Editorial: Artificial intelligence in digital pathology image analysis
Yi Liu, Xiaoyan Liu, Hantao Zhang, Junlin Liu, Chaofan Shan, Yinglu Guo, Xun Gong, Min Tang · Frontiers in Bioinformatics · 2023
Artificial intelligence in digital pathology image analysisIn the 21st century, cancer is the top cause of death in hospitals and the key limitation of life expectancy in most countries (Luo et al.).The analysis of medical images including histopathological slides, radiological images such as magnetic resonance imaging (MRI) and CT, and ultrasound images, etc. is an essential tool in cancer research, disease diagnosis and treatment.Moreover, the availability of faster networks and cheaper storage solutions make these images easier to manage and share, leading to the emergence of digital pathology images, for example, whole slide imaging (WSI).However, extracting important information from these images for clinical use requires a big effort from pathologist and is also errorprone due to inexperience and fatigue.Recently, Artificial intelligence (AI) such as deep learning (DL) shows clear potential to mine image features from medical images, better quantitative model disease appearance and hence possibly improve prediction of disease aggressiveness and patient outcome.The application of AI not only reduces the burden on pathologists but also saves high costs and time, thus attracting great attention.In this editorial, we presented an account of how AI has greatly facilitated digital pathology image analysis as well as other medical image analysis.This editorial is based on 11 research articles, 1 regular review and a methods article, shedding light on the power of AI to analyze medical images including but not limited to magnetic resonance imaging (MRI), CT images, and digital pathology images, primarily WSI.Five research articles use machine learning to construct prediction models based on pathological and radiological images. Wang D. et al. developed an automated machinelearning framework for predicting IDH1 mutation status in glioma. In their framework, a random forest algorithm is applied to select relevant features in regions of interest (ROIs) extracted from high-resolution pathology slides and multi-sequence MRI scans. The model integrating histopathological and radiological information can predict glioma IDH genotype with greater accuracy and reliability (Wang D. et al.).Zhao F. et al. used random forest combined with hyperparameter tuning for feature selection and radiomics prediction modeling to distinguish invasive adenocarcinoma (IAC) and minimal invasive adenocarcinoma (MIA) presenting as ground-glass nodules (GGNs). The result of ROC curve showed that their model effectively distinguished IAC from MIA presenting as GGNs and represented a non-invasive, low-cost, rapid, and reproducible preoperative prediction method for clinical application (Zhao F. et al.).In Wang X. et al.'s study, Mann-Whitney U