METHOD FOR ANALYZING OUTPUT OF NEURAL NETWORK, AND SYSTEM THEREFOR

    公开(公告)号:US20240055104A1

    公开(公告)日:2024-02-15

    申请号:US18271231

    申请日:2021-03-29

    Applicant: DEEP BIO INC.

    CPC classification number: G16H30/40 G06N3/047 G06N3/08 G16H50/70

    Abstract: A method for analyzing an output of a neural network that analyzes an output result of a neural network trained so as to output a disease expression probability for each biological image pixel includes, depending on whether an output result value of the neural network for each pixel is equal to or greater than a reference value, an output analysis system determines the optimal output result value which is a reference value for detecting whether a disease is expressed in the corresponding pixel; determining an optimal cut-off value for determining whether a detected lesion site is effective with respect to a detected lesion in a biological image; and, when the output analysis system receives an output result corresponding to a diagnostic biometric image to be diagnosed, performing an output analysis on the output result by using the optimal reference value and the optimal cut-off value.

    DATA PROCESSING METHOD AND SYSTEM USING AUTOTHRESHOLDING

    公开(公告)号:US20220277812A1

    公开(公告)日:2022-09-01

    申请号:US17626795

    申请日:2020-07-10

    Abstract: A method for automatically calculating a threshold for classifying clusters from a reference data set and processing data by using same, and a system for performing same is included herein. A data processing method using auto-thresholding includes the steps of receiving, by a data processing system, as an input, a plurality of individual numerical values included in a reference data set having two or more clusters; on the basis of each of the numerical values included in the reference data set received as an input, calculating, by the data processing system, a threshold for classifying a cluster of the reference data set; and classifying, by the data processing system, into different clusters by using the threshold, each of at least one data set to be analyzed, having a plurality of individual numerical values.

Patent Agency Ranking