Few-shot learning based histopathological image classification of colorectal cancer  

作　　者：Rui Li Xiaoyan Li Hongzan Sun Jinzhu Yang Md Rahaman Marcin Grzegozek Tao Jiang Xinyu Huang Chen Li 

机构地区：[1]Key Laboratory of Intelligent Computing in Medical Image,Ministry of Education,Northeastern University,Shenyang,Liaoning 110167,China [2]Cancer Hospital,China Medical University,Shenyang,Liaoning 110122,China [3]Shengjing Hospital,China Medical University,Shenyang.Liaoning 110000,China [4]Institute for Medical Informatics,University of Luebeck,Germany [5]Department of Knowledge Engineering,University of Economics in Katowice,Poland [6]School of Intelligent Medicine,Chengdu University of Traditional Chinese Medicine,Chengdu,Sichuan 610075,China [7]International Joint Institute of Robotics and Intelligent Systems,Chengdu University of Information Technology,Chengdu,Sichuan 610225,China [8]School of Computer Science and Engineering,University of New South Wales,Sydney,NSW 2052,Australia

出　　处：《Intelligent Medicine》2024年第4期256-267,共12页智慧医学(英文)

基　　金：supported by National Natural Science Foundation of China(Grant No.82220108007);Liaoning Province Applied Basic Research Program(Grant No.2023JH2/101600016);Sichuan Science and Technology Planning Project(Grant No.2024YFHZ0320);Special Project for Traditional Chinese Medicine Research of Sichuan Administration of Traditional Chinese Medicine(Grant No.2024zd030).

摘　　要：Background Colorectal cancer is a prevalent and deadly disease worldwide,posing significant diagnostic challenges.Traditional histopathologic image classification is often inefficient and subjective.Although some histopathologists use computer-aided diagnosis to improve efficiency,these methods depend heavily on exten-sive data and specific annotations,limiting their development.To address these challenges,this paper proposes a method based on few-shot learning.Methods This study introduced a few-shot learning approach that combines transfer learning and contrastive learning to classify colorectal cancer histopathology images into benign and malignant categories.The model comprises modules for feature extraction,dimensionality reduction,and classification,trained using a combi-nation of contrast loss and cross-entropy loss.In this paper,we detailed the setup of hyperparameters:n-way,κ-shot,β,and the creation of support,query,and test datasets.Results Our method achieved over 98% accuracy on a query dataset with 35 samples per category using only 10 training samples per category.We documented the model’s loss,accuracy,and the confusion matrix of the results.Additionally,we employed the t-SNE algorithm to analyze and assess the model’s classification performance.Conclusion The proposed model may demonstrate significant advantages in accuracy and minimal data depen-dency,performing robustly across all tested n-way,κ-shot scenarios.It consistently achieved over 93% accuracy on comprehensive test datasets,including 1916 samples,confirming its high classification accuracy and strong generalization capability.Our research could advance the use of few-shot learning in medical diagnostics and also lays the groundwork for extending it to deal with rare,difficult-to-diagnose cases.

关 键 词：Colorectal cancer Few-shot learning Transfer learning Contrastive learning Histopathological images Benign and malignant categories 

分 类 号：R735.34[医药卫生—肿瘤] TP391.41[医药卫生—临床医学]