Abstract: System and computer-implemented method for detecting and categorizing pathologies through an analysis of pulsatile blood flow. The method has a pulsatile blood flow signal of a subject, extracting a set of features from the pulsatile blood flow signal; and categorizing a pathology based on the extracted features. The extracted features may be predetermined features or features learned through a machine learning algorithm. For the categorization, a classification or a regression algorithm may be used to provide an index or a value score as a biomarker. Additional static features of the subject may be used in the categorization.
Type:
Grant
Filed:
September 11, 2019
Date of Patent:
July 18, 2023
Assignees:
FUNDACIO INSTITUT DE CIENCIES FOTONIQUES, INSTITUCIO CATALANA DE RECERCA I ESTUDIS AVANCATS, HEMOPHOTONICS, S.L.
Inventors:
Turgut Durduran, Jonas Fischer, Ameer Ghouse, Udo Weigel