Deep self‐supervised transformation learning for leukocyte classification. Issue 3 (2nd December 2022)

• Record Type: Journal Article
• Authors: Chen, Xinwei; Zheng, Guolin; Zhou, Liwei; Li, Zuoyong; Fan, Haoyi
• Is Part Of: Journal of biophotonics. Volume 16:Issue 3(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-02
• Subjects: deep learning -- image transformation -- leukocyte classification -- self‐supervised learning; Photonics -- Periodicals; Optical materials -- Periodicals; Optics -- Periodicals; Medical instruments and apparatus -- Periodicals; 621.3605
• DOI: 10.1002/jbio.202200244
• Languages: English
• ISSNs: 1864-063X

Abstract

Abstract: The scarcity of training annotation is one of the major challenges for the application of deep learning technology in medical image analysis. Recently, self‐supervised learning provides a powerful solution to alleviate this challenge by extracting useful features from a large number of unlabeled training data. In this article, we propose a simple and effective self‐supervised learning method for leukocyte classification by identifying the different transformations of leukocyte images, without requiring a large batch of negative sampling or specialized architectures. Specifically, a convolutional neural network backbone takes different transformations of leukocyte image as input for feature extraction. Then, a pretext task of self‐supervised transformation recognition on the extracted feature is conducted by a classifier, which helps the backbone learn useful representations that generalize well across different leukocyte types and datasets. In the experiment, we systematically study the effect of different transformation compositions on useful leukocyte feature extraction. Compared with five typical baselines of self‐supervised image classification, experimental results demonstrate that our method performs better in different evaluation protocols including linear evaluation, domain transfer, and finetuning, which proves the effectiveness of the proposed method. Abstract : Scarcity of training annotations hinders the development of deep learning techniques in medical image analysis. And we found that the common spatial and geometric transformations (e.g., image rotation) are not applicable to leukocyte images. Therefore, we use appearance‐modified data augmentation strategy and apply them to our proposed simple and efficient self‐supervised method for leukocyte classification by identifying the different transformations of leukocyte images.