# US 12,770,742 B2 · Determining erroneous codes in medical reports

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

- Number: US 12,770,742 B2
- Document ID: US12770742B2
- Status: Granted
- Kind: B2
- Application number: 16/980,141
- Filed: 2019-03-14
- Published: 2021-01-14
- Granted: 2026-10-20
- Earliest claimed: 2018-03-14
- Assignees: Koninklijke Philips N.V.
- Applicants: Koninklijke Philips N.V.
- Inventors: Xin Wang; Yugang Jia; Merlijn Sevenster
- Web page: https://patentgrep.com/patent/US12770742B2

## CPC

- G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients
- G06F 16/2379: {Updates performed during online database operations; commit processing}
- G06N 20/00: Machine learning
- 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)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems for determining an erroneous code in a medical report comprising a plurality of codes, each code representing a comment in the medical report the system, comprise a memory comprising instruction data representing a set of instructions and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to determine a respective vector representation for each of the plurality of codes in the medical report, wherein relative values of any selected pair of vector representations are correlated with a co-occurrence of the corresponding codes in a set of reference medical reports. The set of instructions when executed by the processor further cause the processor to determine an erroneous code in the medical report, based on the vector representations.

## Claims

### Claim 1

A computer-implemented method of determining an erroneous code in a medical report, comprising:
obtaining the medical report comprising a plurality of codes, each code representing a comment comprising clinical information;
generating a respective vector representation for each of the plurality of codes in the medical report, wherein values of a vector representation for a code of the plurality of codes are generated based on relation of the code in a set of reference medical reports to other codes in the set of reference medical reports, wherein relative values of a pair of vector representations are correlated with a co-occurrence of corresponding codes in the set of reference medical reports,
wherein the respective vector representation is generated by converting the code into a plurality of position values that each represent a position in a dimension of a hyper dimensional vector space, each position value being determined to reflect a frequency that the code co-occurs with a respective one of the other codes in the set of reference medical reports and assigned to a vector component of the respective vector representation;
identifying at least one of the generated vector representations as an outlying vector representation relative to other vector representations for the plurality of codes in the medical report, wherein said identifying comprises determining that the at least one generated vector representations is separated by more than a threshold separation from the other vector representations based on the position values in the hyper dimensional vector space; and
determining, based on the identification of the at least one generated vector representation as being an outlier and the position values in the hyper dimensional space, that the code represented by the at least one generated vector is an erroneous code in the medical report.

### Claim 2 (depends on 1)

The method of claim 1, wherein determining a respective vector representation for each of the plurality of codes comprises:
for each code:
determining a plurality of word vector representations, the plurality of word vector representations comprising a vector representation for each word in the comment represented by the code, wherein relative values of any selected pair of vector representations in the plurality of word vector representations are correlated with a co-occurrence of the corresponding words in the set of reference medical reports.

### Claim 3 (depends on 2)

The method of claim 2, further comprising, for each code:
concatenating the plurality of word vector representations to obtain a concatenated word vector representation for the code.

### Claim 4 (depends on 3)

The method of claim 3, wherein the concatenated word vector representation is used as the vector representation for the code.

### Claim 5 (depends on 3)

The method of claim 3, further comprising for each code:
concatenating the concatenated word vector representation with the vector representation for the code to create a combined vector representation for the code.

### Claim 6 (depends on 2)

The method of claim 2, further comprising for each code:
weighting each vector representation in the plurality of word-vector representations using the vector representation for the code;
determining an average of the weighted word-vector representations to create a weighted-average vector representation for the code.

### Claim 7 (depends on 1)

The method of claim 1, wherein determining a respective vector representation for each of the plurality of codes comprises:
using a machine learning process to determine each vector representation, based on a co-occurrence of the corresponding codes in the set of reference medical reports.

### Claim 8 (depends on 7)

The method of claim 7, wherein the machine learning process comprises a Word2Vec process.

### Claim 9 (depends on 1)

The method of claim 1, further comprising determining:
an average separation of the vector representations in a vector space; or
an average separation of a predetermined number of a most separated vector representations in a vector space.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
determining a probability that each code comprises an outlying code, based on a normalised measure of separation of the plurality of codes in a vector space.

### Claim 11 (depends on 1)

The method of claim 1, further comprising predicting at least one additional code that could be missing from the medical report, based on the vector representations.

### Claim 12

A non-transitory computer readable medium having stored instruction which, when executed by a processor, cause the processor to:
obtain a medical report comprising a plurality of codes, each code representing a comment comprising clinical information;
generate a respective vector representation for each of the plurality of codes in the medical report, wherein values of a vector representation for a code of the plurality of codes are generated based on relation of the code in a set of reference medical reports to other codes in the set of reference medical reports, wherein relative values of a pair of vector representations are correlated with a co-occurrence of corresponding codes in the set of reference medical reports,
wherein the respective vector representation is generated by converting the code into a plurality of position values that each represent a position in a dimension of a hyper dimensional vector space, each position value determined to reflect a frequency that the code co-occurs with a respective one of the other codes in the set of reference medical reports and assigned to a vector component of the respective vector representation;
identify at least one of the generated vector representations as an outlying vector representation relative to other vector representations for the plurality of codes in the medical report, wherein said identifying comprises determining that the at least one generated vector representations is separated by more than a threshold separation from the other vector representations based on the position values in the hyper dimensional vector space; and
determine, based on the identification of the at least one generated vector representation as being an outlier and the position values in the hyper dimensional space, that the code represented by the at least one generated vector is an erroneous code in the medical report.

### Claim 13

A system for determining an erroneous code in a medical report;
the system comprising:
a memory comprising a set of instructions;
a processor configured to communicate with the memory and to execute the set of instructions, wherein the set of instructions, when executed by the processor, cause the processor to:
obtain a medical report comprising a plurality of codes, each code representing a comment comprising clinical information;
generate a respective vector representation for each of the plurality of codes in the medical report, wherein values of a vector representation for a code of the plurality of codes are generated based on relation of the code in a set of reference medical reports to other codes in the set of reference medical reports, wherein relative values of any pair of vector representations are correlated with a co-occurrence of corresponding codes of the pair of vector representations in the set of reference medical reports,
wherein the respective vector representation is generated by converting the code into a plurality of position values that each represent a position in a dimension of a hyper dimensional vector space, each position value determined to reflect a frequency that the code co-occurs with a respective one of the other codes in the set of reference medical reports and assigned to a vector component of the respective vector representation;
identify at least one of the generated vector representations as an outlying vector representation relative to other vector representations for the plurality of codes in the medical report, wherein said identifying comprises determining that the at least one generated vector representations is separated by more than a threshold separation from the other vector representations based on the position values in the hyper dimensional vector space; and
determine, based on the identification of the at least one generated vector representation as being an outlier and the position values in the hyper dimensional space, that the code represented by the at least one generated vector is an erroneous code in the medical report.

### Claim 14 (depends on 13)

The system of claim 13, wherein the determination of a respective vector representation for each of the plurality of codes comprises, for each code:
determining a plurality of word vector representations, the plurality of word vector representations comprising a vector representation for each word in the comment represented by the code, wherein relative values of any selected pair of vector representations in the plurality of word vector representations are correlated with a co-occurrence of the corresponding words in the set of reference medical reports.

### Claim 15 (depends on 14)

The system of claim 14, wherein the set of instructions, when executed by the processor, further causes the processor, for each code, to;
concatenate the plurality of word vector representations to obtain a concatenated word vector representation for the code, wherein the concatenated word vector representation is used as the vector representation for the code; and
concatenate the concatenated word vector representation with the vector representation for the code to create a combined vector representation for the code.

### Claim 16 (depends on 14)

The system of claim 14, wherein the set of instructions, when executed by the processor, further causes the processor, for each code, to weight each vector representation in the plurality of word-vector representations using the vector representation for the code, and determine an average of the weighted word-vector representations to create a weighted-average vector representation for the code.

### Claim 17 (depends on 13)

The system of claim 13, wherein the set of instructions further causes the processor to determine an average separation of the vector representations in a vector space, or an average separation of a predetermined number of a most separated vector representations in a vector space.