Employing Raman Spectroscopy and Machine Learning for the Identification of Breast Cancer
Problem
Surgeons need to quickly distinguish cancerous breast tissue from healthy tissue during surgery to ensure complete tumor removal.
Why it matters
Incomplete tumor removal risks recurrence, while removing healthy tissue causes unnecessary harm to patients.
What the AI did
Random Forest, SVM, and CNN models classified Raman spectra of normal versus late-stage cancerous mouse mammary tissue, and spectral peaks contributing to classification were identified.
Results
Classification accuracy: 94.47% (RF), 96.76% (SVM), 97.58% (CNN)
Impact
Demonstrates ML-based Raman classification of cancerous tissue in mice; human clinical use not shown.
Record
| Venue | Biological Procedures Online |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Ya Zhang, Zheng Li, Zhongqiang Li, Huaizhi Wang, Dinkar Regmi, Jian Zhang, Ji‐Ming Feng, Shaomian Yao et al. |
| Citations | 15 (OpenAlex, at import time) |
| References | 62 |
| Model or method | Random Forest, Support Vector Machine, Convolutional Neural Network |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1186/s12575-024-00255-0 |
| Open access | gold |
| Keywords | Raman spectroscopy, machine learning, breast cancer, tissue classification, CNN, SVM |
Importance rating
significant — thousands to millions affected
Abstract states breast cancer poses significant health risk to women worldwide and that intraoperative discrimination is crucial for complete tumor removal; validated only in mice so far.