Background: In the evolving field of health care, multimodal generative artificial intelligence (AI) systems, such as ChatGPT-4 with vision (ChatGPT-4V), represent a significant advancement, as they integrate visual data with text data. This integration has the potential to revolutionize clinical diagnostics by offering more comprehensive analysis capabilities. However, the impact on diagnostic accuracy of using image data to augment ChatGPT-4 remains unclear. Objective: This study aims to assess the impact of adding image data on ChatGPT-4’s diagnostic accuracy and provide insights into how image data integration can enhance the accuracy of multimodal AI in medical diagnostics. Specifically, this study endeavored to compare the diagnostic accuracy between ChatGPT-4V, which processed both text and image data, and its counterpart, ChatGPT-4, which only uses text data. Methods: We identified a total of 557 case reports published in the American Journal of Case Reports from January 2022 to March 2023. After excluding cases that were nondiagnostic, pediatric, and lacking image data, we included 363 case descriptions with their final diagnoses and associated images. We compared the diagnostic accuracy of ChatGPT-4V and ChatGPT-4 without vision based on their ability to include the final diagnoses within differential diagnosis lists. Two independent physicians evaluated their accuracy, with a third resolving any discrepancies, ensuring a rigorous and objective analysis. Results: The integration of image data into ChatGPT-4V did not significantly enhance diagnostic accuracy, showing that final diagnoses were included in the top 10 differential diagnosis lists at a rate of 85.1% (n=309), comparable to the rate of 87.9% (n=319) for the text-only version (P=.33). Notably, ChatGPT-4V’s performance in correctly identifying the top diagnosis was inferior, at 44.4% (n=161), compared with 55.9% (n=203) for the text-only version (P=.002, χ2 test). Additionally, ChatGPT-4’s self-reports showed that image data accounted for 30% of the weight in developing the differential diagnosis lists in more than half of cases. Conclusions: Our findings reveal that currently, ChatGPT-4V predominantly relies on textual data, limiting its ability to fully use the diagnostic potential of visual information. This study underscores the need for further development of multimodal generative AI systems to effectively integrate and use clinical image data. Enhancing the diagnostic performance of such AI systems through improved multimodal data integration could significantly benefit patient care by providing more accurate and comprehensive diagnostic insights. Future research should focus on overcoming these limitations, paving the way for the practical application of advanced AI in medicine.

Background: Generative artificial intelligence (GAI) systems by Google have recently been updated from Bard to Gemini and Gemini Advanced as of December 2023. Gemini is a basic, free-to-use model after a user’s login, while Gemini Advanced operates on a more advanced model requiring a fee-based subscription. These systems have the potential to enhance medical diagnostics. However, the impact of these updates on comprehensive diagnostic accuracy remains unknown. Objective: This study aimed to compare the accuracy of the differential diagnosis lists generated by Gemini Advanced, Gemini, and Bard across comprehensive medical fields using case report series. Methods: We identified a case report series with relevant final diagnoses published in the American Journal Case Reports from January 2022 to March 2023. After excluding nondiagnostic cases and patients aged 10 years and younger, we included the remaining case reports. After refining the case parts as case descriptions, we input the same case descriptions into Gemini Advanced, Gemini, and Bard to generate the top 10 differential diagnosis lists. In total, 2 expert physicians independently evaluated whether the final diagnosis was included in the lists and its ranking. Any discrepancies were resolved by another expert physician. Bonferroni correction was applied to adjust the P values for the number of comparisons among 3 GAI systems, setting the corrected significance level at P value <.02. Results: In total, 392 case reports were included. The inclusion rates of the final diagnosis within the top 10 differential diagnosis lists were 73% (286/392) for Gemini Advanced, 76.5% (300/392) for Gemini, and 68.6% (269/392) for Bard. The top diagnoses matched the final diagnoses in 31.6% (124/392) for Gemini Advanced, 42.6% (167/392) for Gemini, and 31.4% (123/392) for Bard. Gemini demonstrated higher diagnostic accuracy than Bard both within the top 10 differential diagnosis lists (P=.02) and as the top diagnosis (P=.001). In addition, Gemini Advanced achieved significantly lower accuracy than Gemini in identifying the most probable diagnosis (P=.002). Conclusions: The results of this study suggest that Gemini outperformed Bard in diagnostic accuracy following the model update. However, Gemini Advanced requires further refinement to optimize its performance for future artificial intelligence–enhanced diagnostics. These findings should be interpreted cautiously and considered primarily for research purposes, as these GAI systems have not been adjusted for medical diagnostics nor approved for clinical use.

The utility of remote auscultation was unknown. This study aimed to evaluate internet-connected real-time remote auscultation using cardiopulmonary simulators. In this open-label randomized controlled trial, the physicians were randomly assigned to the real-time remote auscultation group (intervention group) or the classical auscultation group (control group). After the training session, the participants had to classify the ten cardiopulmonary sounds in random order as the test session. In both sessions, the intervention group auscultated with an internet-connected electronic stethoscope. The control group performed direct auscultation using a classical stethoscope. The total scores for correctly identified normal or abnormal cardiopulmonary sounds were 97/100 (97%) in the intervention group and 98/100 (98%) in the control group with no significant difference between the groups (p > 0.99). In cardiac auscultation, the test score in the control group (94%) was superior to that in the intervention group (72%, p < 0.05). Valvular diseases were not misclassified as normal sounds in real-time remote cardiac auscultation. The utility of real-time remote cardiopulmonary auscultation using an internet-connected electronic stethoscope was comparable to that of classical auscultation. Classical cardiac auscultation was superior to real-time remote auscultation. However, real-time remote cardiac auscultation is useful for classifying valvular diseases and normal sounds.

BACKGROUND Clinical decision support systems (CDSSs) are widely used in various health care settings. In Japan, pressure ulcers are becoming a major concern in an aging society due to their increasing prevalence. However, management is often handled by nonspecialists in wound care due to regional disparities in specialist availability. OBJECTIVE To provide support for nonspecialists in wound care, we developed a prototype smartphone-based CDSS for pressure ulcer management. The system prompts users to answer questions about the wound's condition and recommends appropriate ointments and wound dressings by using a safety-first approach. This study aims to evaluate the utility of this system. METHODS We conducted a randomized crossover pilot study involving 28 general internal medicine (GIM) physicians. Participants were randomly assigned to group A (intervention-control) or group B (control-intervention). Participants evaluated 10 standardized pressure ulcer photographs and selected the most appropriate ointment and wound dressing for each. The unit of analysis was the individual response to each question (N=280 total observations). We used generalized estimating equations with an exchangeable correlation structure to account for within-subject clustering and adjust for potential period and sequence effects. RESULTS The overall correct response rate during the intervention phase was significantly higher than that during the control phase (49.3% vs 4.3%, respectively). After adjusting for clustering and crossover biases, the use of CDSS was associated with a 29.1-fold increase in the odds of a correct response (95% CI 8.2-103; P<.001). Secondary analyses revealed significant improvements in ointment selection (adjusted odds ratio [aOR] 2.4, 95% CI 1.5-3.8; P<.001) and wound dressing selection (aOR 8.9, 95% CI 4.9-16.1; P<.001). However, no significant period (P=.11) or sequence (P=.25) effects were observed for the primary outcome. CONCLUSIONS The prototype CDSS improved the accuracy of treatment decisions made by GIM physicians in a pilot study that used photographs and fixed options. Within the parameters of this investigation, CDSS effectively guided participants toward standardized, safety-oriented choices as defined by our scoring criteria. CLINICALTRIAL UMIN Clinical Trials Registry UMIN000057294; https://tinyurl.com/36a6vvah

The aim of this study was to evaluate short-term outcomes and safety of elobixibat in home-care patients with chronic constipation (CC) and explore the changes in caregiver burden related to bowel management. This multicenter prospective observational study evaluated bowel movement status, defecation-related symptoms, and caregiver burden in home-care patients diagnosed with CC who had hard stools classified as Bristol Stool Form Scale (BSFS) type 1 or 2. Elobixibat was administered to home-care patients once a day for 4 weeks before a meal. The primary endpoint was the proportion of patients with normal stools at week 2 in patients who were able to excrete independently. Secondary endpoints included the proportion of patient with normal stools at week 4, stool consistency, stool frequency, defecation-related symptoms at weeks 2 and 4, caregiver burden at week 4, concomitant laxative use during the study period, and adverse events. Analyses were based on available data at each assessment time point. The proportion of patient with normal stools at week 2 in patients who could excrete independently was 53% (8/15; 95% CI, 30–75%). Weekly defecation frequency significantly increased at week 4 (n = 20, p = 0.02, without multiplicity adjustment) and was associated with potential improvements in defecation-related symptoms (straining, abdominal bloating) and patient satisfaction. The most common adverse event was mild diarrhea (17%), with no new safety signals observed over 4 weeks. The caregiver questionnaire, evaluated in a limited subset of seven respondents, showed preliminary descriptive changes in physical, psychological, and social burdens. This uncontrolled, exploratory observational study suggests that elobixibat was associated with improvements in defecation-related outcomes in home-care patients with CC. However, these findings should be interpreted with caution due to the small sample size and potential confounding from concomitant laxative use. Caregiver burden findings were descriptive and hypothesis-generating and require confirmation in larger, controlled studies.

Background: Clinical decision support systems (CDSSs) are widely used in various health care settings. In Japan, pressure ulcers are becoming a major concern in an aging society due to their increasing prevalence. However, management is often handled by nonspecialists in wound care due to regional disparities in specialist availability. Objective: To provide support for nonspecialists in wound care, we developed a prototype smartphone-based CDSS for pressure ulcer management. The system prompts users to answer questions about the wound's condition and recommends appropriate ointments and wound dressings by using a safety-first approach. This study aims to evaluate the utility of this system. Methods: We conducted a randomized crossover pilot study involving 28 general internal medicine (GIM) physicians. Participants were randomly assigned to group A (intervention-control) or group B (control-intervention). Participants evaluated 10 standardized pressure ulcer photographs and selected the most appropriate ointment and wound dressing for each. The unit of analysis was the individual response to each question (N=280 total observations). We used generalized estimating equations with an exchangeable correlation structure to account for within-subject clustering and adjust for potential period and sequence effects. Results: The overall correct response rate during the intervention phase was significantly higher than that during the control phase (49.3% vs 4.3%, respectively). After adjusting for clustering and crossover biases, the use of CDSS was associated with a 29.1-fold increase in the odds of a correct response (95% CI 8.2-103; P<.001). Secondary analyses revealed significant improvements in ointment selection (adjusted odds ratio [aOR] 2.4, 95% CI 1.5-3.8; P<.001) and wound dressing selection (aOR 8.9, 95% CI 4.9-16.1; P<.001). However, no significant period (P=.11) or sequence (P=.25) effects were observed for the primary outcome. Conclusions: The prototype CDSS improved the accuracy of treatment decisions made by GIM physicians in a pilot study that used photographs and fixed options. Within the parameters of this investigation, CDSS effectively guided participants toward standardized, safety-oriented choices as defined by our scoring criteria. Trial Registration: UMIN Clinical Trials Registry UMIN000057294; https://tinyurl.com/36a6vvah