Nair Sandeep, Lushington Gerald H, Purushothaman Mohan, Rubin Bernard, Jupe Eldon, Gattam Santosh
Progentec Diagnostics, Inc., 755 Research Pkwy, Oklahoma City, OK 73104, USA.
BioTech (Basel). 2025 Mar 6;14(1):15. doi: 10.3390/biotech14010015.
Systemic lupus erythematosus (SLE) is a complex autoimmune disease that poses serious long-term patient burdens. 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. 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. 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. 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.
系统性红斑狼疮（SLE）是一种复杂的自身免疫性疾病，给患者带来严重的长期负担。背景：SLE患者的分类和护理常常因病例异质性（症状和严重程度的多种变化）而变得复杂。大语言模型（LLMs）和生成式人工智能（genAI）可以通过分析医疗记录来评估关键医学标准，从而缓解这一挑战。方法：为了展示基于genAI的分析，使用美国风湿病学会（ACR）1997年SLE分类标准来定义与医学相关的LLM提示。通过五次genAI重复运行，对78名先前研究过的患者（45名被分类为患有SLE；33名不确定或阴性）的记录进行了计算分析。结果：genAI对“盘状红斑”和“胸膜炎或心包炎”分类标准的判定与临床分类完全一致，而一些因素如“免疫紊乱”（准确率56%）在统计学上不可靠。与临床分类相比，我们的genAI方法实现了72%的预测成功率。结论：genAI分类可能具有足够的预测性，有助于医学专业人员评估SLE患者并制定护理策略。对于个别标准，准确性似乎与临床判定的复杂性呈负相关，这意味着人工智能患者分析工具的改进可能源于临床分类效能的持续提高。
