US 11,786,151 B2
Apparatus and method for obtaining target signal spectrum
Ka Ram Choi, Hwaseong-si (KR); So Young Lee, Daejeon (KR); Jun Ho Lee, Incheon (KR); Sang Kyu Kim, Yongin-si (KR); and Hyeong Seok Jang, Seoul (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Dec. 11, 2020, as Appl. No. 17/119,634.
Claims priority of application No. 10-2020-0106001 (KR), filed on Aug. 24, 2020.
Prior Publication US 2022/0054050 A1, Feb. 24, 2022
Int. Cl. A61B 5/1455 (2006.01); A61B 5/145 (2006.01); A61B 5/00 (2006.01); G16H 40/60 (2018.01)
CPC A61B 5/1455 (2013.01) [A61B 5/14532 (2013.01); A61B 5/14546 (2013.01); A61B 5/4866 (2013.01); A61B 5/7203 (2013.01); A61B 5/7257 (2013.01); G16H 40/60 (2018.01); A61B 2560/0431 (2013.01); A61B 2562/0233 (2013.01)] 17 Claims
OG exemplary drawing
 
1. An apparatus for non-invasively estimating bio-information based on a target signal spectrum, the apparatus comprising:
a spectrum measurer configured to measure a plurality of spectra from an object; and
a processor configured to obtain a difference spectrum matrix by subtracting a reference spectrum from each of the plurality of spectra, to obtain a spectrum of a target signal based on the obtained difference spectrum matrix, and to estimate bio-information of the object based on the spectrum of the target signal by using a bio-information estimation model,
wherein the processor is further configured to obtain a symmetric matrix by calculating an inner product between the difference spectrum matrix and a transposed matrix of the difference spectrum matrix, select one row or one column from the obtained symmetric and obtain the spectrum of the target signal based on the selected one rover or one column,
wherein the target signal comprises at least one body substance and at least one noise component, and
wherein the processor is configured to, based on a spectrum of the at least one noise component being obtained, update the bio-information estimation model by reflecting the spectrum of the at least one noise component in the bio-information estimation model.