公开(公告)号：US12183465B2

公开(公告)日：2024-12-31

申请号：US18135851

申请日：2023-04-18

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bryan Conroy ,  Jonathan Rubin

IPC: G16H50/20 ,  G06F17/16 ,  G06N5/01 ,  G06N20/00

Abstract: A method of explaining a machine learning model, including: receiving a plurality of disease states for a patient over time from a database, wherein the disease states have a plurality of features; generating a plurality of locally faithful explanation models for the patient for the disease states based upon the machine learning model; calculating an explanation with respect to one feature of the plurality of features over time using the locally faithful explanation models; and calculating the importance of the one feature of the plurality of features over time based upon the plurality of locally faithful explanation models.

公开(公告)号：US20210098092A1

公开(公告)日：2021-04-01

申请号：US17032019

申请日：2020-09-25

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bishal Lamichhane ,  Mohammad Shahed Sorower

IPC: G16H10/60 ,  G06F21/62 ,  G06F16/14

Abstract: Implementations set forth herein relate to a peer-to-peer search system for patient medical records for leveraging benefits of identifying similar patient cases in a secured singular network. The peer-to-peer search system can use a distributed ledger to securely correlate similar instances of medical data, located in various other systems, to a globally accessible network that is available via the peer-to-peer search system. For instance, medical data from various sources can be hashed by a hash technique, such as locality-sensitive hashing, and stored in a hash database. When the hash database is queried via the peer-to-peer search system, hash data corresponding to query results can be provided and, optionally, ranked according to similarities between a hashing of input query to a hashing of documents embodying the query results.

公开(公告)号：US20200160201A1

公开(公告)日：2020-05-21

申请号：US16679842

申请日：2019-11-11

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bryan Conroy ,  Jack He ,  Jonathan Rubin

IPC: G06N7/00 ,  G06N3/08 ,  G06N3/04 ,  G06N20/20 ,  G06F17/18 ,  G06F17/16

Abstract: A method for training a probabilistic encoder-decoder having a latent space, the method including: extracting different types of medical data for a group of individuals; creating a data matrix X including the extracted medical data, wherein each row of the data matrix X includes data for one of the group of individuals; creating condition matrix C including features to define a clinical condition, wherein each row of the condition matrix C includes the condition data for one of the group of individuals; and training the encoder and the decoder to learn the latent space by minimizing the reconstruction loss and using a regularization effect to force clinically similar inputs to be close together in the latent space.

公开(公告)号：US11928610B2

公开(公告)日：2024-03-12

申请号：US16679842

申请日：2019-11-11

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bryan Conroy ,  Jack He ,  Jonathan Rubin

IPC: G06F17/16 ,  G06F17/18 ,  G06N3/045 ,  G06N3/08 ,  G06N7/01 ,  G06N20/20

CPC classification number: G06N7/01 ,  G06F17/16 ,  G06F17/18 ,  G06N3/045 ,  G06N3/08 ,  G06N20/20

Abstract: A method for training a probabilistic encoder-decoder having a latent space, the method including: extracting different types of medical data for a group of individuals; creating a data matrix X including the extracted medical data, wherein each row of the data matrix X includes data for one of the group of individuals; creating condition matrix C including features to define a clinical condition, wherein each row of the condition matrix C includes the condition data for one of the group of individuals; and training the encoder and the decoder to learn the latent space by minimizing the reconstruction loss and using a regularization effect to force clinically similar inputs to be close together in the latent space.

公开(公告)号：US11657920B2

公开(公告)日：2023-05-23

申请号：US16880250

申请日：2020-05-21

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bryan Conroy ,  Jonathan Rubin

IPC: G16H50/20 ,  G06F17/16 ,  G06N5/00 ,  G06N20/00

CPC classification number: G16H50/20 ,  G06F17/16 ,  G06N5/003 ,  G06N20/00

Abstract: A method of explaining a machine learning model, including: receiving a plurality of disease states for a patient over time from a database, wherein the disease states have a plurality of features; generating a plurality of locally faithful explanation models for the patient for each disease state based upon the machine learning model; calculating an explanation with respect to one feature of the plurality of features over time using the locally faithful explanation models; and calculating the importance of the one feature of the plurality of features over time based upon the plurality of locally faithful explanation models.

公开(公告)号：US20210012897A1

公开(公告)日：2021-01-14

申请号：US16880250

申请日：2020-05-21

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： Gajendra Jung Katuwal ,  Bryan Conroy ,  Jonathan Rubin

IPC: G16H50/20 ,  G06N20/00 ,  G06N5/00 ,  G06F17/16

Abstract: A method of explaining a machine learning model, including: receiving a plurality of disease states for a patient over time from a database, wherein the disease states have a plurality of features; generating a plurality of locally faithful explanation models for the patient for each disease state based upon the machine learning model; calculating an explanation with respect to one feature of the plurality of features over time using the locally faithful explanation models; and calculating the importance of the one feature of the plurality of features over time based upon the plurality of locally faithful explanation models .