Multi-Cancer Classification
Deep-learning CNNs that classify breast and brain tumour scans, separating benign, malignant and healthy cases.
Overview
A computer-vision project that trains convolutional neural networks on medical imaging to detect breast and brain cancers. The dataset is split 70/20/10 into training, validation and test sets, and the notebooks document the model architecture and per-class predictions.
- Year
- 2026
- Area
- AI & Vision
- Role
- Deep learning
Stack
- Python
- Deep Learning
- CNNs
- Jupyter
01
Two domains
Separate classification tasks for breast (benign vs malignant) and brain (tumour vs no tumour) scans.
02
CNN architecture
Custom convolutional models, with parameter summaries documented in the notebook.
03
Clean evaluation
A 70/20/10 train/validation/test split, with visual prediction examples.