Detection of HER2 from Haematoxylin-Eosin Slides Through a Cascade of Deep Learning Classifiers via Multi-Instance Learning

• Published in: Journal of imaging Vol. 6; no. 9; p. 82
• Main Authors: La Barbera, David, Polónia, António, Roitero, Kevin, Conde-Sousa, Eduardo, Della Mea, Vincenzo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 23.08.2020; MDPI
• Subjects: Algorithms; Artificial neural networks; Biopsy; Breast cancer; Classification; Classifiers; convolutional neural networks; Datasets; Decision making; Deep learning; deep learning classification; digital pathology; Digitization; HER2; Machine learning; Metastasis; Morphology; multiple instance learning; Researchers; Systems design; whole slide image processing; whole slide image processing; convolutional neural networks; HER2; digital pathology; deep learning classification; multiple instance learning
• ISSN: 2313-433X; 2313-433X
• DOI: 10.3390/JIMAGING6090082

Breast cancer is the most frequently diagnosed cancer in woman. The correct identification of the HER2 receptor is a matter of major importance when dealing with breast cancer: an over-expression of HER2 is associated with aggressive clinical behaviour; moreover, HER2 targeted therapy results in a significant improvement in the overall survival rate. In this work, we employ a pipeline based on a cascade of deep neural network classifiers and multi-instance learning to detect the presence of HER2 from Haematoxylin-Eosin slides, which partly mimics the pathologist's behaviour by first recognizing cancer and then evaluating HER2. Our results show that the proposed system presents a good overall effectiveness. Furthermore, the system design is prone to further improvements that can be easily deployed in order to increase the effectiveness score.