Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling

Systemic lupus erythematosus (SLE) is a complex autoimmune disease that poses serious long-term patient burdens. <b>(1)</b> Background: SLE patient classification and care are often complicated by case heterogeneity (diverse variations in symptoms and severity). Large language models (LL...

Full description

Saved in:
Bibliographic Details
Main Authors: Sandeep Nair, Gerald H. Lushington, Mohan Purushothaman, Bernard Rubin, Eldon Jupe, Santosh Gattam
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:BioTech
Subjects:
Online Access:https://www.mdpi.com/2673-6284/14/1/15
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850089489303601152
author Sandeep Nair
Gerald H. Lushington
Mohan Purushothaman
Bernard Rubin
Eldon Jupe
Santosh Gattam
author_facet Sandeep Nair
Gerald H. Lushington
Mohan Purushothaman
Bernard Rubin
Eldon Jupe
Santosh Gattam
author_sort Sandeep Nair
collection DOAJ
description Systemic lupus erythematosus (SLE) is a complex autoimmune disease that poses serious long-term patient burdens. <b>(1)</b> Background: SLE patient classification and care are often complicated by case heterogeneity (diverse variations in symptoms and severity). Large language models (LLMs) and generative artificial intelligence (genAI) may mitigate this challenge by profiling medical records to assess key medical criteria. <b>(2)</b> Methods: To demonstrate genAI-based profiling, ACR (American College of Rheumatology) 1997 SLE classification criteria were used to define medically relevant LLM prompts. Records from 78 previously studied patients (45 classified as having SLE; 33 indeterminate or negative) were computationally profiled, via five genAI replicate runs. <b>(3)</b> Results: GenAI determinations of the “Discoid Rash” and “Pleuritis or Pericarditis” classification criteria yielded perfect concurrence with clinical classification, while some factors such as “Immunologic Disorder” (56% accuracy) were statistically unreliable. Compared to clinical classification, our genAI approach achieved a 72% predictive success rate. <b>(4)</b> Conclusions: GenAI classifications may prove sufficiently predictive to aid medical professionals in evaluating SLE patients and structuring care strategies. For individual criteria, accuracy seems to correlate inversely with complexities in clinical determination, implying that improvements in AI patient profiling tools may emerge from continued advances in clinical classification efficacy.
format Article
id doaj-art-a83f22c93d474ee5a6259782ab5ff843
institution DOAJ
issn 2673-6284
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series BioTech
spelling doaj-art-a83f22c93d474ee5a6259782ab5ff8432025-08-20T02:42:46ZengMDPI AGBioTech2673-62842025-03-011411510.3390/biotech14010015Prediction of Lupus Classification Criteria via Generative AI Medical Record ProfilingSandeep Nair0Gerald H. Lushington1Mohan Purushothaman2Bernard Rubin3Eldon Jupe4Santosh Gattam5Progentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USAProgentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USAProgentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USAProgentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USAProgentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USAProgentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USASystemic lupus erythematosus (SLE) is a complex autoimmune disease that poses serious long-term patient burdens. <b>(1)</b> Background: SLE patient classification and care are often complicated by case heterogeneity (diverse variations in symptoms and severity). Large language models (LLMs) and generative artificial intelligence (genAI) may mitigate this challenge by profiling medical records to assess key medical criteria. <b>(2)</b> Methods: To demonstrate genAI-based profiling, ACR (American College of Rheumatology) 1997 SLE classification criteria were used to define medically relevant LLM prompts. Records from 78 previously studied patients (45 classified as having SLE; 33 indeterminate or negative) were computationally profiled, via five genAI replicate runs. <b>(3)</b> Results: GenAI determinations of the “Discoid Rash” and “Pleuritis or Pericarditis” classification criteria yielded perfect concurrence with clinical classification, while some factors such as “Immunologic Disorder” (56% accuracy) were statistically unreliable. Compared to clinical classification, our genAI approach achieved a 72% predictive success rate. <b>(4)</b> Conclusions: GenAI classifications may prove sufficiently predictive to aid medical professionals in evaluating SLE patients and structuring care strategies. For individual criteria, accuracy seems to correlate inversely with complexities in clinical determination, implying that improvements in AI patient profiling tools may emerge from continued advances in clinical classification efficacy.https://www.mdpi.com/2673-6284/14/1/15systemic lupus erythematosus (SLE)medical records (MRs)large language model (LLM)natural language processing (NLP)generative artificial intelligence (genAI)American College of Rheumatology (ACR)
spellingShingle Sandeep Nair
Gerald H. Lushington
Mohan Purushothaman
Bernard Rubin
Eldon Jupe
Santosh Gattam
Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
BioTech
systemic lupus erythematosus (SLE)
medical records (MRs)
large language model (LLM)
natural language processing (NLP)
generative artificial intelligence (genAI)
American College of Rheumatology (ACR)
title Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
title_full Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
title_fullStr Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
title_full_unstemmed Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
title_short Prediction of Lupus Classification Criteria via Generative AI Medical Record Profiling
title_sort prediction of lupus classification criteria via generative ai medical record profiling
topic systemic lupus erythematosus (SLE)
medical records (MRs)
large language model (LLM)
natural language processing (NLP)
generative artificial intelligence (genAI)
American College of Rheumatology (ACR)
url https://www.mdpi.com/2673-6284/14/1/15
work_keys_str_mv AT sandeepnair predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling
AT geraldhlushington predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling
AT mohanpurushothaman predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling
AT bernardrubin predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling
AT eldonjupe predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling
AT santoshgattam predictionoflupusclassificationcriteriaviagenerativeaimedicalrecordprofiling