Evaluation of ChatGPT-Generated Differential Diagnosis for Common Diseases With Atypical Presentation: Descriptive Research.

Background: The persistence of diagnostic errors, despite advances in medical knowledge and diagnostics, highlights the importance of understanding atypical disease presentations and their contribution to mortality and morbidity. Artificial intelligence (AI), particularly generative pre-trained transformers like GPT-4, holds promise for improving diagnostic accuracy, but requires further exploration in handling atypical presentations. Objective: This study aimed to assess the diagnostic accuracy of ChatGPT in generating differential diagnoses for atypical presentations of common diseases, with a focus on the model's reliance on patient history during the diagnostic process. Methods: We used 25 clinical vignettes from the Journal of Generalist Medicine characterizing atypical manifestations of common diseases. Two general medicine physicians categorized the cases based on atypicality. ChatGPT was then used to generate differential diagnoses based on the clinical information provided. The concordance between AI-generated and final diagnoses was measured, with a focus on the top-ranked disease (top 1) and the top 5 differential diagnoses (top 5). Results: ChatGPT's diagnostic accuracy decreased with an increase in atypical presentation. For category 1 (C1) cases, the concordance rates were 17% (n=1) for the top 1 and 67% (n=4) for the top 5. Categories 3 (C3) and 4 (C4) showed a 0% concordance for top 1 and markedly lower rates for the top 5, indicating difficulties in handling highly atypical cases. The χ2 test revealed no significant difference in the top 1 differential diagnosis accuracy between less atypical (C1+C2) and more atypical (C3+C4) groups (χ²1=2.07; n=25; P=.13). However, a significant difference was found in the top 5 analyses, with less atypical cases showing higher accuracy (χ²1=4.01; n=25; P=.048). Conclusions: ChatGPT-4 demonstrates potential as an auxiliary tool for diagnosing typical and mildly atypical presentations of common diseases. However, its performance declines with greater atypicality. The study findings underscore the need for AI systems to encompass a broader range of linguistic capabilities, cultural understanding, and diverse clinical scenarios to improve diagnostic utility in real-world settings.

HPLC/PDA determination of carminic acid and 4-aminocarminic acid using relative molar sensitivities with respect to caffeine.

To accurately determine carminic acid (CA) and its derivative 4-aminocarminic acid (4-ACA), a novel, high-performance liquid chromatography with photodiode array detector (HPLC/PDA) method using relative molar sensitivity (RMS) was developed. The method requires no analytical standards of CA and 4-ACA; instead it uses the RMS values with respect to caffeine (CAF), which is used as an internal standard. An off-line combination of 1H-quantitative nuclear magnetic resonance spectroscopy (1H-qNMR) and HPLC/PDA was able to precisely determine the RMSs of CA274nm/CAF274nm and 4-ACA274nm/CAF274nm. To confirm the performance of the HPLC/PDA method using RMSs, the CA and 4-ACA contents in test samples were tested using four different HPLC-PDA instruments and one HPLC-UV. The relative standard deviations of the results obtained from five chromatographs and two columns were less than 2.7% for CA274nm/CAF274nm and 1.1% for 4-ACA274nm/CAF274nm. The 1H-qNMR method was directly employed to analyse the CA and 4-ACA contents in test samples. The differences between the quantitative values obtained from both methods were less than 5% for CA and 3% for 4-ACA. These results demonstrate that the HPLC/PDA method using RMSs to CAF is a simple and reliable quantification method that does not require CA and 4-ACA certified reference materials.

Possible contribution of rubiadin, a metabolite of madder color, to renal carcinogenesis in rats.

Madder color (MC) has been shown to exert carcinogenic potential in the rat kidney in association with degeneration, karyomegaly, increased cell proliferation of renal tubule cells and increased renal 8-OHdG levels. To clarify the causal relationship of components and metabolites of MC to renal carcinogenesis, male F344 rats were fed lucidin-3-O-primeveroside (LuP) or alizarin (Alz), and the genotoxic LuP metabolites lucidin (Luc) or rubiadin (Rub) for up to 26 weeks. After one week and four weeks, Luc did not induce any renal changes. In contrast, after one week, cortical tubule degeneration was apparent in the Alz and LuP groups, and cytoplasmic swelling with basophilic change and karyomegaly in the outer medulla was observed only in the Rub group. LuP and Rub increased the proliferative activity of tubule cells in the outer medulla, and Alz and LuP increased renal 8-OHdG levels. After 26 weeks, Rub but not Alz induced atypical tubules, a putative preneoplastic lesion, and karyomegaly in the outer medulla. These results indicate that Rub may be a potent carcinogenic metabolite of MC, targeting proximal tubule cells in the outer medulla, although oxidative stress increased by Alz or LuP might also be involved in renal carcinogenesis by MC.