CellaScope helped lab staff identify white blood cells on smears

What happened

Beijing Xiaoying Technology built CellaScope, software that uses deep learning to segment and classify white blood cells in images of blood smears. It helps laboratory staff recheck patients whose routine blood count shows abnormal white cells. The Center for Medical Device Evaluation (CMDE) of China's National Medical Products Administration reviewed the company's application for Class III registration. The trial was the maker's own, with 494 cases at Peking Union Medical College Hospital, the Second Xiangya Hospital and Dongyang People's Hospital. Investigators read each smear alone and with the software, and an expert panel set the reference. Accuracy was higher with the software in all nine categories, for example 70.22% against 39.10% for abnormal white cells. In leukemia and suspected leukemia cases, total agreement was 86.44% with the software and 81.17% without it. On 2 November 2022, CMDE found that the product “可提高检验人员识别各类白细胞的准确性” (can improve laboratory staff's accuracy in identifying the various types of white blood cells). It judged that benefits outweigh risks and recommended approval. The report does not say whether the readings guided any patient's care.