Application of transfer learning and ensemble learning in image-level classification for breast histopathology
Application of transfer learning and ensemble learning in image-level classification for breast histopathology作者机构:Microscopic Image and Medical Image Analysis GroupCollege of Medicine and Biological Information EngineeringNortheastern UniversityShenyangLiaoning 110819 China China Medical UniversityShenyangLiaoning 110122 China Institute of Medical InformaticsUniversity of LuebeckLuebeckGermany
出 版 物:《智慧医学(英文)》 (Intelligent Medicine)
年 卷 期:2023年第3卷第2期
页 面:115-128页
核心收录:
学科分类:12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 1002[医学-临床医学] 1001[医学-基础医学(可授医学、理学学位)] 081104[工学-模式识别与智能系统] 08[工学] 100214[医学-肿瘤学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术(可授工学、理学学位)] 10[医学]
基 金:supported by the National Natural Science Foundation of China(Grant No.61806047)
主 题:Convolutional neural network Transfer learning Ensemble learning Image classification Histopathological image Breast cancer
摘 要:Background Breast cancer has the highest prevalence among all cancers in women *** classification of histopathological images in the diagnosis of breast cancers is an area of clinical *** computer-aided diagnosis,most traditional classification models use a single network to extract features,although this approach has significant ***,many networks are trained and optimized on patient-level datasets,ignoring lower-level data *** This paper proposed a deep ensemble model based on image-level labels for the binary classification of breast histopathological images of benign and malignant ***,the BreaKHis dataset was randomly divided into training,validation,and test ***,data augmentation techniques were used to balance the numbers of benign and malignant ***,based on their transfer learning performance and the complementarity between networks,VGG16,Xception,ResNet50,and DenseNet201 were selected as base *** In a ensemble network model with accuracy as the weight,the image-level binary classification achieved an accuracy of 98.90%.To verify the capabilities of our method,it was experimentally compared with the latest transformer and multilayer perception(MLP)models on the same *** ensemble model showed a 5%-20%advantage,emphasizing its far-reaching abilities in classification *** This research focuses on improving the performance of a classification model with an ensemble *** learning has an essential role in classification of small datasets,improving training speed and *** model may outperform many existing approaches with respect to accuracy and has applications in the field of auxiliary medical diagnosis.