System and method for obtaining health data using photoplethysmography

Abstract

A photoplethysmography (PPG) circuit or non-contact camera obtains PPG signals at a plurality of wavelengths. A signal processing module obtains at least a first spectral response around a first wavelength and a second spectral response around a second wavelength. The signal processing device generates PPG input data using the PPG signals, wherein the PPG input data includes one or more parameters obtained from each of the first spectral response and the second spectral response. A neural network processing device generates an input vector including the PPG input data and determines an output vector including health data, wherein the health data includes for example, an oxygen saturation level, a heart rate, or indication of a septic condition.

1 . A device, comprising:
 a plurality of light emitting diodes configured to emit light at a plurality of wavelengths onto tissue of a user, wherein at least a first wavelength is in a range of 380 nm to 410 nm; and   at least one signal processing device configured to detect a plurality of photoplethysmography (PPG) signals reflected from or transmitted through the tissue of the user at the plurality of wavelengths, wherein a first PPG signal is obtained from light reflected from or transmitted through the tissue of the user at the first wavelength; and   a neural network processing device implementing a machine learning algorithm, wherein the neural network processing device is configured to:
 receive input data, wherein the input data is obtained using at least the first PPG signal of the plurality of PPG signals; and 
 determine a risk of sepsis of the user from the input data. 
   
     
     
         2 . The device of  claim 1 , wherein the neural network processing device is configured to:
 receive the input data, wherein the input data includes at least the first PPG signal of the plurality of PPG signals and a second PPG signal of the plurality of PPG signals obtained from light reflected from or transmitted through the tissue of the user at a second wavelength in one of: a visible range or an infrared (IR) range.   
     
     
         3 . The device of  claim 1 , wherein the neural network processing device is configured to:
 use one or more parameters in the machine learning algorithm to determine the risk of sepsis, wherein the one or more parameters of the machine learning algorithm are determined using a training set that includes training input data obtained from a healthy population.   
     
     
         4 . The device of  claim 3 , wherein the training set that further includes training input data obtained from a population with a known septic condition. 
     
     
         5 . The device of  claim 4 , wherein the training input data is obtained using PPG signals from the healthy population and from the population with the know septic condition, wherein the PPG signals are obtained at least at the first wavelength in the range of 370 nm to 410 nm. 
     
     
         6 . The device of  claim 5 , wherein the neural network processing device is further configured to:
 periodically receive an updated learning vector, wherein the updated learning vector is generated from an updated training set, wherein the updated training set includes updated PPG input data obtained using updated PPG signals obtained at the first wavelength and at the second wavelength and corresponding known glucose concentration levels; and   reconfigure the one or more parameters of the machine learning algorithm using the updated learning vector.   
     
     
         7 . The device of  claim 1 , wherein the neural network processing device is configured to:
 determine a level of nitric oxide in blood flow using the input data.   
     
     
         8 . The device of  claim 2 , wherein the PPG input data includes:
 a value L λ1  generated using the first PPG signal, wherein the value L λ1  isolates the first PPG signal due to pulsating blood flow; and   a value L λ2  generated using the second PPG signal, wherein the value L λ2  isolates the PPG signal due to pulsating blood flow.   
     
     
         9 . The device of  claim 8 , wherein the PPG input data further includes:
 a value R λ1,λ2  obtained from a ratio including the value L λ1  and the value L λ2 .   
     
     
         10 . The device of  claim 2 , wherein the PPG input data includes:
 a first alternating current (AC) component signal I AC  generated using the first PPG signal; and   a second alternating current (AC) component signal I AC  generated using the second PPG signal.   
     
     
         11 . The device of  claim 10 , wherein the PPG input data further includes:
 a value R λ1,λ2  obtained from a ratio including the first AC component signal I AC  and the second AC component signal I AC .   
     
     
         12 . A device, comprising:
 one or more processing circuits configured to:
 receive a plurality of photoplethysmography (PPG) signals, wherein a first PPG signal is obtained from light at a first wavelength reflected from or transmitted through skin tissue of a patient and a second PPG signal is obtained from light at a second wavelength reflected from or transmitted through skin tissue of the patient, wherein the first wavelength is between 380 nm and 410 nm and wherein the second wavelength is equal to or greater than 660 nm; and 
 generate PPG input data using the first PPG signal and the second PPG signal; and 
   a neural network processing device configured to:
 receive the PPG input data; and 
 determine a risk of sepsis of the patient from the input data. 
   
     
     
         13 . The device of  claim 12 , wherein the one or more processing circuits is configured to:
 generate the PPG input data using the first PPG signal and the second PPG signal by generating characteristic features related to the shape of the PPG waveform from each of the first spectral response and the second spectral response.   
     
     
         14 . The device of  claim 13 , wherein the characteristic features related to the shape of the PPG waveform include one or more of: a pulse shape, an average distance between pulses, a variance, an instant energy information, or an energy variance. 
     
     
         15 . The device of  claim 12 , wherein the one or more processing circuits is configured to:
 generate the PPG input data using the first PPG signal and the second PPG signal by:
 determining a value L λ1  using the first PPG signal, wherein the value L λ1  isolates the first PPG signal due to pulsating blood flow; and 
 determining a value L λ2  using the second PPG signal, wherein the value L λ2  isolates the PPG signal due to pulsating blood flow. 
   
     
     
         16 . The device of  claim 15 , wherein the one or more processing circuits is configured to:
 generate the PPG input data using the first PPG signal and the second PPG signal by:   determining a value R λ1,λ2  from a ratio including the value L λ1  and the value L λ2 .   
     
     
         17 . A method in a system for determining a septic condition, comprising:
 obtaining by an optical circuit a plurality of photoplethysmography (PPG) signals, wherein a first PPG signal is obtained from light at a first wavelength reflected from or transmitted through skin tissue of a patient and a second PPG signal is obtained from light at a second wavelength reflected from or transmitted through skin tissue of the patient, wherein the first wavelength is between 380 nm and 410 nm and wherein the second wavelength is equal to or greater than 660 nm;   generating by at least a first processing circuit PPG input data using the first PPG signal and the second PPG signal; and   determining a septic condition in the patient from the input data using a machine learning algorithm operating in at least a second processing device.   
     
     
         18 . The method of  claim 17 , further comprising:
 determining by at least the first processing circuit a value L λ1  using the first PPG signal, wherein the value L λ1  isolates the first PPG signal due to pulsating blood flow; and   determining by at least the first processing circuit a value L λ2  using the second PPG signal, wherein the value L λ2  isolates the PPG signal due to pulsating blood flow.   
     
     
         19 . The method of  claim 18 , further comprising:
 determining by at least the first processing circuit a value R λ1,λ2  from a ratio including the value L λ1  and the value L λ2 .   
     
     
         20 . The method of  claim 17 , further comprising:
 configuring one or more parameters utilized by the machine learning algorithm by at least the second processing circuit to determine the risk of sepsis, wherein the one or more parameters of the machine learning algorithm are determined using a training set that includes training input data obtained from a healthy population and a population with a septic condition.