medsync-ai-weights

MedSync AI Trained Model Weights

This repository contains the official, verified, and validated trained model weights for the MedSync AI inference microservice.

These are specifically trained MedSync project models.

Disclaimer: These models are prototypes developed for the MedSync project. They have not been clinically validated and should NOT be used for actual medical diagnosis without physician supervision.

Models Included

1. Bone Fracture Detection

• File: bone.pt
• Architecture: YOLO (detection)
• Class Mapping:
  • 0 → fracture

2. Brain Tumour Detection

• File: brain.pt
• Architecture: YOLO (detection)
• Class Mapping:
  • 0 → glioma
  • 1 → meningioma
  • 2 → pituitary

3. Kidney Stone Detection

• File: kidney.pt
• Architecture: YOLO (detection)
• Class Mapping:
  • 0 → kidney-stone

4. Skin Disease Classification

• File: skin_model.pt
• Architecture: PyTorch EfficientNet-B0
• Class Mapping:
  • 0 → acne
  • 1 → eczema
  • 2 → fungal
  • 3 → infection
  • 4 → normal
  • 5 → psoriasis
  • 6 → tumor