Master list

[haranrk]

Paper skeleton

Abstract

• Model performance on datasets (Liver, Colon, Prostrate(maybe))
• Main content of the paper

Introduction

1. Motivation
   1. Include why
   2. Pathologists aren’t accurate and slow
2. Previous work
   1. Papers (at least 10)
3. Our contribution in brief

Methodology

Data

• [ ] Explain all 3 datasets

Sampling strategies

Currently, doing equal tumor and normal tissue sampling

• [x] Some paper - prostrate cancer - deepmind/googleai - use a different ratio investigate

• [x] Sampling randomly? Or visit some papers on sampling,

Data pre-processing

Stain normalisation

• [x] Augmentations

Network architectures

• [x] Explain 3 architectures in detail

Training strategies

Inference strategies

Interpretability

• [ ] Data flow analysis - gradcam
• [x] Uncertainty analysis -

Results & Discussion

Implementation details

Conclusion

Future work

Acknowledgements

[haranrk]

Experiments to run

• [ ] 9-way ensemble
• [ ] Color-jitter vs Stain normalisation
• [ ] Interpretability - uncertainty
• [ ] Interpretability - gradcam
• [ ] Check transfer learning between the Histopathology domains