Research presented here is motivated by the strong need of an affordable, easy to use sensing technology for the real-time noninvasive continuous measurement of hemoglobin concentration in arterial blood, with minimal interaction. Presently, the concentration of hemoglobin is measured using sodium lauryl sulphate (SLS) method, which is invasive and demands blood to be drawn from a vein using a syringe and chemically analyzed in a laboratory. The above method has the disadvantages like pain, causing stress, and infection risk to the patients. Hence a novel noninvasive method, using the magnetoplethysmogram(MPG) system is proposed in this thesis.



The magnetic flux required for obtaining an MPG is generated by a permanent magnet placed on the wrist surface such that the generated flux couples with the blood flow in the radial artery. As a first step, simulations were carried out using COMSOL Multiphysics Finite Element method, to optimize the MPG sensor head. Optimization technique will be discussed here. Initially the Hall Effect Sensor is used as the magnetic flux sensor for acquiring the MPG signal. As the MPG probe is placed at the pulse point in contact with the skin, it is realized that the signal is corrupted due to synchronous motion artefacts. Hence a noncontact MPG system is proposed here.



Using the noncontact MPG probe with the GMR sensor, clinical validation for the measurement of hemoglobin concentration is done for the 64 patients in the institute hospital of IIT Madras. From the analysis, it is found that the worst-case error in ascertaining the Hb concentration using the proposed MPG system is only 6.5 %, which is much less than that of the existing noninvasive hemoglobinometers reported so far.