ResearchStatus: Research
Medical AI Research
University research across medical imaging, health chatbots, and wildfire detection
PythonTensorFlowPyTorchHuggingFaceScikit-Learn
The Problem
Exploring applied AI across healthcare imaging, conversational health guidance, and environmental monitoring.
The Solution & Architecture
Conducted university research projects covering medical imaging analysis, a nasal health chatbot, and a wildfire detection model.
Pipeline Architecture Breakdown
1.DICOM & Medical Imaging Preprocessing -> Spatial normalization, intensity scaling, and data augmentation
2.Convolutional & Transformer Backbones -> Feature extraction with transfer learning fine-tuning
3.Explainability Pipeline -> Grad-CAM and saliency mapping for medical practitioner validation
Core Highlights & Functionality
Medical MRI Scan Classification: Deep learning pipelines for volumetric anomaly detection and segmentation
Specialized Nasal Health Chatbot: NLP pipeline for triaging ENT symptoms and patient guidance
Wildfire Early Detection: Remote sensing and thermal pattern recognition using CNN architectures
Rigorous Validation: Cross-validation, sensitivity analysis, and clinical interpretability maps (Grad-CAM)