Title:
INTELLIGENT MULTI-DATA FUSION DISEASE-IDENTIFYING METHOD
Document Type and Number:
WIPO Patent Application WO/2022/126799
Kind Code:
A1
Abstract:
An intelligent multi-data fusion disease-identifying method, comprising: using an artificial intelligence theory and intelligent recognition and intelligent analysis technologies to determine, according to an improved neural network method, a ratio to analyze a recognition degree; determining an input item as a hard input item or a soft input item according to a recognition degree corresponding to a feature; dynamically inputting input items to a neural network model; and adaptively adjusting a hard determination rate and a soft determination rate by means of an intermediate layer calculation, to finally determine a disease. By using the intelligent multi-data fusion disease-identifying method, the problems in the existing technology data recognition difficulty, data recognition and determining inaccuracy, and the like are solved.
More Like This:
| WO/2021/247253 | PRECIPITATION REMOVAL FROM VIDEO |
| WO/2020/010561 | METHOD AND APPARATUS FOR MEASURING OBJECT PARAMETERS |
| JP2643960 | [Title of Invention] Character recognition method |
Inventors:
TAN SICONG (CN)
YU HAO (CN)
YU MENGFEI (CN)
YU HAO (CN)
YU MENGFEI (CN)
Application Number:
PCT/CN2021/000233
Publication Date:
June 23, 2022
Filing Date:
December 08, 2021
Export Citation:
Assignee:
TAN SICONG (CN)
YU HAO (CN)
YU MENGFEI (CN)
YU HAO (CN)
YU MENGFEI (CN)
International Classes:
G06K9/62
Foreign References:
| CN112509694A | 2021-03-16 | |||
| CN111081379A | 2020-04-28 | |||
| CN109543047A | 2019-03-29 | |||
| CN111861975A | 2020-10-30 | |||
| US20170091528A1 | 2017-03-30 |
Download PDF: