EpiScan — Cancer Detection
Built with TensorFlow, Scikit-Learn, Python
The Problem
Early cancer detection significantly improves treatment outcomes, but manual analysis of medical images is slow and error-prone. Machine learning can assist pathologists by providing fast, consistent second-opinion classifications with high accuracy.
The Solution
EpiScan is a machine learning model for cancer detection that achieves 92% accuracy on medical image classification tasks. Built with TensorFlow and Scikit-Learn, the model uses transfer learning, data augmentation, and ensemble techniques to classify medical images. Comprehensive data preprocessing with Pandas and NumPy ensures clean training data for reliable predictions.
Key Features
92% classification accuracy
Deep learning model architecture
Medical image classification
Data augmentation pipeline
Transfer learning approach
Comprehensive data preprocessing
Technology Stack
Related Projects
AI Document Search (RAG)
RAG-based chatbot for PDF queries using LangChain and OpenAI APIs, delivering document-grounded, accurate responses.
React, FastAPI, OpenAI, LangChain, FAISS
Brain Blood Clot Detection
Deep learning model for detecting blood clots in brain CT scans using convolutional neural networks and image segmentation techniques.
TensorFlow, OpenCV, Python, Deep Learning