There are few decision yet which specifically address the question to what extent machine-learning aspects make a technical contribution. This decision explores whether using machine learning for improving the accuracy of medical billing codes makes a technical contribution.

Here are the practical takeaways from the decision  T 0755/18 (Semi-automatic answering/3M INNOVATIVE PROPERTIES) of 11.12.2020 of Technical Board of Appeal 3.5.07:

Key takeaways

If neither the output of a machine-learning computer program nor the output's accuracy contribute to a technical effect, an improvement of the machine achieved automatically through supervised learning to generate a more accurate output is not in itself a technical effect (Catchword)

A billing code is non-technical administrative data. Generating a billing code is a cognitive task. The process of generating a billing code on the basis of documents is thus a non-technical administrative task, which, as such, is not patentable.

The invention

This European patent application concerns billing codes for medical billing. Such billing codes may relate to a hospital stay of a patient based on a collection of the documents containing information about the medical procedures that were performed on the patient during the stay and other billable activities performed by hospital staff. This set of documents may be viewed as a corpus of evidence for the billing codes that need to be generated and provided to an insurer for reimbursement.

The patent application starts from known computer-based support systems that guide human coders through the process of generating billing codes. Such systems typically include "concept extraction components" (e.g. to extract concepts like "allergy" or "prescription" from a medical report) and an "inference engine" that generates appropriate billing codes.

The invention sets out to improve the accuracy of such automatically generated billing codes.

To this end, the invention allows a human operator to provide input as to whether the generated billing codes are accurate (e.g. a verification status). The system may automatically interpret the feedback, and the reasoning process may be inverted in a probabilistic way to assign blame and/or praise for an incorrectly/correctly generated billing code to the constituent logic clauses which led to the generation of the billing code.

1050364a.jpg

Fig. 4 of EP2619661

Here is how the invention was defined in claim 1:

Claim 1 (main request)

A method performed by at least one computer processor executing computer program instructions tangibly stored on at least one non-transitory computer-readable medium, the method for use with a system including a data source and a first billing code, the first billing code being derived from a set of forward logic applied to first and second concept extraction components, the concept extraction components able [sic] to extract concepts from the data source, the method comprising:

(A) receiving input from a user, wherein the input represents a verification status of the first billing code;

(B) applying first inverse logic to the input, the billing code, and the set of forward logic, to identify the first and second concept extraction components; and

(C) applying reinforcement to the first and second concept extraction components, comprising: (C)(1) determining whether the verification status indicates that the first billing code is accurate; (C)(2) if the verification status indicates that the first billing code is inaccurate, then applying negative reinforcement to the first and second concept extraction components, comprising apportioning the negative reinforcement between the first and second concept extraction components."X. Claim 1 of the first auxiliary request reads as follows:"A method performed by at least one computer processor executing computer program instructions tangibly stored on at least one non-transitory computer-readable medium, the method for improving the accuracy of billing codes to be generated by a system, the system including a data source and having generated a first billing code, the first billing code being derived from a set of forward logic applied to the output of first and second concept extraction components, the concept extraction components being able to extract concepts from the data source, the method comprising: (A) receiving input from a user, wherein the input represents a verification status of the first billing code; (B) applying first inverse logic to the input, the billing code, and the set of forward logic, to identify the first and second concept extraction components, and (C) applying a reinforcement output to the first and second concept extraction components to thereby improve the accuracy of billing codes to be generated by the system, the applying comprising: (C)(1) determining whether the verification status indicates that the first billing code is accurate; (C)(2) if the verification status indicates that the first billing code is inaccurate, then applying a negative reinforcement output to the first and second concept extraction components, comprising apportioning the negative reinforcement output between the first and second concept extraction components."XI. Claim 1 of the second auxiliary request differs from that of the main request in that the text "and to generate concept codes" has been inserted after "the concept extraction components able to extract concepts from the data source", the word "and" at the end of step (B) has been removed, and the following text has been inserted at the end of the claim:", wherein a first reliability score is associated with the first concept extraction component, wherein the first reliability score represents an estimate of a first degree to which the first concept extraction component generates concept codes accurately, and wherein applying the negative reinforcement comprises associating a second reliability score with the first concept extraction component, wherein the second reliability score represents an estimate of a second degree to which the first concept extraction component generates concept codes accurately, wherein the second degree is lower than the first degree;

(D) determining whether the first concept extraction component is unreliable at generating concept codes; and

(E) if the first concept extraction component is determined to be unreliable at generating concept codes, then:

(E)(1) at the first concept extraction component, when generating a further concept code requiring human review of the further concept code before adding the further concept code to the data source.

Is it patentable?

The first-instance examining division had refused the application based on lack of inventive step over a standard general purpose computer.

On the appeal stage, the board of appeal assessed which of the features of the invention actually makes a technical contribution, and took the view:

A billing code is non-technical administrative data which may take the form of a textual representation, for instance "Unspecified diabetes" (see paragraph [0050] of the international publication). Generating a billing code (see also point 1. above) is a cognitive task (paragraphs [0002] and [0015]). The process of generating a billing code on the basis of documents is thus a non-technical administrative task, which, as such, is not patentable pursuant to Article 52(2) and (3) EPC.

The appellant had argued that simply because a certain feature offers a solution to an administrative, economic or business problem, it did not in and of itself prohibit the same feature from simultaneously solving a technical problem for which an applicant was entitled to seek protection. The board agreed that the presence of non-technical features in the claim does not mean that the claimed subject-matter is not patentable and that features which are non-technical when taken in isolation but which interact with technical features of the invention to solve a technical problem should be taken into account in assessing inventive step.

Moreover, the appellant argued that the invention used machine-learning techniques to improve the accuracy of the machine output. According to the appellant, the invention was technical because it improved the system so that it would generate more accurate billing codes in the future.

The board did not follow this argument:

In the board's opinion, if neither the output of a learning-machine computer program nor the machine output's accuracy contributes to a technical effect, an improvement of the machine achieved automatically through supervised learning for producing a more accurate output is not in itself a technical effect. In this case, the learning machine's output is a billing code, which is non-technical administrative data. The accuracy of the billing code refers to "administrative accuracy" regarding, for example, whether the billing code is consistent with information represented by a spoken audio stream or a draft transcript (paragraph [0051]) or is "justified by the given corpus of documents, considering applicable rules and regulations" (paragraph [0002]). Therefore, improving the learning machine to generate more accurate billing codes or, equivalently, improving the accuracy of the billing codes generated by the system, is as such not a technical effect.

Also the further arguments made by the appellant were not successful, and the main request was found to lack an inventive step. In addition, since none of the auxiliary requests was found to be allowable either, the appeal was dismissed in the end.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.