RCMNet: A deep learning model assists CAR-T therapy for leukemia
RCMNet: A deep learning model assists CAR-T therapy for leukemia
复制标题
DOI：
10.1016/j.compbiomed.2022.106084
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Peiwu Qin
中科院分区：
文献类型：
--
作者：
Ruitao Zhang;Xueying Han;Zhengyang Lei;Chenyao Jiang;Ijaz Gul;Qiuyue Hu;Shiyao Zhai;Hong Lu;Lijin Lian;Ying Liu;Yongbing Zhang;Yuhan Dong;Canyang Zhang;Tsz Kwan Lam;Yuxing Han;Dongmei Yu;Jin Zhou;Peiwu Qin
Acute leukemia is a type of blood cancer with a high mortality rate. Current therapeutic methods include bone marrow transplantation, supportive therapy, and chemotherapy. Although a satisfactory remission of the disease can be achieved, the risk of recurrence is still high. Therefore, novel treatments are demanding. Chimeric antigen receptor-T (CAR-T) therapy has emerged as a promising approach to treating and curing acute leukemia. To harness the therapeutic potential of CAR-T cell therapy for blood diseases, reliable cell morphological identification is crucial. Nevertheless, the identification of CAR-T cells is a big challenge posed by their phenotypic similarity with other blood cells. To address this substantial clinical challenge, herein we first construct a CAR-T dataset with 500 original microscopy images after staining. Following that, we create a novel integrated model called RCMNet (ResNet18 with Convolutional Block Attention Module and Multi-Head Self-Attention) that combines the convolutional neural network (CNN) and Transformer. The model shows 99.63% top-1 accuracy on the public dataset. Compared with previous reports, our model obtains satisfactory results for image classification. Although testing on the CAR-T cell dataset, a decent performance is observed, which is attributed to the limited size of the dataset. Transfer learning is adapted for RCMNet and a maximum of 83.36% accuracy is achieved, which is higher than that of other state-of-the-art models. This study evaluates the effectiveness of RCMNet on a big public dataset and translates it to a clinical dataset for diagnostic applications. Peripheral Blood Cells (PBC) dataset is used to train our RCMNet model. Blood samples are collected from patients receiving CAR-T therapy. The Cells are stained by May Grünwald-Giemsa for wide field image acquisition. RCMNet is transferred from PBC dataset to our CAR-T cell dataset. • The first dataset for CAR-T cell and non-CAR-T cell classification. • A deep neural network (RCMNet) consisting of CNN and self-attention for classification tasks. • Enhance performance of RCMNet on CAR-T dataset by transfer learning. • RCMNET performs well in identifying CAR-T cell, which shows the potential to aid in clinical decision-making.
