A Novel Noninvasive Hemoglobin Sensing Device for Anemia Screening

• Published in: IEEE sensors journal Vol. 21; no. 13; pp. 15318 - 15329
• Main Authors: Kumar, R. Dinesh, Guruprasad, S., Kansara, Krity, Rao, K. N. Raghavendra, Mohan, Murali, Reddy, Manjunath Ramakrishna, Prabhu, Uday Haleangadi, Prakash, P., Chakraborty, Sushovan, Das, Sreetama, Madhusoodanan, K. N.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.07.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Anemia; Blood; Computer architecture; Fiber optics; Fingers; Hemoglobin; Incident light; Laboratories; Light emitting diodes; Light sources; Luminous intensity; noninvasive; photo plethysmography; Photodetectors; Physicians; point light source; Probes; Screening; Sensors; Spectrophotometry
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2021.3070971

Anemia is a condition that affects more than a billion people globally, with the highest prevalence noticed among women and children. In large parts of the world anemia is detected by using invasive laboratory techniques that require blood draw by trained doctors, nurses or laboratory staff. A low cost and easy to use noninvasive device that can be used at the doorstep or bedside of a patient has the potential to create a significant impact in anemia screening. In this paper, we present the architecture and development of such a device. Using a specific arrangement of light emitting diodes and fiber optics, we have created a novel multiwavelength spectrophotometry sensing platform to detect anemia. A mechanical lever operated fiber optics based sensor eliminates the problems associated with conventional noninvasive finger probe, aiding in the detection of hemoglobin levels as low as 1.6 g/dL. This study also proposes a method wherein incident light intensity can be measured along with attenuated light intensity to support hemoglobin estimation. The device could detect the light signals as low as 0.001% of the incident light intensity. Performance of the proposed platform was validated in conjunction with an algorithm to estimate hemoglobin. The device was able to estimate hemoglobin with an RMSE of 1.47 ± 0.042 g/Dl and a correlation of 0.79 ± 0.03 between the predicted and actual value of Hb.