancer in its early stage using ML. The system analyzes measurements from breast mass characteristics to classify tumors as malignant or benign with high accuracy, providing medical professionals with a powerful decision-support tool

This project is designed for a simple LLM agent with tools on a clinical knowledge graph in Neo4j.