Brain Tumor Detection and Classification Using Deep Learning Models
summary
Automated system trains VGG-16, VGG-19 and EfficientNet-B1 (~94% accuracy) on 7,023 MRI images with resizing, normalization and augmentation, using Adam optimizer, early stopping and a train/validation split; reported accuracies are ~88% (VGG-16), ~90% (VGG-19) and ~94% (EfficientNet-B1), with ensemble deployment via Flask/FastAPI loading model files and a simple web UI that demoed 98–99% confidences on example scans. Also integrates a RAG chatbot with Chroma vector DB for medical Q&A from verified documents.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:20Early detection need
- 1:45Solution and system overview
- 2:29Deep learning models
- 3:11How CNNs work
- 4:56VGG models overview
- 6:00EfficientNet B1 advantages
- 7:04Model performance comparison
- 7:25System architecture
- 8:30Backend and Flask app
- 9:11RAG chatbot implementation
- 9:32Training process and optimizer
- 10:15Prediction system and dataset
- 10:36Live demo
- 11:19Code overview
- 12:02Testing with MRI images
- 13:26Chatbot interaction
- 14:30Conclusion and future
related talks
49mBuilding AI That Learns and Adapts: A Case Study in MRI Diagnostics
48mNeural Networks and Brains
43mBuilding AI That Learns and Adapts: A Case Study in MRI Diagnostics
50mBuilding AI That Learns and Adapts: A Case Study in MRI Diagnostics
42mBuilding AI That Learns and Adapts: A Case Study in MRI Diagnostics
25mDeep Learning in Biomedicine: Challenges and Opportunities
29mCancer diagnostics enabled by machines: finding and classifying single tumor....
18mCancer Screening using Deep Learning