公开(公告)号：US11642087B2

公开(公告)日：2023-05-09

申请号：US16749514

申请日：2020-01-22

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Ibrahim Abdul Majeed ,  Sujit Jos ,  Rahul Arora ,  Ka Ram Choi ,  Sang Kon Bae

IPC: A61B5/00 ,  A61B5/024

CPC classification number: A61B5/7214 ,  A61B5/02416 ,  A61B5/725 ,  A61B5/7278

Abstract: A method and apparatus for pre-processing a photoplethysmogram (PPG) signal include receiving, by an apparatus, the PPG signal including a plurality of pulses, obtaining, by the apparatus, a smoothed PPG signal based on smoothing the PPG signal using a moving average filter, obtaining, by the apparatus, a drift removed PPG signal based on removing drift from the smoothed PPG signal using a fitted polynomial function, and obtaining, by the apparatus, a rectified PPG signal based on correcting a motion artifact of the drift removed PPG signal by replacing each of the plurality of pulses using a combination of a global signal pulse and a corresponding local signal pulse.

公开(公告)号：US11607178B2

公开(公告)日：2023-03-21

申请号：US16828364

申请日：2020-03-24

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Rahul Arora ,  Sujit Jos ,  Ibrahim Abdul Majeed ,  Ka Ram Choi ,  Sang Kon Bae

IPC: A61B5/00 ,  A61B5/024 ,  A61B5/021 ,  A61B5/145

Abstract: The present disclosure relates to a method and system for processing a photoplethysmogram (PPG) signal to improve accuracy of measurement of physiological parameters of a subject. The method may include removing a baseline drift from the PPG signal; obtaining a drift removed signal; filtering the drift removed signal; obtaining a filtered signal; performing motion artifact correction on the filtered signal; and obtaining a corrected signal.

公开(公告)号：US11796465B2

公开(公告)日：2023-10-24

申请号：US17163857

申请日：2021-02-01

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Sujit Jos ,  Kiran Bynam ,  Ibrahim A ,  Rahul Arora ,  Gorish Aggarwal ,  So Young Lee

IPC: G01N21/3577 ,  G01N21/359 ,  G01N33/49 ,  G06N5/04 ,  G06N20/00

CPC classification number: G01N21/3577 ,  G01N21/359 ,  G01N33/49 ,  G06N5/04 ,  G06N20/00 ,  G01N2201/12

Abstract: A method of predicting a blood compound concentration of a target may include receiving, by a system, spectral data associated with a region of the target, using near-infrared (NIR) spectroscopy. The method may include classifying, by the system, each of the plurality of data instances of the spectral data to one of a plurality of labelled classes. The method may include obtaining, by the system, one or more best fit models from a plurality of prediction models based on the classification. The method may include determining, by the system, blood compound concentration values corresponding to each of the one or more best fit models. The method may include predicting, by the system, the blood compound concentration of the target using the blood compound concentration values predicted using the best fit models.

公开(公告)号：US11089981B2

公开(公告)日：2021-08-17

申请号：US16459785

申请日：2019-07-02

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Gorish Aggarwal ,  Kiran Bynam ,  So Young Lee ,  Ibrahim A ,  Sujit Jos ,  Rahul Arora

IPC: A61B5/1455 ,  A61B5/145 ,  A61B5/00 ,  G06N20/20 ,  A61B5/1495

Abstract: A method includes processing near infrared spectroscopy (NIR) spectra and pure spectra, to obtain a preprocessed NIR spectra, wherein the preprocessed NIR spectra includes any one or any combination of a training data of a plurality of subjects, a calibration data of a test subject and a validation data of the test subject, extracting a dominant feature set from the preprocessed NIR spectra, wherein the dominant feature set includes at least one preprocessed NIR spectrum corresponding to each of the training data, the calibration data and the validation data, and determining a blood glucose concentration of the test subject, using the training data of the plurality of subjects and based on the extracted dominant feature set.

公开(公告)号：US20200022627A1

公开(公告)日：2020-01-23

申请号：US16459785

申请日：2019-07-02

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Gorish Aggarwal ,  Kiran Bynam ,  So Young Lee ,  Ibrahim A ,  Sujit Jos ,  Rahul Arora

IPC: A61B5/145 ,  A61B5/00 ,  G06N20/20 ,  A61B5/1495

Abstract: A method includes processing near infrared spectroscopy (NIR) spectra and pure spectra, to obtain a preprocessed NIR spectra, wherein the preprocessed NIR spectra includes any one or any combination of a training data of a plurality of subjects, a calibration data of a test subject and a validation data of the test subject, extracting a dominant feature set from the preprocessed NIR spectra, wherein the dominant feature set includes at least one preprocessed NIR spectrum corresponding to each of the training data, the calibration data and the validation data, and determining a blood glucose concentration of the test subject, using the training data of the plurality of subjects and based on the extracted dominant feature set.