公开(公告)号：US12198454B2

公开(公告)日：2025-01-14

申请号：US17513473

申请日：2021-10-28

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Xue Chen ,  Jinnv Liu ,  Yude Li ,  Xianqiang Ni ,  Shuobin Liang ,  Li Zhou

IPC: G06V20/64 ,  A61B5/30 ,  A61B5/308 ,  G06F18/2137 ,  G06F18/214 ,  G06F18/241 ,  G06V10/147 ,  G06V10/42 ,  G06V10/94 ,  G16H10/60 ,  G16H50/20

Abstract: An attention mechanism-based 12-lead electrocardiogram (ECG) classification method is described, the method including acquiring an original image of a 12-lead ECG, segmenting waveform data recorded in the original image to obtain segmented waveform data for each lead in the 12-lead ECG, performing depth feature extraction on the segmented waveform data of said each lead to obtain a first feature map of said each lead, performing feature transformation on the first feature map of said each lead based on an attention mechanism to obtain a depth feature of said each lead, and classifying the depth feature of said each lead to obtain a classification result for the original image. The classification method can make full use of the 12-lead ECG for overall classification and improve the accuracy of image classification.

公开(公告)号：US12193826B2

公开(公告)日：2025-01-14

申请号：US17605278

申请日：2020-09-28

Applicant: BOE TECHNOLOGY GROUP CO., LTD. ,  Peking University

Inventor： Li Jiang ,  Yongyue Sun ,  Siyang Liang ,  Meng Wu ,  Yude Li ,  Hongyan Li ,  Litong Han

IPC: G06F16/23 ,  A61B5/35 ,  A61B5/364 ,  A61B5/366

Abstract: A method for processing time series data comprises: dividing time series data into a plurality of data fragments according to an objective function, the plurality of data fragments having a greatest similarity; in response to determining that at least one data fragment does not satisfy an iteration termination condition, performing following iteration operations on the at least one data fragment; and constructing a time series base pattern library by using a plurality of time series base patterns. The iteration operations includes: using the at least one data fragment as at least one update time series fragment; dividing each update time series fragment into a plurality of update data fragments; using each update data fragment that does not satisfy the iteration termination condition as a new update time series fragment; and using each update data fragment that satisfies the iteration termination condition as a time series base pattern.

公开(公告)号：US20240170161A1

公开(公告)日：2024-05-23

申请号：US18282018

申请日：2023-01-04

Applicant: BOE Technology Group Co., Ltd.

Inventor： Xue Chen ,  Jinnv Liu ,  Yude Li ,  Xiying Zhang ,  Zhenzhong Zhang ,  Li Zhou

IPC: G16H50/70 ,  G06F18/25 ,  G06F40/30 ,  G16H50/20 ,  G16H50/30

CPC classification number: G16H50/70 ,  G06F18/25 ,  G06F40/30 ,  G16H50/20 ,  G16H50/30

Abstract: A method for processing medical data, an apparatus and a storage medium, the method includes: acquiring a case-history datum, and performing a target process to obtain a disease-analysis vector, wherein the target process includes: generating a case-history semantic vector of the case-history datum; for each of preset diseases in a preset-disease set, determining a first possibility weight of the case-history datum caused by the preset disease according to the case-history semantic vector, to obtain a first weight vector; according to case-history symptoms and case-history diseases in the case-history datum, determining from a predetermined knowledge graph a candidate disease that is capable of generating generate the case-history datum; determining a second possibility weight of the case-history datum caused by the candidate disease, to obtain a second weight vector; and fusing the first weight vector and the second weight vector, to obtain the disease-analysis vector corresponding to the case-history datum.