Explainable convolutional neural networks for assessing head and neck cancer histopathology

• Published in: Diagnostic pathology Vol. 18; no. 1; pp. 1 - 121
• Main Authors: Dörrich, Marion, Hecht, Markus, Fietkau, Rainer, Hartmann, Arndt, Iro, Heinrich, Gostian, Antoniu-Oreste, Eckstein, Markus, Kist, Andreas M
• Format: Journal Article
• Language: English
• Published: London BioMed Central Ltd 03-11-2023; BioMed Central; BMC
• Subjects: Accuracy; Algorithms; Annotations; Artificial intelligence; Artificial neural networks; Cancer; Classification; Explainable AI; Head & neck cancer; Head and neck cancer; Head and neck carcinoma; Histochemistry; Histopathology; Image analysis; Image processing; Image segmentation; Larynx; Libraries; Localization; Medical imaging; Medical imaging equipment; Medical prognosis; Neural networks; Patients; Semantic segmentation; Squamous cell carcinoma; Tissues; Tumors
• ISSN: 1746-1596; 1746-1596
• DOI: 10.1186/s13000-023-01407-8

Abstract
Purpose
Although neural networks have shown remarkable performance in medical image analysis, their translation into clinical practice remains difficult due to their lack of interpretability. An emerging field that addresses this problem is Explainable AI.
Methods
Here, we aimed to investigate the ability of Convolutional Neural Networks (CNNs) to classify head and neck cancer histopathology. To this end, we manually annotated 101 histopathological slides of locally advanced head and neck squamous cell carcinoma. We trained a CNN to classify tumor and non-tumor tissue, and another CNN to semantically segment four classes - tumor, non-tumor, non-specified tissue, and background. We applied Explainable AI techniques, namely Grad-CAM and HR-CAM, to both networks and explored important features that contributed to their decisions.
Results
The classification network achieved an accuracy of 89.9% on previously unseen data. Our segmentation network achieved a class-averaged Intersection over Union score of 0.690, and 0.782 for tumor tissue in particular. Explainable AI methods demonstrated that both networks rely on features agreeing with the pathologist’s expert opinion.
Conclusion
Our work suggests that CNNs can predict head and neck cancer with high accuracy. Especially if accompanied by visual explanations, CNNs seem promising for assisting pathologists in the assessment of cancer sections.