This is the code repository for our paper: Weakly supervised identification of microscopic human breast cancer-related optical signatures from normal-appearing breast tissue [https://doi.org/10.1364/BOE.480687].
We propose MM-MIL for the discovery of novel optical signatures when only coarse-grained and ambiguous annotations are available. We applied the proposed method to the investigation of human breast cancer-related optical signatures based on Simultaneous Label-free Autofluorescence Multiharmonic (SLAM) microscopy and unveiled non-obvious cancer-related optical signatures in peri-tumoral regions.
- Ubuntu 18.04
- Python
- Pytorch
- Nvidia GPU + CUDA
- jsondiff
- tdpm
- tifffile
- Captum
(Will be updated)
(Will be updated)
If you use this code and relevant data, please cite our paper:
@article{shi2023weakly,
title={Weakly supervised identification of microscopic human breast cancer-related optical signatures from normal-appearing breast tissue},
author={Shi, Jindou and Tu, Haohua and Park, Jaena and Marjanovic, Marina and Higham, Anna M and Luckey, Natasha N and Cradock, Kimberly A and Liu, Z George and Boppart, Stephen A},
journal={Biomedical Optics Express},
volume={14},
number={4},
pages={1339--1354},
year={2023},
publisher={Optica Publishing Group}
}