Cell damage evaluation by intelligent imaging flow cytometry

Abstract

Essential thrombocythemia (ET) is an uncommon situation in which the body produces too many platelets. This can cause blood clots anywhere in the body and results in various symptoms and even strokes or heart attacks. Removing excessive platelets using acoustofluidic methods receives extensive attention due to their high efficiency and high yield. While the damage to the remaining cells, such as erythrocytes and leukocytes is yet evaluated. Existing cell damage evaluation methods usually require cell staining, which are time‐consuming and labor‐intensive. In this paper, we investigate cell damage by optical time‐stretch (OTS) imaging flow cytometry with high throughput and in a label‐free manner. Specifically, we first image the erythrocytes and leukocytes sorted by acoustofluidic sorting chip with different acoustic wave powers and flowing speed using OTS imaging flow cytometry at a flowing speed up to 1 m/s. Then, we employ machine learning algorithms to extract biophysical phenotypic features from the cellular images, as well as to cluster and identify images. The results show that both the errors of the biophysical phenotypic features and the proportion of abnormal cells are within 10% in the undamaged cell groups, while the errors are much greater than 10% in the damaged cell groups, indicating that acoustofluidic sorting causes little damage to the cells within the appropriate acoustic power, agreeing well with clinical assays. Our method provides a novel approach for high‐throughput and label‐free cell damage evaluation in scientific research and clinical settings.

Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment

“…To enhance the diversity of the dataset, we may need to adopt techniques such as pitch shifting [9], time stretching [10], and adding background noise for sound data. Sound data can have features extracted using methods such as Mel Frequency Cepstrum Coefficients (MFCCs).…”

Is AI essential? Examining the need for deep learning in image-activated sorting of Saccharomyces cerevisiae

“…7,8 Among these, imaging flow cytometry has seen considerable integration with AI for automated handling and classification of large volumes of cell image data. 9–18 Traditionally, AI-based cell image classification has been chiefly realized through feature extraction followed by machine learning (ML) to construct classifier algorithms. 9 Preset features ( e.g.…”