Many artificial intelligence models have been developed to predict clinically relevant biomarkers for colorectal cancer (CRC), including microsatellite instability (MSI). However, existing deep learning networks require large training datasets, which are often hard to obtain. In this study, based on the latest Hierarchical Vision Transformer using Shifted Windows (Swin Transformer [Swin‐T]), we developed an efficient workflow to predict biomarkers in CRC (MSI, hypermutation, chromosomal instability, CpG island methylator phenotype, and BRAF and TP53 mutation) that required relatively small datasets. Our Swin‐T workflow substantially achieved the state‐of‐the‐art (SOTA) predictive performance in an intra‐study cross‐validation experiment on the Cancer Genome Atlas colon and rectal cancer dataset (TCGA‐CRC‐DX). It also demonstrated excellent generalizability in cross‐study external validation and delivered a SOTA area under the receiver operating characteristic curve (AUROC) of 0.90 for MSI, using the Molecular and Cellular Oncology dataset for training (N = 1,065) and the TCGA‐CRC‐DX (N = 462) for testing. A similar performance (AUROC = 0.91) was reported in a recent study, using ~8,000 training samples (ResNet18) on the same testing dataset. Swin‐T was extremely efficient when using small training datasets and exhibited robust predictive performance with 200–500 training samples. Our findings indicate that Swin‐T could be 5–10 times more efficient than existing algorithms for MSI prediction based on ResNet18 and ShuffleNet. Furthermore, the Swin‐T models demonstrated their capability in accurately predicting MSI and BRAF mutation status, which could exclude and therefore reduce samples before subsequent standard testing in a cascading diagnostic workflow, in turn reducing turnaround time and costs.
Due to lack of annotated pathological images, transfer learning has been the predominant approach in the field of digital pathology. Pre-trained neural networks based on ImageNet database are often used to extract “off-the-shelf” features, achieving great success in predicting tissue types, molecular features, and clinical outcomes, etc. We hypothesize that fine-tuning the pre-trained models using histopathological images could further improve feature extraction, and downstream prediction performance. We used 100 000 annotated H&E image patches for colorectal cancer (CRC) to fine-tune a pre-trained Xception model via a 2-step approach. The features extracted from fine-tuned Xception (FTX-2048) model and Image-pretrained (IMGNET-2048) model were compared through: (1) tissue classification for H&E images from CRC, same image type that was used for fine-tuning; (2) prediction of immune-related gene expression, and (3) gene mutations for lung adenocarcinoma (LUAD). Five-fold cross validation was used for model performance evaluation. Each experiment was repeated 50 times. The extracted features from the fine-tuned FTX-2048 exhibited significantly higher accuracy (98.4%) for predicting tissue types of CRC compared to the “off-the-shelf” features directly from Xception based on ImageNet database (96.4%) (P value = 2.2 × 10−6). Particularly, FTX-2048 markedly improved the accuracy for stroma from 87% to 94%. Similarly, features from FTX-2048 boosted the prediction of transcriptomic expression of immune-related genes in LUAD. For the genes that had significant relationships with image features (P < 0.05, n = 171), the features from the fine-tuned model improved the prediction for the majority of the genes (139; 81%). In addition, features from FTX-2048 improved prediction of mutation for 5 out of 9 most frequently mutated genes (STK11, TP53, LRP1B, NF1, and FAT1) in LUAD. We proved the concept that fine-tuning the pretrained ImageNet neural networks with histopathology images can produce higher quality features and better prediction performance for not only the same-cancer tissue classification where similar images from the same cancer are used for fine-tuning, but also cross-cancer prediction for gene expression and mutation at patient level.
31 May 2011 in Journal of Pharmacokinetics and Pharmacodynamics