TY - JOUR
T1 - Non-invasive Estimation of Clinical Severity of Anemia Using Hierarchical Ensemble Classifiers
AU - Chakraborty, Sushovan
AU - Kansara, Krity
AU - Dinesh Kumar, R.
AU - Swaminathan, Dhivya
AU - Aatre, Kiran
AU - Acharya, Soumyadipta
N1 - Funding Information:
We would like to acknowledge the valuable inputs and efforts of Dr. Sanjay Pattanshetty, Dr. Navya Vyas and Dr. Shilpa Drakshi from Prasanna School of Public Health, MAHE for facilitating the study. We would also like to show our gratitude towards Dr. Keerthinath Ballal, Chief Medical Officer from TMA Pai Rotary Hospital, Karkala; and Health officials and staff of Gangavathi Taluk Hospital and Koppal District Hospital for their assistance in data collection.
Publisher Copyright:
© 2022, Taiwanese Society of Biomedical Engineering.
PY - 2022/12
Y1 - 2022/12
N2 - Purpose: Current techniques of anemia classification are either invasive, expensive or inaccurate, making them ill-suited for community health-worker based screening programs. In this study, we propose an Artificial Intelligence (AI) based anemia classification method using a multi-wavelength non-invasive photometry device. Methods: A finger mounted photo-plethysmogram (PPG) device was designed to acquire PPG signals at four wavelengths (590, 660, 810, and 940 nm). A set of 13 attenuation and ratio-of-ratio features, derived using the peak and trough information extracted from the PPG signals, were used to develop a three-way hierarchical ensemble classification scheme using a machine-learning algorithm. PPG data from the device and true hemoglobin data from laboratory-based cell counters was collected for 1583 women of childbearing age and subjects were classified into either healthy (Hemoglobin, Hb > 11 g/dL), anemic (Hb: 7–11 g/dL) or severely anemic (Hb < 7 g/dL) categories. Results: We report a classification sensitivity of 92% (p < 0.05) and specificity of 84% (p < 0.05) in differentiating anemic and non-anemic women. We also report a sensitivity of 76% (p < 0.05), and specificity of 74% (p < 0.05) in identifying severe anemia. Conclusion: We believe that the proposed anemia classification algorithm, along with the associated sensor has the potential to be productized as a low-cost non-invasive anemia-screening device to rapidly determine next steps in clinical decision making in widespread community interventions.
AB - Purpose: Current techniques of anemia classification are either invasive, expensive or inaccurate, making them ill-suited for community health-worker based screening programs. In this study, we propose an Artificial Intelligence (AI) based anemia classification method using a multi-wavelength non-invasive photometry device. Methods: A finger mounted photo-plethysmogram (PPG) device was designed to acquire PPG signals at four wavelengths (590, 660, 810, and 940 nm). A set of 13 attenuation and ratio-of-ratio features, derived using the peak and trough information extracted from the PPG signals, were used to develop a three-way hierarchical ensemble classification scheme using a machine-learning algorithm. PPG data from the device and true hemoglobin data from laboratory-based cell counters was collected for 1583 women of childbearing age and subjects were classified into either healthy (Hemoglobin, Hb > 11 g/dL), anemic (Hb: 7–11 g/dL) or severely anemic (Hb < 7 g/dL) categories. Results: We report a classification sensitivity of 92% (p < 0.05) and specificity of 84% (p < 0.05) in differentiating anemic and non-anemic women. We also report a sensitivity of 76% (p < 0.05), and specificity of 74% (p < 0.05) in identifying severe anemia. Conclusion: We believe that the proposed anemia classification algorithm, along with the associated sensor has the potential to be productized as a low-cost non-invasive anemia-screening device to rapidly determine next steps in clinical decision making in widespread community interventions.
KW - Anemia
KW - Ensemble classifiers
KW - Hemoglobin
KW - Photoplethysmograph
UR - http://www.scopus.com/inward/record.url?scp=85140123655&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85140123655&partnerID=8YFLogxK
U2 - 10.1007/s40846-022-00750-3
DO - 10.1007/s40846-022-00750-3
M3 - Article
AN - SCOPUS:85140123655
SN - 1609-0985
VL - 42
SP - 828
EP - 838
JO - Journal of Medical and Biological Engineering
JF - Journal of Medical and Biological Engineering
IS - 6
ER -