# US 2024/0194356 A1 · Medical document analysis method and system thereof

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

- Number: US 2024/0194356 A1
- Document ID: US20240194356A1
- Status: Published application
- Kind: A1
- Application number: 18/319,842
- Filed: 2023-05-18
- Published: 2024-06-13
- Earliest claimed: 2022-12-09
- Assignees: ASG Inspiration Laboratory Ltd.
- Applicants: ASG Inspiration Laboratory Ltd.
- Inventors: Johnson LEE; Jao Juen HUNG; Sung Tsai YU; Shih Pan CHAO; Hao-Wei HUANG
- Web page: https://patentgrep.com/patent/US20240194356A1

## CPC

- G16H 70/00: ICT specially adapted for the handling or processing of medical references
- G06F 16/3334: {Selection or weighting of terms from queries, including natural language queries}
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G16H 70/40: relating to drugs, e.g. their side effects or intended usage
- G16H 10/60: for patient-specific data, e.g. for electronic patient records
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A medical document analysis method includes following steps: performing recursive search based on keywords extracted from first medical documents to obtain second medical documents; analyzing feature labels of the second medical documents; projecting the second medical documents onto a multi-dimensional map according to the feature labels; estimating second symptoms from a first symptom; selecting third medical documents based on the first symptom and the second symptom from the multi-dimensional map; analyzing correlation between the third medical documents and their respective feature labels in the multi-dimensional map to form a label topology map; and selecting a target branch path from branch paths in the label topology map, and displaying information about the target branch path.

## Claims

### Claim 1

A medical document analysis method, comprising:
performing a recursive search based on keywords of a plurality of first medical documents to obtain a plurality of second medical documents;
analyzing contents of the plurality of second medical documents, and marking each of the plurality of second medical documents with a plurality of feature labels;
projecting the plurality of second medical documents onto different coordinate positions in a multi-dimensional map respectively based on at least one of the plurality of feature labels of each of the plurality of second medical documents, wherein dimensions of the multi-dimensional map correspond to the plurality of feature labels, and distances between the coordinate positions correspond to a correlation between the plurality of second medical documents
in response to selecting a first symptom, estimating a plurality of second symptoms related to the first symptom;
selecting a plurality of coordinates matching the first symptom and the plurality of second symptoms from the multi-dimensional map, and selecting a plurality of third medical documents from the plurality of second medical documents based on a node adjacent to the plurality of coordinates;
analyzing a correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents in the multi-dimensional map to form a label topology map, wherein a top layer node in the label topology map is the first symptom, the label topology map comprises a plurality of branch paths, and each of the plurality of branch paths starts from the first symptom and passes through a drug formulation, a receptor name and a drug dosage form; and
selecting a target branch path from the plurality of branch paths, and displaying information related to the target branch path.

### Claim 2 (depends on 1)

The medical document analysis method of claim 1, wherein the step of performing the recursive search based on the keywords of the plurality of first medical documents comprises:
(a1) selecting a plurality of keywords from the plurality of first medical documents by using a natural language model;
(a2) searching for a plurality of related medical documents based on the plurality of keywords;
(a3) determining whether the plurality of related medical documents meet a convergence condition;
(a4) in response to the convergence condition not being met, updating the plurality of keywords by extracting the plurality of related medical documents by using the natural language model, and back to the step (a3); and
(a5) in response to the convergence condition being met, taking the plurality of related medical documents as the plurality of second medical documents.

### Claim 3 (depends on 2)

The medical document analysis method of claim 2, wherein after the step (a1), further comprises:
adding a new keyword to the plurality of keywords, or removing a excluded keyword from the plurality of keywords; and
assigning a plurality of weights to each of the plurality of keywords;
wherein the step (a2) further comprises:
searching for the plurality of related medical documents based on the plurality of weighted keywords.

### Claim 4 (depends on 1)

The medical document analysis method of claim 1, wherein the plurality of feature labels in the step of marking each of the plurality of second medical documents with the plurality of feature labels comprise a published time, a publication, a drug name label, a symptom label, an author label, a drug receptor label, and a dosage form label.

### Claim 5 (depends on 1)

The medical document analysis method of claim 1, wherein the step of projecting the plurality of second medical documents onto the different coordinate positions in the multi-dimensional map respectively comprises:
(b1) determining a word vector of each of the plurality of second medical documents on dimensions of the plurality of feature labels by using a natural language model; and
(b2) projecting the plurality of second medical documents onto the multi-dimensional map based on the word vector of each of the plurality of second medical documents.

### Claim 6 (depends on 1)

The medical document analysis method of claim 1, wherein the step of forming the label topology map based on the correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents comprises:
(c1) analyzing an experimental result mentioned in each of the plurality of third medical documents selected, and the experimental result comprises a drug effect, a drug antagonism, or a drug side effect;
(c2) selecting the plurality of third medical documents with a corresponding feature label for the first symptom, the drug formulation, the receptor name to the drug dosage form of each of the plurality of branch paths in the label topology map, and performing a strategy neural network simulation or a value neural network simulation to obtain a simulation success rate of each of the plurality of branch paths; and
(c3) in response to the simulation success rate of one of the plurality of branch paths in the label topology map being lower than a threshold, determining the one of the plurality of branch paths to be invalid.

### Claim 7 (depends on 6)

The medical document analysis method of claim 6, wherein the step of forming the label topology map based on the correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents comprises:
(c4) searching in at least one patent database for the first symptom, the drug formulation, the receptor name, and the drug dosage form passed by each of the plurality of branch paths in the label topology map; and
(c5) in response to a valid patent corresponding to the one of the plurality of branch paths being searched from the at least one patent database, determining the one of the plurality of branch paths in the label topology map to be invalid.

### Claim 8 (depends on 7)

The medical document analysis method of claim 7, wherein the step of selecting the target branch path from the plurality of branch paths comprises:
sorting the plurality of valid branch paths based on the simulation success rate of each of the plurality of valid branch paths; and
selecting the target branch path based on the plurality of sorted branch paths.

### Claim 9 (depends on 7)

The medical document analysis method of claim 7, wherein the step of selecting the target branch path from the plurality of branch paths comprises:
estimating a commercial value of each of the plurality of valid branch paths;
sorting the plurality of valid branch paths based on the simulation success rate and the commercial value of each of the plurality of valid branch paths; and
selecting the target branch path based on the plurality of sorted branch paths.

### Claim 10 (depends on 1)

The medical document analysis method of claim 1, wherein the step of displaying the information related to the target branch path further comprises:
selecting at least one fourth medical documents related to the target branch path from the plurality of third medical documents; and
displaying the at least one fourth medical documents.

### Claim 11

A medical document analysis system, comprising:
a communication transceiver unit, configured to communicatively connected to an academic database;
a user interface; and
a process unit, coupled to the communication transceiver unit and the user interface, and the process unit is configured to:
performing a recursive search based on keywords of a plurality of first medical documents to obtain a plurality of second medical documents;
analyzing contents of the plurality of second medical documents, and marking each of the plurality of second medical documents with a plurality of feature labels;
projecting the plurality of second medical documents onto different coordinate positions in a multi-dimensional map respectively based on at least one of the plurality of feature labels of each of the plurality of second medical documents, wherein dimensions of the multi-dimensional map correspond to the plurality of feature labels, and distances between the coordinate positions correspond to a correlation between the plurality of second medical documents;
in response to selecting a first symptom, estimating a plurality of second symptoms related to the first symptom;
selecting a plurality of coordinates matching the first symptom and the plurality of second symptoms from the multi-dimensional map, and selecting a plurality of third medical documents from the plurality of second medical documents based on a node adjacent to the plurality of coordinates;
analyzing a correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents in the multi-dimensional map to form a label topology map, wherein a top layer node in the label topology map is the first symptom, the label topology map comprises a plurality of branch paths, and each of the plurality of branch paths starts from the first symptom and passes through a drug formulation, a receptor name, and a drug dosage form; and
selecting a target branch path from the plurality of branch paths, and displaying information related to the target branch path.

### Claim 12 (depends on 11)

The medical document analysis system of claim 11, wherein the operation of performing the recursive search based on the keywords of the plurality of first medical documents comprises:
(a1) selecting a plurality of keywords from the plurality of first medical documents by using a natural language model;
(a2) searching for a plurality of related medical documents based on the plurality of keywords;
(a3) determining whether the plurality of related medical documents meet a convergence condition;
(a4) in response to the convergence condition not being met, updating the plurality of keywords by extracting the plurality of related medical documents by using the natural language model, and back to the operation (a3); and
(a5) in response to the convergence condition being met, taking the plurality of related medical documents as the plurality of second medical documents.

### Claim 13 (depends on 12)

The medical document analysis system of claim 12, wherein after the operation (a1), further comprises:
adding a new keyword to the plurality of keywords, or removing a excluded keyword from the plurality of keywords; and
assigning a plurality of weights to each of the plurality of keywords;
wherein the operation (a2) further comprises:
searching for the plurality of related medical documents based on the plurality of weighted keywords.

### Claim 14 (depends on 11)

The medical document analysis system of claim 11, wherein the plurality of feature labels in the operation of marking each of the plurality of second medical documents with the plurality of feature labels comprise a published time, a publication, a drug name label, a symptom label, an author label, a drug receptor label, and a dosage form label.

### Claim 15 (depends on 11)

The medical document analysis system of claim 11, wherein the operation of projecting the plurality of second medical documents onto the different coordinate positions in the multi-dimensional map respectively comprises:
(b1) determining a word vector of each of the plurality of second medical documents on dimensions of the plurality of feature labels by using a natural language model; and
(b2) projecting the plurality of second medical documents onto the multi-dimensional map based on the word vector of each of the plurality of second medical documents.

### Claim 16 (depends on 11)

The medical document analysis system of claim 11, wherein the operation of forming the label topology map based on the correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents comprises:
(c1) analyzing an experimental result mentioned in each of the plurality of third medical documents selected, and the experimental result comprises a drug effect, a drug antagonism, or a drug side effect;
(c2) selecting the plurality of third medical documents with a corresponding feature label for the first symptom, the drug formulation, the receptor name to the drug dosage form of each of the plurality of branch paths in the label topology map, and performing a strategy neural network simulation or a value neural network simulation to obtain a simulation success rate of each of the plurality of branch paths; and
(c3) in response to the simulation success rate of one of the plurality of branch paths in the label topology map being lower than a threshold, determining the one of the plurality of branch paths to be invalid.

### Claim 17 (depends on 16)

The medical document analysis system of claim 16, wherein the operation of forming the label topology map based on the correlation between the plurality of third medical documents and the at least one of the plurality of feature labels of each of the plurality of third medical documents comprises:
(c4) searching in at least one patent database for the first symptom, the drug formulation, the receptor name, and the drug dosage form passed by each of the plurality of branch paths in the label topology map; and
(c5) in response to a valid patent corresponding to the one of the plurality of branch paths being searched from the at least one patent database, determining the one of the plurality of branch paths in the label topology map to be invalid.

### Claim 18 (depends on 17)

The medical document analysis system of claim 17, wherein the operation of selecting the target branch path from the plurality of branch paths comprises:
sorting the plurality of valid branch paths based on the simulation success rate of each of the plurality of valid branch paths; and
selecting the target branch path based on the plurality of sorted branch paths.

### Claim 19 (depends on 17)

The medical document analysis system of claim 17, wherein the operation of selecting the target branch path from the plurality of branch paths comprises:
estimating a commercial value of each of the plurality of valid branch paths;
sorting the plurality of valid branch paths based on the simulation success rate and the commercial value of each of the plurality of valid branch paths; and
selecting the target branch path based on the plurality of sorted branch paths.

### Claim 20 (depends on 11)

The medical document analysis system of claim 11, wherein the operation of displaying the information related to the target branch path further comprises:
selecting at least one fourth medical documents related to the target branch path from the plurality of third medical documents; and
displaying the at least one fourth medical documents.
