Clinical content analytics engine

Inventors

DeJori, Mathaeus

Assignees

Zynx Health Inc

Abstract

Clinical content analytics engines and associated processes are described. An engine receives a clinical decision support document, accesses corresponding reference content, identifies and extracts medical intervention content from the clinical decision support document, segments extracted medical intervention content into a first plurality of segments including at least a first segment comprising a first set of text, determines if the first segment corresponds to at least a first item included in the reference content, the first item comprising a second set of text comprising terminology different than that found in the first set of text, and in response to determining that the first segment corresponds to the first item included in the reference content, causing a report to include an indication that the first segment corresponds to the first item included in the reference content.

Core Innovation

The invention is a computerized, machine learning system and an analyzing method for comparing a received clinical decision support (CDS) document with reference content comprising clinical guidelines. It identifies and extracts medical intervention content from the clinical decision support document, segments at least a portion of the extracted content into a first plurality of segments, and determines using the machine learning module whether a segment corresponds to a reference content item, including cases where terminology is different.

The machine learning module is configured to process fuzzy rules to identify whether a segment from the CDS document satisfies a corresponding segment from the reference content. The system improves machine learning system accuracy in identifying CDS document deficiencies and consistencies with respect to the reference content by repeatedly training the machine learning module based on new incoming data and collected positive and negative cases from CDS documents.

At least partly in response to the correspondence determinations, the system dynamically generates a version of the CDS document that includes a visual indication of which segments correspond to reference content items and which segments fail to include corresponding items. The document generation visually marks positive matches and missing items, supporting deficiencies and consistencies assessment relative to the reference content.

Claims Coverage

The provided material includes two independent claims, each centered on a fuzzy-rules-based machine learning module that segments extracted medical intervention content from a CDS document, determines correspondence to items in reference clinical guideline content, and dynamically generates a CDS document version with visual indications of matches and missing items. Across the independent claims, the core coverage includes repeated training with positive and negative cases, correspondence determination between segments and reference items including terminology differences, and dynamic generation of a visually annotated document.

Across the independent claims, the inventive approach is covered by combining a fuzzy-rules-based machine learning module with repeated training using positive and negative CDS cases, performing extraction and segmentation of medical intervention content, determining correspondence between CDS segments and reference guideline items, and dynamically generating a version of the CDS document that includes visual indications of correspondence and missing items.

Stated Advantages

Documented Applications

No documented applications found