Background: Medical history contributes approximately 80% to a diagnosis, although physical examinations and laboratory investigations increase a physician’s confidence in the medical diagnosis. The concept of artificial intelligence (AI) was first proposed more than 70 years ago. Recently, its role in various fields of medicine has grown remarkably. However, no studies have evaluated the importance of patient history in AI-assisted medical diagnosis. Objective: This study explored the contribution of patient history to AI-assisted medical diagnoses and assessed the accuracy of ChatGPT in reaching a clinical diagnosis based on the medical history provided. Methods: Using clinical vignettes of 30 cases identified in The BMJ , we evaluated the accuracy of diagnoses generated by ChatGPT. We compared the diagnoses made by ChatGPT based solely on medical history with the correct diagnoses. We also compared the diagnoses made by ChatGPT after incorporating additional physical examination findings and laboratory data alongside history with the correct diagnoses. Results: ChatGPT accurately diagnosed 76.6% (23/30) of the cases with only the medical history, consistent with previous research targeting physicians. We also found that this rate was 93.3% (28/30) when additional information was included. Conclusions: Although adding additional information improves diagnostic accuracy, patient history remains a significant factor in AI-assisted medical diagnosis. Thus, when using AI in medical diagnosis, it is crucial to include pertinent and correct patient histories for an accurate diagnosis. Our findings emphasize the continued significance of patient history in clinical diagnoses in this age and highlight the need for its integration into AI-assisted medical diagnosis systems.
Key Clinical Message: Postherpetic pseudohernia must be suspected when a patient develops motor dysfunction coincident with or following a herpes zoster eruption.
Article For any disease a human can imagine, ChatGPT can generate a fake report was published on February 23, 2024 in the journal Diagnosis (volume 0, issue 0).
1 November 2020 in International Journal of General Medicine
Purpose: Fundoscopic exams are conducted during outpatient consultations to assess intracranial disease and complications from chronic diseases. Low level of confidence in these skills which physicians have is one of the causes that implementation frequency is low. Research has not yet identified specific measures through which the healthcare system may increase the implementation of fundoscopic exams nor a qualitative process that enables physicians to gain confidence in their fundoscopic exam skills. We introduced a checklist and conducted a mixed-methods study. Methods: This study is a before-and-after study, within an embedded-experimental mixed-methods design. We sampled 15 physicians in the department of general medicine at a university hospital assigned to initial consultation. We introduced a checklist to verify whether the fundoscopic exam was implemented. Measures are implementation ratio of the fundoscopic exam to the total number of indication cases, and Visual Analog Scale (VAS) values for the confidence in procedure are measured before and after the intervention. We obtained qualitative data from interviews and used the Modified Grounded Theory Approach. Results: We observed significant increases in the implementation ratio (19.2% (29/151 cases) vs 64.8% (105/162 cases), p< 0.001) and in the VAS value for the confidence (1.89 mm vs 4.68 mm (p< 0.001)). Analysis of the interviews revealed the following. To increase the implementation, it is necessary to reduce the and , which is prevented by the checklist. The leads to a . Repeated executions result in . Conclusion: The intervention increased the implementation ratio, thereby increasing successful experiences and confidence among physicians. The growth of confidence boosted motivations to implement fundoscopic exams.
A 37-year-old Indonesian man was referred to our hospital with a 30-day history of fever, chills and dry cough. He also reported night sweats and weight loss. Chest radiography at the previous hospital 2 weeks prior revealed no abnormalities ( Fig. 1a). His vital signs at presentation were unremarkable with no fever. Chest auscultation revealed late inspiratory fine crackles in the basal area of the lungs. Repeated chest radiography revealed diffuse, bilateral and small lung nodules ( Fig. 1b). High-resolution computed tomography (HRCT) of the chest revealed innumerable miliary nodules in both lungs. A sputum acid-fast bacilli (AFB) smear and the QuantiFERON-TB Gold (QFT) were negative. Subsequent two sputum AFB smears were also negative. Twelve days after presentation, Mycobacterium tuberculosis complex was identified from one of the sputum samples, and the patient was diagnosed with miliary tuberculosis (TB). He was initiated on isoniazid, rifampin, pyrazinamide and ethambutol.