Method and device for capacitive touch panel based biosignal measurement

A device for capacitive touch panel based biosignal measurement, the device comprising:

• a memory storing instructions;
• one or more Input/Output (I/O) interfaces;
• and
• one or more hardware processors coupled to the memory via the one or more I/O interfaces, wherein the one or more hardware processors are configured by the instructions to:
• detect a touch event performed by a subject on a capacitive touch panel of a device comprising an array of sensors by scanning the array at regular predefined time intervals;
• obtain a raw signal corresponding to change in a raw capacitance value generated by the capacitive touch panel on detection of the touch event;
• derive a normalized signal by normalizing the raw signal using a normalization model, wherein the normalization model derives the normalized signal based on inter-relationships between a plurality of variables that affect the raw capacitance value, wherein the plurality of variables comprise a) a set of device specific parameters associated with the capacitive touch panel of the device, b) a set of ethnographic parameters, and c) metadata associated with the subject, wherein the set of device specific parameters comprises capacitive sensing elements density, dynamic range of change in the capacitance value with respect to the touch event, and delay in scanning capacitance array, which is dependent on processing power/speed of touch pad controller unit;
• wherein the set of ethnographic parameters comprises average size of fingers, normalized static potential, skin color and dermal thickness;
• wherein the metadata associated with the subject includes age and profession;
• and wherein the normalization model generated using a combination of statistical techniques and Machine Learning (ML), techniques, wherein the normalization model is generated by a central server and stored in an external repository to be accessed by the device, wherein the generation comprises:
• recording a set of changes in each raw capacitance value generated for a set of touch events performed by a set of sample subjects on a set of sample capacitive touch panels, representing the set of device specific parameters, along with a corresponding set of ethnographic parameters and metadata of the set of sample subjects, wherein the set of device specific parameters, the set of ethnographic parameters and the metadata associated with the subject represents the plurality of variables for generating the normalization model;
• simultaneously recording corresponding true measurements of a primary biosignal for the set of sample subjects;
• establishing a correlation among the true measurements of the primary biosignal for the set of sample subjects, the corresponding set of ethnographic parameters, the metadata of the set of sample subjects and the set of sample capacitive touch panels, wherein the correlation is established using the statistical techniques and the ML techniques;
• and
• deriving a set of normalization models from a plurality of normalization models for multiple capacitive touch panels and multiple configurations of multiple ethnographic parameters, for a varying set of metadata;
• extract from the normalized signal, a first set of parameters associated with the primary biosignal among a plurality of biosignals corresponding to the subject in accordance with a capacitance signal to a biosignal conversion model, wherein the first set of parameters comprise heart rate and pulse rate;
• derive a second set of parameters associated with one or more secondary biosignals among the plurality of biosignals from the first set of parameters associated with the primary biosignal, wherein the second set of parameters comprise heart rate variability;
• analyzing the first set of parameters associated with the primary biosignal and the second set of parameters associated with the secondary biosignals to generate an alert indicative of health condition of the subject by comparing the analyzed first set of parameters and the second set of parameters with corresponding subject specific standard values;
• and
• display analysis on the device indicating current health condition of the subject.

One or more non-transitory machine-readable information storage mediums comprising one or more instructions, which when executed by one or more hardware processors cause a method for:

• detecting a touch event performed by a subject on a capacitive touch panel of a device comprising an array of sensors by scanning the array at regular predefined time intervals;
• obtaining a raw signal corresponding to change in a raw capacitance value generated by the capacitive touch panel on detection of the touch event;
• deriving a normalized signal by normalizing the raw signal using a normalization model, wherein the normalization model derives the normalized signal based on inter-relationships between a plurality of variables that affect the raw capacitance value, wherein the plurality of variables comprise a) a set of device specific parameters associated with the capacitive touch panel of the device, b) a set of ethnographic parameters, and c) metadata associated with the subject, wherein the set of device specific parameters comprises capacitive sensing elements density, dynamic range of change in the capacitance value with respect to the touch event, and delay in scanning capacitance array, which is dependent on processing power/speed of touch pad controller unit;
• wherein the set of ethnographic parameters comprises average size of fingers, normalized static potential, skin color and dermal thickness;
• wherein the metadata associated with the subject includes age and profession;
• and wherein the normalization model generated using a combination of statistical techniques and Machine Learning;
• (ML), techniques, wherein the normalization model is generated by a central server and stored in an external repository to be accessed by the device, wherein the generation comprises:
• recording a set of changes in each raw capacitance value generated for a set of touch events performed by a set of sample subjects on a set of sample capacitive touch panels, representing the set of device specific parameters, along with a corresponding set of ethnographic parameters and metadata of the set of sample subjects, wherein the set of device specific parameters, the set of ethnographic parameters and the metadata associated with the subject represents the plurality of variables for generating the normalization model;
• simultaneously recording corresponding true measurements of a primary biosignal for the set of sample subjects;
• establishing a correlation among the true measurements of the primary biosignal for the set of sample subjects, the corresponding set of ethnographic parameters, the metadata of the set of sample subjects and the set of sample capacitive touch panels, wherein the correlation is established using the statistical techniques and the ML techniques;
• and
• deriving a set of normalization models from a plurality of normalization models for multiple capacitive touch panels and multiple configurations of multiple ethnographic parameters, for a varying set of metadata;
• extracting from the normalized signal, a first set of parameters associated with the primary biosignal among a plurality of biosignals corresponding to the subject in accordance with a capacitance signal to a biosignal conversion model, wherein the first set of parameters comprise heart rate and pulse rate;
• deriving a second set of parameters associated with one or more secondary biosignals among the plurality of biosignals from the first set of parameters associated with the primary biosignal, wherein the second set of parameters comprise heart rate variability;
• analyzing the first set of parameters associated with the primary biosignal and the second set of parameters associated with the secondary biosignals to generate an alert indicative of health condition of the subject by comparing the analyzed first set of parameters and the second set of parameters with corresponding subject specific standard values;
• and
• displaying the analysis on the device indicating current health condition of the subject.