# US 2012/0254083 A1 · System and method for automatically generating a medical code

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

- Number: US 2012/0254083 A1
- Document ID: US20120254083A1
- Status: Published application
- Kind: A1
- Application number: 13/179,607
- Filed: 2011-07-11
- Published: 2012-10-04
- Earliest claimed: 2011-03-31
- Assignees: Infosys Technologies Limited
- Inventors: Rajesh Lawrence PATRAO; Gururaj Bhimsen RAO; Guruprasad HOLLA
- Web page: https://patentgrep.com/patent/US20120254083A1

## CPC

- G16H 15/00: ICT specially adapted for medical reports, e.g. generation or transmission thereof
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

Disclosed herein is a system which address the problem of multiple mappings of a source ICD code to a target ICD code by using an adaptive mechanism to automatically convert to a target ICD code with the help of accompanying data and associated rules. The mechanism is based on analysis of the ICD code description, and analysis of accompanying data to determine a set of parameters to assist in the conversion. The rules are assigned numerical values to indicate the confidence of conversion. An adaptive mechanism is also disclosed that automatically tracks user overrides to refine the rules and the indicator values over time.

## Claims

### Claim 1

A computer-implemented method for a probabilistic target mapping of a source ICD code to one or more target ICD codes, the method comprising:
generating scenario tokens, supplementary tokens and identifier tokens from the target ICD codes descriptions;
calculating an attribute score for each of:
a. at least one of a scenario;
b. at least one of choice list determined using supplementary tokens; and
c. at least one of the choice list determined using identifier tokens;
calculating a total code score for each of the target ICD codes descriptions; and
generating a probabilistic target ICD code based on the total code score.

### Claim 2 (depends on 1)

The method of claim 1, wherein the scenario tokens, supplementary tokens and the identifier tokens are compared to one or more configured parameters.

### Claim 3 (depends on 1)

The method of claim 1, wherein at least one scenario is determined by comparing the scenario tokens to medical notes.

### Claim 4 (depends on 1)

The method of claim 1, wherein at least one choice list is determined by comparing the supplementary tokens and the identifier tokens to medical notes.

### Claim 5 (depends on 1)

The method of claim 1, wherein the attribute score is calculated using a search percentage match and an attribute weightage.

### Claim 6 (depends on 5)

The method of claim 5, wherein the search percentage match is calculated based on a search performed within a scenario slop, a supplementary slop and an identifier slop.

### Claim 7 (depends on 5)

The method of claim 5, wherein the attribute weightages for each of the scenario and the choice list is retrieved from a database.

### Claim 8 (depends on 1)

The method of claim 1, wherein the total score is calculated for at least one of the scenario data, the supplementary data and the identifier data or a combination of the scenario data, the supplementary data and the identifier data.

### Claim 9 (depends on 1)

The method of claim 1, wherein the scenario tokens are generated from target ICD code descriptions.

### Claim 10 (depends on 1)

The method of claim 1, wherein the scenario tokens, the supplementary tokens and the identifier tokens are generated using a synonyms database.

### Claim 11 (depends on 1)

The method of claim 1, wherein the supplementary tokens are generated by searching for a common sub-sequence from at least one of the choice list.

### Claim 12 (depends on 1)

The method of claim 1, wherein the identifier tokens are generated by selecting at least one singular word within the proximity of the common sub-sequence used to generate the supplementary tokens

### Claim 13 (depends on 1)

The method of claim 1, wherein the probabilistic target ICD code is identified by comparing the total code score for each of the target ICD codes with a lower threshold value and an upper threshold value, wherein the lower threshold value and the upper threshold values are retrieved from a database.

### Claim 14 (depends on 1)

The method of claim 1, wherein a virtual scenario is created in the event the target code descriptions contains non-multiple scenarios.

### Claim 15 (depends on 1)

The method of claim 1, further comprising an adaptive engine to adapt particulars entered by a user, wherein the adaptive engine is triggered in the event the system is unable to determine at least one of the scenario data.

### Claim 16 (depends on 15)

The method of claim 15, wherein the particulars entered by a user is adapted in subsequent mapping of the source ICD code to the one or more target ICD code.

### Claim 17 (depends on 1)

The method of claim 1, wherein the source ICD code is one of an ICD-9 code and an ICD-10 code.

### Claim 18 (depends on 1)

The method of claim 1, wherein each of the one or more target ICD codes is one of an ICD-9 code and an ICD-10 code.

### Claim 19

An automated system for a probabilistic target mapping of a source ICD code to one or more target ICD codes, the system comprising:
a token generating module for generating scenario tokens, supplementary tokens and identifier tokens from the target ICD codes descriptions;
a processor configured to calculate an attribute score for each of:
a. at least one of the scenario data;
b. at least one of the choice list determined using supplementary tokens; and
c. at least one of the choice list determined using identifier tokens;
calculate a total code score for each of the equivalent target ICD codes; and
a code generator module for generating a probabilistic target ICD code based on the total code score.

### Claim 20 (depends on 19)

The automated system of claim 19, further comprising a rule engine to compare the scenario tokens, supplementary tokens and the identifier tokens to one or more configured parameters.

### Claim 21 (depends on 20)

The automated system of claim 20, wherein the rule engine compares the scenario tokens, supplementary tokens and the identifier tokens to medical notes.

### Claim 22 (depends on 19)

The automated system of claim 19, wherein the attribute score is calculated using a search percentage match and an attribute weightage.

### Claim 23 (depends on 22)

The automated system of claim 22, wherein the search percentage match is calculated based on a search performed within a scenario slop, a supplementary slop and an identifier slop.

### Claim 24 (depends on 22)

The automated system of claim 22, wherein the attribute weightages for each of the scenario and the choice list is retrieved from a database.

### Claim 25 (depends on 19)

The automated system of claim 19, wherein the total score is calculated for at least one of the scenario data, the supplementary data and the identifier data or a combination of the scenario data, the supplementary data and the identifier data.

### Claim 26 (depends on 19)

The automated system of claim 19, wherein the scenario tokens are generated from target ICD code descriptions.

### Claim 27 (depends on 19)

The automated system of claim 19, wherein the supplementary tokens are generated by searching for a common sub-sequence from at least one of the choice list.

### Claim 28 (depends on 19)

The automated system of claim 19, wherein the identifier tokens are generated by selecting at least one singular word within the proximity of the common sub-sequence used to generate the supplementary tokens.

### Claim 29 (depends on 19)

The automated system of claim 19, wherein the scenario tokens, the supplementary tokens and the identifier tokens are generated using a synonyms database.

### Claim 30 (depends on 19)

The automated system of claim 19, wherein the probabilistic target ICD code is identified by comparing the total code score for each of the target ICD codes with a lower threshold value and an upper threshold value, wherein the lower threshold value and the upper threshold value is retrieved from the database.

### Claim 31 (depends on 19)

The automated system of claim 19, further comprising an adaptive engine to adapt particulars entered by a user, wherein the adaptive engine is triggered in the event the system is unable to determine at least of the scenario data.

### Claim 32 (depends on 31)

The automated system of claim 31, wherein the particulars entered by a user is adapted in subsequent mapping of the source ICD code to the one or more target ICD code.

### Claim 33 (depends on 19)

The automated system of claim 19, wherein a virtual scenario is created in the event the target code descriptions contains a large number of target codes

### Claim 34 (depends on 19)

The automated system of claim 19, wherein the source ICD code is one of an ICD-9 code and an ICD-10 code.

### Claim 35 (depends on 19)

The automated system of claim 19, wherein each of the one or more target ICD codes is one of an ICD-9 code and an ICD-10 code.
