The diagnostic accuracy of differential diagnoses generated by artificial intelligence (AI) chatbots, including the generative pretrained transformer 3 (GPT-3) chatbot (ChatGPT-3) is unknown. This study evaluated the accuracy of differential-diagnosis lists generated by ChatGPT-3 for clinical vignettes with common chief complaints. General internal medicine physicians created clinical cases, correct diagnoses, and five differential diagnoses for ten common chief complaints. The rate of correct diagnosis by ChatGPT-3 within the ten differential-diagnosis lists was 28/30 (93.3%). The rate of correct diagnosis by physicians was still superior to that by ChatGPT-3 within the five differential-diagnosis lists (98.3% vs. 83.3%, p = 0.03). The rate of correct diagnosis by physicians was also superior to that by ChatGPT-3 in the top diagnosis (53.3% vs. 93.3%, p < 0.001). The rate of consistent differential diagnoses among physicians within the ten differential-diagnosis lists generated by ChatGPT-3 was 62/88 (70.5%). In summary, this study demonstrates the high diagnostic accuracy of differential-diagnosis lists generated by ChatGPT-3 for clinical cases with common chief complaints. This suggests that AI chatbots such as ChatGPT-3 can generate a well-differentiated diagnosis list for common chief complaints. However, the order of these lists can be improved in the future.

Background There is an increasing number of elderly patients with pulmonary arterial hypertension, and their characteristics differ from those of young or middle-aged patients with this condition. Case Summary A 73-year-old woman with a history of myocardial infarction and cardiovascular risk factors was admitted to the hospital with 2-week exertional dyspnoea. Her initial diagnosis was heart failure with preserved ejection fraction, but the symptoms persisted despite receiving treatment with diuretics. Additional tests showed a significant decrease in diffusing capacity of carbon monoxide and findings suggestive of severe pulmonary hypertension. Contrast-enhanced computed tomography of the chest, and pulmonary angiography, showed no narrowing or obstruction of the pulmonary arteries. Right heart catheterization revealed hemodynamic data implying pre-capillary pulmonary hypertension. Her condition gradually deteriorated to World Health Organization functional class IV, and sequential combination therapy with tadalafil, macitentan, and selexipag was initiated with a pulmonary arterial hypertension diagnosis; however, she died 1 month later. Pathological findings in autopsy were consistent with pulmonary arterial hypertension, and some parts of the lungs revealed the presence of obstructive and interstitial lung disease. Discussion The majority of elderly patients with pulmonary arterial hypertension might have multimorbidity. However, there is no specific treatment strategy. It is associated with diagnostic delay and worse prognosis; therefore, early suspicion and comprehensive tests, including right heart catheterization, are essential for better management.

No abstract

Purpose The data on the diagnostic contribution of general internal medicine (GIM) consultations for undiagnosed health problems from specialists are scarce. This study aims to explore the role of generalists as diagnostic medicine consultants in tertiary care settings. Patients and Methods We conducted a retrospective observational study at a Japanese university hospital. GIM consultations for diagnosis from other departments on outpatients aged ≧ 20 years from January 2018 to December 2020 were included. Data were extracted from electronic medical records. The primary outcome was new diagnosis rates. The secondary outcomes were new diagnosis rates with clinical significance and clinical outcomes at 90 days from the index visit. Results A total of 328 patients were included. The top five consulting departments were orthopedics (17.0%), cardiovascular (10.3%), otorhinolaryngology (8.8%), neurology (8.8%), and gastroenterology (7.9%). GIM identified 456 chief complaints (CCs), and the top five were fever (10.9%), abnormal laboratory results (8.3%), fatigue (5.9%), and pain (7.4%) or numbness (4.6%) in the extremities. There were 139 (104/328 patients: 31.8%) specialty consultations from GIM, and the top five departments were rheumatology (21.1%), gastroenterology (19.2%), orthopedics (9.6%), psychiatry (9.6%), and neurology (9.6%). In total, 277 new diagnoses were established in 232 patients (70.7%), and 203 patients had new diagnoses with clinical significance (61.8%). Clinical outcomes at 90 days from the time of the index visit were resolution/improvement (60.7%), unchanged/worsened (22.3%), and unknown (17.0%). Conclusion Over 70% of GIM consultations from other departments established new diagnoses with favorable outcomes in >60% of the patients.

No abstract