4.8 Article

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

期刊

BIOSENSORS & BIOELECTRONICS
卷 220, 期 -, 页码 -

出版社

ELSEVIER ADVANCED TECHNOLOGY
DOI: 10.1016/j.bios.2022.114865

关键词

Imaging flow cytometry; Image-activated cell sorting; Artificial intelligence; AI inferencing

向作者/读者索取更多资源

Classification and sorting of cells using image-activated cell sorting (IACS) systems can provide valuable insights to biomedical sciences. However, the limited capabilities and complex implementation of deep learning-assisted IACS systems have hindered their widespread adoption in biomedical research. In this study, the authors present a label-free cell sorting solution based on real-time AI inferencing and fast training of a deep learning model. The system achieves high sorting purity with a short latency time, setting a new record for IACS with sorting by AI inference.
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 pop-ulation. 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.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.8
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据