# US 11,600,391 B1 · Classifying and grouping service descriptors from health provider chargemasters

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

- Number: US 11,600,391 B1
- Document ID: US11600391B1
- Status: Granted
- Kind: B1
- Application number: 16/879,579
- Filed: 2020-05-20
- Granted: 2023-03-07
- Earliest claimed: 2019-05-28
- Applicants: Joanne Rodriques-Craig
- Inventors: Joanne Rodriques-Craig
- Web page: https://patentgrep.com/patent/US11600391B1

## CPC

- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G06F 16/285: {Clustering or classification}
- G06F 40/20: Natural language analysis (semantic analysis of natural language G06F40/30)
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06Q 30/0204: Market segmentation
- G16H 40/20: for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G16H 70/20: relating to practices or guidelines
- 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)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

Embodiments of a computer-assisted method classify service descriptors from chargemasters into national billing code equivalents for easy comparison of health provider prices. Medical libraries are attached and specialized rules are applied to augment service descriptors from chargemasters with supplementary medical information. A computer processor is configured to execute a text-based classifier and/or textual similarity metric to accurately associate service descriptors to national billing coding equivalents for service descriptors from different health providers. The disclosed embodiments form a comprehensive list of service descriptors grouped by national billing code equivalents and by textual similarity.

## Claims

### Claim 1

A computer-implemented method, comprising:
forming a comprehensive list of services provided by a plurality of health providers, the comprehensive list formed from a plurality of chargemasters received in a machine-readable format from each of the health providers, the chargemaster from each of the health providers including a plurality of service descriptors and corresponding prices represented as a plurality of character strings, each of the character strings corresponding to a healthcare service from one of the health providers having a different sequence of characters from a character string for the same healthcare service from another health provider, and modifying the service descriptor from each health provider into an augmented service descriptor in the comprehensive list representing a same healthcare service from each health provider, the modifying including:
storing, in a memory of a computer processor, the plurality of chargemasters, each of the plurality of chargemasters from a different health provider;
operating the computer processor to sort with a first set of input rules the plurality of chargemasters into a group of chargemasters including billing code equivalents and a group of chargemasters without billing code equivalents;
operating the computer processor to sort the group of chargemasters including billing code equivalents into a subgroup with an appropriate national billing code equivalent for each of a plurality of service descriptors and another subgroup without an appropriate national billing code equivalent for each of a plurality of service descriptors;
with the computer processor, applying a plurality of preprocessing rules to modify service descriptors from each chargemaster into the augmented service descriptors, the plurality of preprocessing rules including:
extending abbreviations into complete words;
replacing words with synonyms;
removing stopwords specific to each health provider; and
replacing numbers with equivalent text;
training a text-based classifier operated by the computer processor with the service descriptors from the group of chargemasters with national billing code equivalents;
classifying, with the trained text-based classifier, the service descriptors from the group of chargemasters without a national billing code equivalent into a national billing code equivalent;
applying a textual similarity metric to all the service descriptors without a national billing code equivalent, forming a group of service descriptors having a textual similarity above a selected threshold degree of confidence, and assigning a corresponding national billing code equivalent to the group of service descriptors in the comprehensive list;
receiving a request for a price comparison between the health providers for a healthcare service; and
displaying the national billing code equivalent, the augmented service descriptor, and the price from each of the health providers for the requested healthcare service retrieved from the comprehensive list, the comprehensive list having augmented service descriptors each formed with a higher accuracy and/or a higher consistency than expected for a classification of service descriptors by a human expert.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the classifying into a national billing code equivalent is performed when the classifying is above a threshold degree of confidence.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the subgroup with an appropriate national billing code equivalent includes the chargemasters having specifications corresponding to national billing code equivalents.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising classifying words selected from the service descriptors in each of the chargemasters into a word type according to character-by-character equivalence of the selected words and a list of words for each of the word types.

### Claim 5 (depends on 1)

The computer-assisted method of claim 1, wherein the classifying with the trained text-based classifier of service descriptors from the group of chargemasters without a national billing code equivalent further includes a variable importance factor, the variable importance factor preferentially selecting words, frequencies of words, and types of words most likely to occur in a selected national billing code equivalent.

### Claim 6 (depends on 1)

The computer-assisted method of claim 1, further comprising the calculation of a match score, and for a match score above a threshold degree of confidence, classifying the service descriptor into a national billing code equivalent.

### Claim 7 (depends on 1)

The computer-assisted method of claim 1, further comprising ranking national billing codes applicable to a service descriptor in order of probability.

### Claim 8 (depends on 1)

The computer-assisted method of claim 1, wherein the forming a group of service descriptors having textual similarity further comprises ordering the service descriptors according to a probability of each service descriptor being attached to a national billing code equivalent.

### Claim 9 (depends on 1)

The computer-assisted method of claim 1, further comprising:
selecting at least one additional textual similarity metric;
selecting a weighting factor for each of the textual similarity metrics; and
combing the textual similarity metrics according to the weighting factors.

### Claim 10 (depends on 1)

The computer-assisted method of claim 1, wherein a model for classifying service descriptors with national billing code equivalents includes a loss function.

### Claim 11 (depends on 10)

The computer-assisted method of claim 10, wherein the model comprises an ensemble of classification models, each of the classification models in the ensemble trained on a selected chargemaster service descriptor.

### Claim 12 (depends on 1)

The computer-assisted method of claim 1, further comprising forming a listing of services offered by health providers from the group of service descriptors having textual similarity above the selected threshold degree of confidence and an assigned national billing code equivalent.
