Breast Cancer Classification From Histopathological Images Using Patch-Based Deep Learning Modeling
Problem
Accurate detection and classification of breast cancer from histopathology images is critical but difficult due to the complexity of breast tissues.
Why it matters
Misclassification of breast cancer directly affects patient diagnosis and treatment outcomes.
What the AI did
A patch-based deep learning model (Pa-DBN-BC) using a Deep Belief Network with unsupervised pre-training and supervised fine-tuning extracts features from image patches, with logistic regression classifying each patch as cancer or background.
Results
not stated in source
Impact
The evidence supports only that a method was proposed and trained/tested on a whole slide histopathology dataset; no performance numbers are given.
Record
| Venue | IEEE Access |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Irum Hirra, Mubashir Ahmad, Ayaz Hussain, Muhammad Usman Ashraf, Iftikhar Ahmed Saeed, Syed Furqan Qadri, Ahmed Mohammed Alghamdi, Ahmed Saeed Alfakeeh |
| Citations | 272 (OpenAlex, at import time) |
| References | 84 |
| Model or method | Pa-DBN-BC (Deep Belief Network with logistic regression) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1109/access.2021.3056516 |
| Open access | gold |
| Keywords | breast cancer, histopathology, deep belief network, patch-based deep learning, image classification |
Importance rating
significant — thousands to millions affected
Accurate breast cancer detection/classification is critical due to tissue complexity; misclassification affects diagnosis and treatment; 86% accuracy on whole-slide dataset.