PPsNet: An improved deep learning model for microsatellite instability high prediction in colorectal cancer from whole slide images. (October 2022)

• Record Type: Journal Article
• Authors: Lou, Jingjiao; Xu, Jiawen; Zhang, Yuyan; Sun, Yuhong; Fang, Aiju; Liu, Jixuan; Mur, Luis A.J.; Ji, Bing
• Is Part Of: Computer methods and programs in biomedicine. Volume 225(2022)
• Publication Date: 2022-10
• Subjects: Colorectal cancer -- Microsatellite instability -- Deep learning -- Deep supervision -- Pair-wise learning -- Synergic network -- Parameter sharing -- Whole slide images; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2022.107095
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: An improved deep learning model is built to predict MSI-H in CRC from whole slide images through tumor detection and MSI classification. Deep supervision is firstly introduced into fine-tuning for performance improvements of the pre-training network. A parameter partial sharing strategy is adopted to reduce the heavy burden of memory cost due to large model parameters. The proposed model is validated on self-collected Asian-CRC datasets from three hospitals in China. Experimental results demonstrate that our proposed model yields better performances with an accuracy of 87.28% and AUC of 94.29% than other state-of-the-art methods. Abstract: Background and Objective: Recent studies have shown that colorectal cancer (CRC) patients with microsatellite instability high (MSI-H) are more likely to benefit from immunotherapy. However, current MSI testing methods are not available for all patients due to the lack of available equipment and trained personnel, as well as the high cost of the assay. Here, we developed an improved deep learning model to predict MSI-H in CRC from whole slide images (WSIs). Methods: We established the MSI-H prediction model based on two stages: tumor detection and MSI classification. Previous works applied fine-tuning strategy directly for tumor detection, but ignoring the challenge of vanishing gradient due to the large number of convolutional layers. We added auxiliary classifiers to intermediate layers of pre-trained models to help propagate gradients back through in an effective manner. To predict MSI status, we constructed a pair-wise learning model with a synergic network, named parameter partial sharing network (PPsNet), where partial parameters are shared among two deep convolutional neural networks (DCNNs). The proposed PPsNet contained fewer parameters and reduced the problem of intra-class variation and inter-class similarity. We validated the proposed model on a holdout test set and two external test sets. Results: 144 H&E-stained WSIs from 144 CRC patients (81 cases with MSI-H and 63 cases with MSI-L/MSS) were collected retrospectively from three hospitals. The experimental results indicate that deep supervision based fine-tuning almost outperforms training from scratch and utilizing fine-tuning directly. The proposed PPsNet always achieves better accuracy and area under the receiver operating characteristic curve (AUC) than other solutions with four different neural network architectures on validation. The proposed method finally achieves obvious improvements than other state-of-the-art methods on the validation dataset with an accuracy of 87.28% and AUC of 94.29%. Conclusions: The proposed method can obviously increase model performance and our model yields better performance than other methods. Additionally, this work also demonstrates the feasibility of MSI-H prediction using digital pathology images based on deep learning in the Asian population. It is hoped that this model could serve as an auxiliary tool to identify CRC patients with MSI-H more time-saving and efficiently.