Low-latency label-free image-activated cell sorting using fast deep learning and AI inferencing

• Published in: Biosensors & bioelectronics 2023-01, Vol.220, p.114865-114865, Article 114865
• Main Authors: Tang, Rui, Xia, Lin, Gutierrez, Bien, Gagne, Ivan, Munoz, Adonary, Eribez, Korina, Jagnandan, Nicole, Chen, Xinyu, Zhang, Zunming, Waller, Lauren, Alaynick, William, Cho, Sung Hwan, An, Cheolhong, Lo, Yu-Hwa
• Format: Article
• Language: English
• Subjects: AI inferencing; Artificial intelligence; Image-activated cell sorting; Imaging flow cytometry
• ISSN: 0956-5663; 1873-4235
• DOI: 10.1016/j.bios.2022.114865

Classification and sorting of cells using image-activated cell sorting (IACS) systems can bring significant insight to biomedical sciences. Incorporating deep learning algorithms into IACS enables cell classification and isolation based on complex and human-vision uninterpretable morphological features within a heterogeneous cell population. However, the limited capabilities and complicated implementation of deep learning–assisted IACS systems reported to date hinder the adoption of the systems for a wide range of biomedical research. Here, we present image-activated cell sorting by applying fast deep learning algorithms to conduct cell sorting without labeling. The overall sorting latency, including signal processing and AI inferencing, is less than 3 ms, and the training time for the deep learning model is less than 30 min with a training dataset of 20,000 images. Both values set the record for IACS with sorting by AI inference. . We demonstrated our system performance through a 2-part polystyrene beads sorting experiment with 96.6% sorting purity, and a 3-part human leukocytes sorting experiment with 89.05% sorting purity for monocytes, 92.00% sorting purity for lymphocytes, and 98.24% sorting purity for granulocytes. The above performance was achieved with simple hardware containing only 1 FPGA, 1 PC and GPU, as a result of an optimized custom CNN UNet and efficient use of computing power. The system provides a compact, sterile, low-cost, label-free, and low-latency cell sorting solution based on real-time AI inferencing and fast training of the deep learning model.