Demos
Case Study
Maximizing Patient Care through AI-Enhanced HCC Code Discovery
By leveraging Medical LLMs from John Snow Labs, WVU Medicine identified and extracted relevant HCC codes from clinical notes. Newly found diagnosis codes are then communicated to clinicians for review directly inside the EHR.
-
Streamlined HCC coding
-
Enhanced accuracy and
completeness -
Improved patient outcomes
The Problem
Understanding Clinical Notes is Essential to Accurate Patient Risk Adjustment
Research consistently suggests that using only structured data, such as diagnosis codes from problem lists or claims, can mean missing or wrong codes for almost half of all patients
Accuracy and Completeness
of Clinical Coding Using
ICD-10 for Ambulatory Visits
of Clinical Coding Using
ICD-10 for Ambulatory Visits
Horsky et. Al., AMIA Annu Symp Proc. 2017; 2017: 912–920.
“Just over a half of entered codes were appropriate for a given scenario and about a quarter were omitted.”
Data gaps in electronic health record (EHR) systems: An audit of problem list completeness during the COVID-19 pandemic
Poulus et. al., International Journal of Medical Informatics,
Volume 150, June 2021
Volume 150, June 2021
“recording of medical diagnoses on the structured problem list for inpatients is incomplete, with almost 40% of important diagnoses mentioned only in the free text notes.”
The Solution
01 | Import Patient Charts and Extract Medical Diagnoses
02 | Infer Diagnosis Codes Relevant for HCC Coding
03 | Calculate RAF (Risk Adjustment Factor)
If needed, the software includes current RAF calculators, which automate the process of ranking which patients should be manually reviewed first.
04 | Manual Review
Automated integration into the EHR (for review by clinicians) or the Generative AI Lab (for review by clinical coders).
The user interface includes a full audit trail, work queues, team management, analytics, medical terminologies, and an API for automated integration into downstream workflows.
Solution Accelerator
Automated Patient Risk Adjustment and Medicare HCC Coding from Clinical Notes
See how John Snow Labs and Databricks built a solution that derives missed ICD codes, preventing lower risk adjustment for patients and lost revenue for the provider.
The solution delivers a secure, enterprise-ready environment for payers or at-risk providers.
