Systems and methods for unobtrusive digital health assessment

A system comprising:

• one or more data storage devices operatively coupled to one or more hardware processors and configured to store instructions configured for execution by the one or more hardware processors to:
• obtain a plurality of values associated with a plurality of metadata features associated with a plurality of high risk subjects, wherein
• the plurality of metadata features comprises domain knowledge of the plurality of high risk subjects
• the domain knowledge includes one or more of:
• physical characteristics including height, weight, gender, ethnicity, and age,
• lifestyle habits including smoking, drinking, exercising regularly, and over-eating, and
• prevailing medical conditions,
• the high risk subjects are subjects with predisposing systemic conditions including neuropathy, peripheral arterial disease, diabetes mellitus, proneness to infection, autoimmune disease,
• the plurality of high risk subjects are more likely to get a particular disease than a normal subject,
• a platform integrates the physical characteristics, the lifestyle habits and the prevailing medical conditions including diabetes, hypertension, cardiac illness, anemia, with monitored physical activities and physiological measurements to assess health of the high risk subjects;
• generate groups from the plurality of high risk subjects based on a plurality of combinations of the obtained plurality of metadata features corresponding to one of a combination of the age and the weight associated with the plurality of high risk subjects, the age and the height associated with the plurality of high risk subjects, the age associated with the plurality of high risk subjects, the weight and having cardiac issues associated with the plurality of high risk subjects;
• and
• iteratively obtain a skewed normal distribution of the plurality of high risk subjects;
• automatically generate, based on the obtained skewed normal distribution, a treatment class and a control class for each of the generated groups, wherein the skewed normal distribution is obtained by:
• generating a fuzzy membership function using a neural network, wherein the neural network includes a multi-layer perceptron with at least two hidden layers and a fully connected input layer and an output layer;
• and
• deriving, based on the fuzzy membership function, two feature classes for each of the obtained plurality of metadata features, wherein a feature class of the two feature classes is young or old for the age as a metadata feature of the plurality of metadata features, and the class is tall or short for the height as the metadata feature;
• deriving a plurality of normalized values between 0 and 1 for each of the plurality of high risk subjects in the two derived feature classes, wherein a normalized value of the plurality of normalized values corresponds to a metadata feature;
• obtaining a Manhattan distance between every pair of high risk subjects amongst the plurality of high risk subjects using the plurality of normalized values;
• generating two clusters of the high risk subjects based on the obtained Manhattan distance and using a fuzzy C-means clustering method such that each of the two clusters have equivalent normalized values;
• randomly sampling the two clusters to create the control class and the treatment class such that the control class and the treatment class comprises an equivalent number of plurality of high risk subjects from each of the two clusters;
• reconstructing the plurality of normalized values associated with each of the plurality of high risk subjects by correlating an identifier of the plurality of high risk subjects with a meta-dictionary to obtain actual values corresponding to the associated metadata features for each of the plurality of high risk subjects in the control class and the treatment class;
• identifying a level of deterioration of health of each of the high risk subjects in the treatment class by:
• monitoring, via a wearable device and allied mobile or stationary gateways, physical activity levels related to a time spent and a distance achieved while performing walking measured and physiological measurements of the plurality of high risk subjects from the treatment class;
• classifying each of the high risk subjects into one of a plurality of pre-determined classes from I to IV illustrative of health assessment of the high risk subject and depict stages of a heart failure using a computational model and a correlation between the monitored physical activity levels (MET) and the physiological measurements;
• wherein the computational model is based on New York Heart Association (NYHA) guidelines wherein 4 classes are identified to depict stages of heart failure, wherein Class I is mapped to a condition that ordinary physical activity does not cause undue fatigue, palpitations, dyspnea and/or angina, Class II is mapped to a condition that ordinary physical activity does cause undue fatigue, palpitations, dyspnea and/or angina, Class III is mapped to a condition that less than ordinary physical activity causes undue fatigue, palpitations, dyspnea and/or angina, and Class IV is mapped to a condition that fatigue, palpitations, dyspnea and/or angina occurs at rest,
• wherein a metadata feature fatigue is modelled in form of I=f (MET, W, T, LPA, G, A, N), where I represents an intensity as a function of intensity associated with the MET, W represents body weight of the high risk subject under consideration, T represents Spell's duration, LPA represents level of physical activity in daily life of the high risk subject under consideration, G represents the gender of the high risk subject under consideration, A represents age spectrum and N represents a normalization constant;
• predicting a normalized value for each of the physiological measurements of interest using a Hidden Markov Model (HMM) with stage wise prediction, the physiological parameters being a metabolic equivalents of task (MET), breathing power change, heart rate change, breathing rate change, time taken to return to a Basal heart rate referring to normal heart rate at rest for the high risk subject under consideration and accordingly the Class II is identified if a high risk subject gets tired after predefined minutes of walking;
• computing a measure of deviation from the predicted normalized value using an actual normalized value obtained from the monitored physiological measurements to assess deviation from a healthy condition for each of the high risk subjects;
• eliminating local outliers in the physiological measurements using a Local Outlier Filter (LOF) algorithm to obtain filtered physiological measurements (208 e);
• performing a trend analyses, of the monitored physical activity levels and the filtered physiological measurements, using AutoRegressive Integrated Moving Average (ARIMA), wherein the ARIMA facilitates in identifying a long term trend when the subject is tired while performing the physical activity;
• and
• triggering an alarm when a trend is negative with a slope greater than a pre-defined threshold;
• obtaining feedback from the high risk subject pertaining to the monitored physical activity levels and analyzing the feedback to check if the high risk subject feel tired after 2 minutes of walking and the metadata feature associated with tiredness is identified as a bio-marker for further assessment of the high risk subject, wherein physiological parameters monitored before and after the physical activity for a particular metabolic equivalents (MET) are used to assess the levels of cardiopulmonary fatigue in the high risk subject,
• wherein the fatigue levels of the high risk subject under consideration is normalized considering other high risk subjects in the same group, wherein if a normalized level of the fatigue is higher than a major percentage of population in the group, then the high risk subject is marked for further assessment based on the fatigue level being identified as the bio-marker,
• wherein the marked high risk subject is monitored longitudinally over time to check if the fatigue levels are trending negatively to trigger a timely alarm, wherein if the fatigue levels are not trending negatively, then pre-op monitoring continues to assess the health of the high risk subject in an unobtrusive manner.

A computer program product comprising a non-transitory computer readable medium having a computer readable program embodied therein, wherein the computer readable program, when executed on a computing device, causes the computing device to:

• obtain a plurality of values associated with a plurality of metadata features associated with a plurality of high risk subjects, wherein
• the plurality of metadata features comprises domain knowledge of the plurality of high risk subjects
• the domain knowledge includes one or more of:
• physical characteristics including height, weight, gender, ethnicity, and age,
• lifestyle habits including smoking, drinking, exercising regularly, and over-eating, and prevailing medical conditions,
• the high risk subjects are subjects with predisposing systemic conditions including neuropathy, peripheral arterial disease, diabetes mellitus, proneness to infection, autoimmune disease,
• the plurality of high risk subjects are more likely to get a particular disease than a normal subject,
• a platform integrates the physical characteristics, the lifestyle habits and the prevailing medical conditions including diabetes, hypertension, cardiac illness, anemia, with monitored physical activities and physiological measurements to assess health of the high risk subjects;
• generate groups from of the plurality of high risk subjects based on a plurality of combinations of the obtained plurality of metadata features corresponding to one of a combination of the age and the weight associated with the plurality of high risk subjects, the age and the height associated with the plurality of high risk subjects, the age associated with the plurality of high risk subjects, the weight associated with the plurality of high risk subjects and cardiac issues associated with the plurality of high risk subjects;
• and
• iteratively obtain a skewed normal distribution of the plurality of high risk subjects;
• automatically generating, based on the obtained skewed normal distribution, a treatment class and a control class for each of the generated groups, wherein the skewed normal distribution is obtained by:
• generating a fuzzy membership function using a neural network, wherein the neural network includes a multi-layer perceptron with at least two hidden layers and a fully connected input layer and an output layer and
• deriving, based on the fuzzy membership function, two feature classes for each of the obtained metadata features, wherein feature class is young or old for the age as the metadata feature, and the class is tall or short for the height as the metadata feature, wherein the neural network uses a multi-layer perceptron with at least two hidden layers and a fully connected input layer and output layer;
• deriving a plurality of normalized values between 0 and 1 for each of the high risk subjects in the two derived feature classes, wherein a normalized value in the plurality of normalized values corresponds to a metadata feature;
• obtaining a Manhattan distance between every pair of subjects amongst the high risk subjects using the plurality of normalized values;
• generating two clusters of the high risk subjects based on the obtained Manhattan distance and using a fuzzy C-means clustering method such that each of the two clusters have equivalent normalized values, wherein the fuzzy C-means clustering ensures that an inter-cluster distance is high and an intra-cluster distance is low;
• randomly sampling the two clusters to create the control class and the treatment class such that the control class and the treatment class comprises an equivalent number of high risk subjects from each of the two clusters;
• reconstructing the plurality of normalized values associated with each of the high risk subjects by correlating an identifier of the high risk subject with a meta-dictionary to obtain actual values corresponding to the associated metadata features for each of the high risk subjects in the control class and the treatment class;
• identifying a level of deterioration of health of each of the high risk subjects in the treatment class by:
• monitoring, via a wearable device and allied mobile or stationary gateways, physical activity levels related to a time spent and a distance achieved while performing walking measured and physiological measurements of the plurality of high risk high risk subjects from the treatment class;
• classifying each of the high risk subjects into one of a plurality of pre-determined classes from I to IV illustrative of health assessment of the high risk subject and depict stages of a heart failure using a computational model and a correlation between the monitored physical activity levels (MET) and the physiological measurements, wherein the computational model is based on New York Heart Association (NYHA) guidelines wherein 4 classes are identified to depict stages of heart failure, wherein Class I is mapped to a condition that ordinary physical activity does not cause undue fatigue, palpitations, dyspnea and/or angina, Class II is mapped to a condition that ordinary physical activity does cause undue fatigue, palpitations, dyspnea and/or angina, Class III is mapped to a condition that less than ordinary physical activity causes undue fatigue, palpitations, dyspnea and/or angina, and Class IV is mapped to a condition that fatigue, palpitations, dyspnea and/or angina occurs at rest,
• wherein a metadata feature fatigue is modelled in form of I=f (MET, W, T, LPA, G, A, N), where I represents an intensity as a function of intensity associated with the MET, W represents body weight of the high risk subject under consideration, T represents Spell's duration, LPA represents level of physical activity in daily life of the high risk subject under consideration, G represents the gender of the high risk subject under consideration, A represents age spectrum and N represents a normalization constant;
• predicting a normalized value for each of the physiological measurements of interest using a Hidden Markov Model (HMM) with stage wise prediction, the physiological parameters being a metabolic equivalents of task (MET), breathing power change, heart rate change, breathing rate change, time taken to return to a Basal heart rate referring to normal heart rate at rest for the high risk subject under consideration and accordingly the Class II is identified if a high risk subject gets tired after predefined minutes of walking;
• computing a measure of deviation from the predicted normalized value using an actual normalized value obtained from the monitored physiological measurements to assess deviation from a healthy condition for each of the high risk subjects;
• eliminating local outliers in the physiological measurements using a Local Outlier Filter (LOF) algorithm to obtain filtered physiological measurements (208 e);
• performing a trend analyses, of the monitored physical activity levels and the filtered physiological measurements, using AutoRegressive Integrated Moving Average (ARIMA), wherein the ARIMA facilitates in identifying a long term trend when the subject is tired while performing the physical activity;
• and
• triggering an alarm when a trend is negative with a slope greater than a pre-defined threshold;
• obtaining feedback from the high risk subject pertaining to the monitored physical activity levels and analyzing the feedback to check if the high risk subject feel tired after 2 minutes of walking and the metadata feature associated with tiredness is identified as a bio-marker for further assessment of the high risk subject, wherein physiological parameters monitored before and after the physical activity for a particular metabolic equivalents (MET) are used to assess the levels of cardiopulmonary fatigue in the high risk subject, wherein the fatigue levels of the high risk subject under consideration is normalized considering other high risk subjects in the same group, wherein if a normalized level of the fatigue is higher than a major percentage of population in the group, then the high risk subject is marked for further assessment based on the fatigue level being identified as the bio-marker,
• wherein the marked high risk subject is monitored longitudinally over time to check if the fatigue levels are trending negatively to trigger a timely alarm, wherein if the fatigue levels are not trending negatively, then pre-op monitoring continues to assess the health of the high risk subject in an unobtrusive manner.