Abstract
Tissue biomarker scoring by pathologists is central to defining the appropriate therapy for patients with cancer. Yet, inter-pathologist variability in the interpretation of ambiguous cases can affect diagnostic accuracy. Modern artificial intelligence methods such as deep learning have the potential to supplement pathologist expertise to ensure constant diagnostic accuracy. We developed a computational approach based on deep learning that automatically scores HER2, a biomarker that defines patient eligibility for anti-HER2 targeted therapies in breast cancer. In a cohort of 71 breast tumour resection samples, automated scoring showed a concordance of 83% with a pathologist. The twelve discordant cases were then independently reviewed, leading to a modification of diagnosis from initial pathologist assessment for eight cases. Diagnostic discordance was found to be largely caused by perceptual differences in assessing HER2 expression due to high HER2 staining heterogeneity. This study provides evidence that deep learning aided diagnosis can facilitate clinical decision making in breast cancer by identifying cases at high risk of misdiagnosis.
Highlights
Computer-aided diagnosis holds great promise to facilitate clinical decision making in personalised oncology
The objectives of this study were (1) to evaluate the ability of a convolutional neural network (ConvNets) model to automatically recognize cancer cell types compared to classical machine learning techniques, (2) to evaluate the performance of ConvNets to provide accurate human epidermal growth factor receptor 2 (HER2) status reviews in clinically realistic conditions and (3) to assess the potential utility of computer-aided diagnosis to facilitate clinical decision making
To automatically score HER2 expression in tumour cells, we propose a simple approach wherein images are first processed to detect cells and machine learning is subsequently used to classify candidate cells into one of the following categories (Fig. 2): stroma cells, immune cells, tumour cells displaying strong HER2 overexpression (3+cells), tumour cells displaying moderate HER2 overexpression (2+cells), tumour cells displaying weak HER2 overexpression (1+cells), tumour cells displaying no HER2 overexpression (0 cells) and artefacts
Summary
Computer-aided diagnosis holds great promise to facilitate clinical decision making in personalised oncology. Despite evidence that image analysis improves IHC biomarker scoring accuracy and reproducibility in tumours[8,10,15], the adoption of computer-aided diagnosis by pathologists has remained limited in practice This can be explained by limited evidence of added clinical value and by the surplus of time required to predefine tumour regions in the tissue sample[16]. The objectives of this study were (1) to evaluate the ability of a convolutional neural network (ConvNets) model to automatically recognize cancer cell types compared to classical machine learning techniques, (2) to evaluate the performance of ConvNets to provide accurate HER2 status reviews in clinically realistic conditions and (3) to assess the potential utility of computer-aided diagnosis to facilitate clinical decision making
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have