# US 12,008,313 B2 · Medical data verification method and electronic device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,008,313 B2
- Document ID: US12008313B2
- Status: Granted
- Kind: B2
- Application number: 17/448,065
- Filed: 2021-09-20
- Published: 2022-01-06
- Granted: 2024-06-11
- Earliest claimed: 2020-09-29
- Assignees: Baidu International Technology (Shenzhen) Co., Ltd.
- Applicants: Baidu International Technology (Shenzhen) Co., Ltd.
- Inventors: Zhou Fang; Yabing Shi; Ye Jiang; Chunguang Chai
- Web page: https://patentgrep.com/patent/US12008313B2

## CPC

- G06F 40/194: Calculation of difference between files
- G06F 16/3347: {using vector based model}
- G06F 16/367: {Ontology}
- G06F 16/93: Document management systems
- G06F 40/205: Parsing
- G06F 40/30: Semantic analysis
- G06N 3/045: Combinations of networks
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 3/09: Supervised learning
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G16H 70/60: relating to pathologies
- G06N 3/08: Learning methods
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G16H 10/60: for patient-specific data, e.g. for electronic patient records
- G16H 40/20: for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

The present disclosure provides a medical data verification method, apparatus and an electronic device, related to a field of artificial intelligence technologies, such as AI (artificial intelligence) medical treatment, deep learning, knowledge graphs, natural language processing. A specific implementation is: obtaining medical data to be verified and a candidate document; obtaining feature vectors respectively corresponding to the medical data to be verified and the candidate document by processing the medical data to be verified and the candidate document by using a nature language processing model; obtaining N correlation vectors by calculating correlation between the medical data to be verified and the candidate document based on the feature vectors by using N methods, N being a positive integer greater than 1; and determining a confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the N correlation vectors.

## Claims

### Claim 1

A medical data verification method, comprising:
obtaining medical data to be verified and a candidate document, wherein the medical data to be verified comprises an entity, an attribute, and an attribute value;
obtaining a first feature vector corresponding to the entity, a second feature vector corresponding to the attribute, a third feature vector corresponding to the attribute value, and a fourth feature vector corresponding to the candidate document, by processing the medical data to be verified and the candidate document by using a natural language processing model;
obtaining N first-type correlation vectors by calculating correlation between the first feature vector and the fourth feature vector by respectively using N methods, wherein N is a positive integer greater than 1;
obtaining N second-type correlation vectors by calculating correlation between the second feature vector and the fourth feature vector by respectively using the N methods;
obtaining N third-type correlation vectors by calculating correlation between the third feature vector and the fourth feature vector by respectively using the N methods; and
determining a confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors.

### Claim 2 (depends on 1)

The method according to claim 1, wherein, obtaining the feature vectors respectively corresponding to the medical data to be verified and the candidate document further comprises:
obtaining at least one of: a fifth feature vector corresponding to the entity and the attribute, a sixth feature vector corresponding to the attribute and the attribute value, and a seventh feature vector corresponding to the entity and the attribute value.

### Claim 3 (depends on 1)

The method according to claim 1, wherein determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors comprises:
obtaining a first fusion vector by fusing the N first-type correlation vectors;
obtaining a second fusion vector by fusing the N second-type correlation vectors;
obtaining a third fusion vector by fusing the N third-type correlation vectors; and
determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the first fusion vector, the second fusion vector and the third fusion vector.

### Claim 4 (depends on 1)

The method according to claim 1, wherein determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors comprises:
fusing the N first-type correlation vectors, the N second-type correlation vectors and the N third-type correlation vectors to generate a fourth fusion vector; and
determining the confidence degree of the medical data to be verified to the candidate document by performing calculation on the fourth fusion vector.

### Claim 5 (depends on 1)

The method according to claim 1, wherein performing the fusion calculation on the correlation vectors comprises:
determining a weight value corresponding to each of the N methods; and
performing weighted fusion on the correlation vectors based on the weight value corresponding to each of the N methods.

### Claim 6 (depends on 1)

The method according to claim 1, wherein obtaining the medical data to be verified and the candidate document comprises:
obtaining the medical data to be verified and a description text corresponding to the medical data to be verified; and
obtaining a candidate document associated with the description text from a database based on a correlation between the description text and each document in the database.

### Claim 7

An electronic device, comprising:
at least one processor; and
a memory communicatively couple to the at least one processor;
wherein, the memory has instructions executable by the at least one processor stored thereon that, when executed by the at least one processor, cause the at least one processor to perform a medical data verification method, comprising:
obtaining medical data to be verified and a candidate document, wherein the medical data to be verified comprises an entity, an attribute, and an attribute value;
obtaining a first feature vector corresponding to the entity, a second feature vector corresponding to the attribute, a third feature vector corresponding to the attribute value, and a fourth feature vector corresponding to the candidate document, by processing the medical data to be verified and the candidate document by using a natural language processing model;
obtaining N first-type correlation vectors by calculating correlation between the first feature vector and the fourth feature vector by respectively using N methods, wherein N is a positive integer greater than 1;
obtaining N second-type correlation vectors by calculating correlation between the second feature vector and the fourth feature vector by respectively using the N methods;
obtaining N third-type correlation vectors by calculating correlation between the third feature vector and the fourth feature vector by respectively using the N methods; and
determining a confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors.

### Claim 8 (depends on 7)

The electronic device according to claim 7, wherein, obtaining the feature vectors respectively corresponding to the medical data to be verified and the candidate document further comprises:
obtaining at least one of: a fifth feature vector corresponding to the entity and the attribute, a sixth feature vector corresponding to the attribute and the attribute value, and a seventh feature vector corresponding to the entity and the attribute value.

### Claim 9 (depends on 7)

The electronic device according to claim 7, wherein determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors comprises:
obtaining a first fusion vector by fusing the N first-type correlation vectors;
obtaining a second fusion vector by fusing the N second-type correlation vectors;
obtaining a third fusion vector by fusing the N third-type correlation vectors; and
determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the first fusion vector, the second fusion vector and the third fusion vector.

### Claim 10 (depends on 7)

The electronic device according to claim 7, wherein determining the confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors comprises:
fusing the N first-type correlation vectors, the N second-type correlation vectors and the N third-type correlation vectors to generate a fourth fusion vector; and
determining the confidence degree of the medical data to be verified to the candidate document by performing calculation on the fourth fusion vector.

### Claim 11 (depends on 7)

The electronic device according to claim 7, wherein performing the fusion calculation on the correlation vectors comprises:
determining a weight value corresponding to each of the N methods; and
performing weighted fusion on the correlation vectors based on the weight value corresponding to each of the N methods.

### Claim 12 (depends on 7)

The electronic device according to claim 7, wherein obtaining the medical data to be verified and the candidate document comprises:
obtaining the medical data to be verified and a description text corresponding to the medical data to be verified; and
obtaining a candidate document associated with the description text from a database based on a correlation between the description text and each document in the database.

### Claim 13

A non-transitory computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are configured to cause a computer to perform a medical data verification method, comprising:
obtaining medical data to be verified and a candidate document, wherein the medical data to be verified comprises an entity, an attribute, and an attribute value;
obtaining a first feature vector corresponding to the entity, a second feature vector corresponding to the attribute, a third feature vector corresponding to the attribute value, and a fourth feature vector corresponding to the candidate document, by processing the medical data to be verified and the candidate document by using a natural language processing model;
obtaining N first-type correlation vectors by calculating correlation between the first feature vector and the fourth feature vector by respectively using N methods, wherein N is a positive integer greater than 1;
obtaining N second-type correlation vectors by calculating correlation between the second feature vector and the fourth feature vector by respectively using the N methods;
obtaining N third-type correlation vectors by calculating correlation between the third feature vector and the fourth feature vector by respectively using the N methods; and
determining a confidence degree of the medical data to be verified to the candidate document by performing fusion calculation on the correlation vectors.

### Claim 14 (depends on 13)

The storage medium according to claim 13, wherein, obtaining the feature vectors respectively corresponding to the medical data to be verified and the candidate document further comprises:
obtaining at least one of: a fifth feature vector corresponding to the entity and the attribute, a sixth feature vector corresponding to the attribute and the attribute value, and a seventh feature vector corresponding to the entity and the attribute value.