TY - JOUR
T1 - Label-Free Identification of White Blood Cells Using Machine Learning
AU - Nassar, Mariam
AU - Doan, Minh
AU - Filby, Andrew
AU - Wolkenhauer, Olaf
AU - Fogg, Darin K.
AU - Piasecka, Justyna
AU - Thornton, Catherine A.
AU - Carpenter, Anne E.
AU - Summers, Huw D.
AU - Rees, Paul
AU - Hennig, Holger
N1 - Funding Information:
The authors thank Tim Becker and Markus Wolfien for the helpful discussions. Funding was provided in part by grants from the National Institutes of Health (MIRA R35 GM122547 to A.E.C.), BBSRC BB/P026818/1 to P.R. and H.D.S. and the National Science Foundation/BBSRC joint program (NSF DBI 1458626 and BBSRC BB/N005163/1 to A.E.C. and P.R.)
Publisher Copyright:
© 2019 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.
PY - 2019
Y1 - 2019
N2 - White blood cell (WBC) differential counting is an established clinical routine to assess patient immune system status. Fluorescent markers and a flow cytometer are required for the current state-of-the-art method for determining WBC differential counts. However, this process requires several sample preparation steps and may adversely disturb the cells. We present a novel label-free approach using an imaging flow cytometer and machine learning algorithms, where live, unstained WBCs were classified. It achieved an average F1-score of 97% and two subtypes of WBCs, B and T lymphocytes, were distinguished from each other with an average F1-score of 78%, a task previously considered impossible for unlabeled samples. We provide an open-source workflow to carry out the procedure. We validated the WBC analysis with unstained samples from 85 donors. The presented method enables robust and highly accurate identification of WBCs, minimizing the disturbance to the cells and leaving marker channels free to answer other biological questions. It also opens the door to employing machine learning for liquid biopsy, here, using the rich information in cell morphology for a wide range of diagnostics of primary blood.
AB - White blood cell (WBC) differential counting is an established clinical routine to assess patient immune system status. Fluorescent markers and a flow cytometer are required for the current state-of-the-art method for determining WBC differential counts. However, this process requires several sample preparation steps and may adversely disturb the cells. We present a novel label-free approach using an imaging flow cytometer and machine learning algorithms, where live, unstained WBCs were classified. It achieved an average F1-score of 97% and two subtypes of WBCs, B and T lymphocytes, were distinguished from each other with an average F1-score of 78%, a task previously considered impossible for unlabeled samples. We provide an open-source workflow to carry out the procedure. We validated the WBC analysis with unstained samples from 85 donors. The presented method enables robust and highly accurate identification of WBCs, minimizing the disturbance to the cells and leaving marker channels free to answer other biological questions. It also opens the door to employing machine learning for liquid biopsy, here, using the rich information in cell morphology for a wide range of diagnostics of primary blood.
KW - high-content analysis
KW - imaging flow cytometry
KW - label-free classification
KW - liquid biopsy
KW - lymphocytes
KW - machine learning
KW - personalized medicine
KW - white blood cell count
KW - white blood cells
UR - http://www.scopus.com/inward/record.url?scp=85065730623&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85065730623&partnerID=8YFLogxK
U2 - 10.1002/cyto.a.23794
DO - 10.1002/cyto.a.23794
M3 - Article
C2 - 31081599
AN - SCOPUS:85065730623
VL - 95
SP - 836
EP - 842
JO - Cytometry Part A
JF - Cytometry Part A
SN - 1552-4922
IS - 8
ER -