Search

Low AC loss and high current density NbTi/Cu/CuNi double stage cabling superconductor for quick response superconducting generator (SCG) has been developed. Its characteristics were measured by short sample and several coil models. The critical current of 14kA at 6T and the AC loss of 10kW/[m.sup.3] were achieved, and coil models were excited under high current change rate operation without the quench. The developed superconductor will be able to be applied to a 70MW class quick response SCG.

Symptoms such as vertigo and unsteady gait occur in various diseases and are among the relatively common chief complaints. Even at present, the mechanisms underlying these disorders are unclear. We report a significant correlation between a prolonged period of resolution of benign paroxysmal positional vertigo (BPPV) and histories of lifestyle-related illnesses. We consider the possibility of correlating between BPPV prognosis and arteriosclerotic changes. Using carotid ultrasonography, we examined maximum intima-media thickness (IMT), maximum common carotid artery IMT, and biochemical examinations in 105 patients with peripheral vertigo. We divided patients with BPPV into groups with and without abnormal thickness of the IMT. The maximum IMT was 1.35 mm in patients with peripheral vestibular disorders. The proportion of peripheral vestibular disorder patients with a maximum IMT of ⩾ 1.1 mm (i.e., thickening) was 58%. The rate at which the feeling of positional vertigo remained at the halfway point in the observation period was significantly higher in the group of patients with an IMT of ⩾ 1.1 mm ( p = .0007). Our results indicate that cervical ultrasonography is useful for noninvasive examination of arteriosclerotic changes in patients with peripheral vestibular disorders. We saw indications that such patients show progression of arteriosclerotic changes. This study suggested that the arteriosclerotic change was related to prognosis. [ABSTRACT FROM AUTHOR]

The mechanism underlying pancreaticopleural fistula is considered the focal inflammation of a pancreatic duct branch caused by pancreatic secretion leakage from a ruptured pseudocyst or pancreatic duct. In the present case, chronic pancreatitis due to alcohol abuse resulted in a pancreatic pseudocyst, continuing in the mediastinum, that triggered inflammation of the mediastinal interspace through the retroperitoneum and reading to massive pleural effusion. [Extracted from the article]

Clinical experiences, helping relationships, and reflection are key factors for personal development for physicians. However, few studies have shown which experiences are important for personal growth and how medical residents specifically use their experiences for personal growth. The aim of this study was to identify from the medical residents' perspective which clinical experiences contribute to their personal development. We employed a qualitative design, conducting semi-structured interviews with ten postgraduate year two medical residents at a Japanese teaching hospital. The interviews were transcribed in interview memos, anonymized, and subjected to reflective thematic analysis to generate themes relevant to personal and professional development. Successful clinical experiences with autonomy and responsibility in clinical management were shown to be essential points for personal development as a physician. Autonomy in this study was the attitude of making one's own choices when managing patients. Responsibility was the obligation of the resident to take charge of a patient. Instructing junior trainees, appreciation received from patients, and approval granted by attending physicians reinforced their feelings of personal growth. The realization of what experiences and concepts influence medical residents' personal growth and development will make their professional development more effective.

Orthostatic intolerance (OI) is a key symptom of long COVID; however, the pathophysiology remains unknown. Among 688 long COVID patients who visited our clinic during the period from February 2021 to April 2023, 86 patients who were suspected of having OI and who underwent an active standing test (ST) were investigated to elucidate the clinical characteristics of OI in patients with long COVID. Of the 86 patients, 33 patients (38%) were ST-positive. Nausea and tachycardia in daily life were frequent complaints in the ST-positive group. The increase in heart rate (HR) during the ST was significantly greater during a 10-min period after standing in the ST-positive group (+ 30 bpm) than in the ST-negative group (+ 16 bpm). The initial increase in diastolic blood pressure (DBP) just after standing was significantly greater in the ST-positive group (+ 14 mmHg) than in the ST-negative group (+ 9 mmHg). Serum cortisol levels in the ST-positive patients aged over 20 years were higher and growth hormone levels in the patients under 20 years of age were lower than those in the ST-negative group. Autonomous nervous symptoms, transient DBP rise with increasing HR after standing, and endocrine dysfunctions are helpful for detecting OI related to long COVID., (© 2024. The Author(s).)

Background: Diagnostic performance of generative artificial intelligences (AIs) using large language models (LLMs) across comprehensive medical specialties is still unknown., Objective: We aimed to evaluate the diagnostic performance of generative AIs using LLMs in complex case series across comprehensive medical fields., Methods: We analyzed published case reports from the American Journal of Case Reports from January 2022 to March 2023. We excluded pediatric cases and those primarily focused on management. We utilized three generative AIs to generate the top 10 differential-diagnosis (DDx) lists from case descriptions: the fourth-generation chat generative pre-trained transformer (ChatGPT-4), Google Gemini (previously Bard), and LLM Meta AI 2 (LLaMA2) chatbot. Two independent physicians assessed the inclusion of the final diagnosis in the lists generated by the AIs., Results: Out of 557 consecutive case reports, 392 were included. The inclusion rates of the final diagnosis within top 10 DDx lists were 86.7% (340/392) for ChatGPT-4, 68.6% (269/392) for Google Gemini, and 54.6% (214/392) for LLaMA2 chatbot. The top diagnoses matched the final diagnoses in 54.6% (214/392) for ChatGPT-4, 31.4% (123/392) for Google Gemini, and 23.0% (90/392) for LLaMA2 chatbot. ChatGPT-4 showed higher diagnostic accuracy than Google Gemini ( P  < 0.001) and LLaMA2 chatbot ( P  < 0.001). Additionally, Google Gemini outperformed LLaMA2 chatbot within the top 10 DDx lists ( P  < 0.001) and as the top diagnosis ( P  = 0.010)., Conclusions: This study demonstrated the diagnostic performance of generative AIs including ChatGPT-4, Google Gemini, and LLaMA2 chatbot. ChatGPT-4 exhibited higher diagnostic accuracy than the other platforms. These findings suggest the importance of understanding the differences in diagnostic performance among generative AIs, especially in complex case series across comprehensive medical fields, like general medicine., Competing Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article., (© The Author(s) 2024.)

Objective General fatigue is one of the most frequent chief complaints in primary care, and an accurate assessment of fatigue has a direct impact on a patient's quality of life and treatment decisions. The Fatigue Assessment Scale (FAS), a measure of general fatigue, is useful for assessing fatigue in diverse cultures and diseases. However, there has been no study showing the reliability and validity of the scale in the Japanese context. The present study assessed the reliability and validity of the Japanese version of the FAS. Methods This study was conducted on 649 patients with long COVID who had a high frequency of general fatigue. To test the structural validity of the FAS, the patients were randomly divided into two groups: one in which an exploratory factor analysis (EFA) was conducted and one in which a confirmatory factor analysis (CFA) was conducted. Cronbach's alpha was calculated to assess internal consistency reliability. Results As 58 patients had missing values, we analyzed the data of 591 patients. The EFA led to an FAS comprising two factors. The CFA showed an acceptable fit for this two-factor model. The internal consistency was found to be good (Cronbach's alpha =0.89). Conclusion This study verified the structural validity and internal consistency and reliability of the Japanese version of the FAS. The results indicate that the Japanese version of the FAS is useful for assessing general fatigue in patients with long COVID in Japan.

Purpose: The present study aimed to clarify the effects of a hyperglycemic condition on the clinical consequences of long COVID. Methods: Among 643 patients who visited the outpatient clinic of our hospital from February 2021 to September 2023, long COVID patients were classified into a hyperglycemic (HG) group with casual blood glucose levels above 140 mg/dL and a normoglycemic (NG) group. The patients' backgrounds, clinical symptoms, health status including the QOL evaluation scale (EQ-5D-5L), self-rating depression scale (SDS), and F-scale questionnaire (FSSG), blood test data, and recovery periods were analyzed. Results: The NG group included 607 patients with long COVID and the HG group included 36 patients with long COVID. Patients in the HG group were older than those in the NG group (55 vs. 41 years; p < 0.001) and included a larger percentage of males (67% vs. 44%; p = 0.009). The HG group had a larger percentage of patients with moderate-to-severe conditions in the acute infection phase (28% vs. 12%; p = 0.008), a higher BMI (25 vs. 22 kg/m 2 ; p < 0.001), higher blood pressure (138/81 vs. 122/72 mmHg; p < 0.001), and a larger percentage of patients with an alcohol drinking habit (53% vs. 34%; p = 0.031). Long COVID symptoms and self-rated scales were not differed between the two groups; however, the laboratory data showed that liver and renal functions and metabolic data were significantly worse in the HG group. Although there was no apparent difference between the two groups in duration from the infection to the first visit, the HG group had a significantly longer period of recovery from long COVID (median period of 421 vs. 294 days; p = 0.019). Conclusion: A hyperglycemic state associated with other lifestyle-related diseases is associated with the prolongation of recovery from long COVID.

Purpose: To elucidate the impact of long COVID on menstruation and mental health, medical records of patients with long COVID were evaluated., Methods: Symptoms of long COVID, QOL, mental health, and related endocrine data were compared between two groups with and without menstrual disturbances., Results: Of 349 female patients who visited our clinic between February 2021 and March 2023, 223 patients with long COVID (aged 18-50 years) were included. Forty-four (19.7%) of the patients had menstrual symptoms associated with long COVID. The patients with menstrual symptoms were older than those without menstrual symptoms (42.5 vs. 38 years). The percentage of patients with menstrual symptoms was higher during the Omicron phase (24%) than during the Preceding (13%) and Delta (12%) phases. Cycle irregularity was the most frequent (in 63.6% of the patients), followed by severe pain (25%), heavy bleeding (20.5%), perimenopausal symptoms (18.2%), and premenstrual syndrome (15.9%). Fatigue and depression were the most frequent complications. Scores for fatigue and for QOL were significantly worse in long COVID patients with menstrual symptoms. Results of endocrine examinations showed significantly increased cortisol levels in patients with menstrual complaints., Conclusion: Long COVID has an impact on menstrual conditions and on QOL related to menstrual conditions.

Purpose : The present study aimed to uncover the impact of long COVID on the working situations of Japanese patients. Methods : Changes in the working situations of the patients who visited our long COVID clinic were evaluated from medical records for the aspects of physical status, quality of life (QOL), and mental conditions. Results : Of 846 long COVID patients who visited our clinic from February 2021 to December 2023, 545 employed patients aged between 18 and 65 years were included in this study. A total of 295 patients (54.1%) with long COVID (median age: 43 years, female: 55.6%) experienced changes in their working status. Those patients included 220 patients (40.4%) who took a leave of absence, 53 patients (9.7%) who retired, and 22 patients (4%) with reduced working hours. Most of the patients (93.2%) with changes in working conditions had mild disease severity in the acute phase of COVID-19. The majority of those patients with mild disease severity (58.8%) were infected in the Omicron-variant phase and included 65.3% of the female patients. The major symptoms in long COVID patients who had changes in their working situations were fatigue, insomnia, headache, and dyspnea. Scores indicating fatigue and QOL were worsened in long COVID patients who had changes in their working situations. In addition, 63.7% of the long COVID patients with changes in their working situations had decreases in their incomes. Conclusions: Changes in the working situation of long COVID patients who were employed had a negative impact on the maintenance of their QOL.

Background: The potential of artificial intelligence (AI) chatbots, particularly ChatGPT with GPT-4 (OpenAI), in assisting with medical diagnosis is an emerging research area. However, it is not yet clear how well AI chatbots can evaluate whether the final diagnosis is included in differential diagnosis lists., Objective: This study aims to assess the capability of GPT-4 in identifying the final diagnosis from differential-diagnosis lists and to compare its performance with that of physicians for case report series., Methods: We used a database of differential-diagnosis lists from case reports in the American Journal of Case Reports, corresponding to final diagnoses. These lists were generated by 3 AI systems: GPT-4, Google Bard (currently Google Gemini), and Large Language Models by Meta AI 2 (LLaMA2). The primary outcome was focused on whether GPT-4's evaluations identified the final diagnosis within these lists. None of these AIs received additional medical training or reinforcement. For comparison, 2 independent physicians also evaluated the lists, with any inconsistencies resolved by another physician., Results: The 3 AIs generated a total of 1176 differential diagnosis lists from 392 case descriptions. GPT-4's evaluations concurred with those of the physicians in 966 out of 1176 lists (82.1%). The Cohen κ coefficient was 0.63 (95% CI 0.56-0.69), indicating a fair to good agreement between GPT-4 and the physicians' evaluations., Conclusions: GPT-4 demonstrated a fair to good agreement in identifying the final diagnosis from differential-diagnosis lists, comparable to physicians for case report series. Its ability to compare differential diagnosis lists with final diagnoses suggests its potential to aid clinical decision-making support through diagnostic feedback. While GPT-4 showed a fair to good agreement for evaluation, its application in real-world scenarios and further validation in diverse clinical environments are essential to fully understand its utility in the diagnostic process., (©Takanobu Hirosawa, Yukinori Harada, Kazuya Mizuta, Tetsu Sakamoto, Kazuki Tokumasu, Taro Shimizu. Originally published in JMIR Formative Research (https://formative.jmir.org), 26.06.2024.)

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, χ 2 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., (©Takanobu Hirosawa, Yukinori Harada, Kazuki Tokumasu, Takahiro Ito, Tomoharu Suzuki, Taro Shimizu. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 09.04.2024.)

Determination of long COVID requires ruling out alternative diagnoses, but there has been no report on the features of alternative diagnoses. This study was a single-center retrospective study of outpatients who visited our clinic between February 2021 and June 2023 that was carried out to determine the characteristics of alternative diagnoses in patients with post-COVID-19 symptoms. In a total of 731 patients, 50 patients (6.8%) were newly diagnosed with 52 diseases requiring medical intervention, and 16 (32%) of those 50 patients (2.2% of the total) were considered to have priority for treatment of the newly diagnosed disorders over long COVID treatment. The proportion of patients with a new diagnosis increased with advance of age, with 15.7% of the patients aged 60 years or older having a new diagnosis. Endocrine and metabolic diseases and hematological and respiratory diseases were the most common, being detected in eight patients (16%) each. Although 35 of the 52 diseases (67%) were related to their symptoms, endocrine and metabolic diseases were the least associated with specific symptoms. Other disorders that require attention were found especially in elderly patients with symptomatic long COVID. Thus, appropriate assessment and differentiation from alternative diagnoses are necessary for managing long COVID., (© 2024. The Author(s).)

Introduction: This study aimed to elucidate the prevalence and clinical characteristics of patients with long COVID (coronavirus disease 2019), especially focusing on 50% hemolytic complement activity (CH50)., Methods: This retrospective observational study focused on patients who visited Okayama University Hospital (Japan) for the treatment of long COVID between February 2021 and March 2023. CH50 levels were measured using liposome immunometric assay (Autokit CH50 Assay, FUJIFILM Wako Pure Chemical Corporation, Japan); high CH50 was defined as ≥59 U/mL. Univariate analyses assessed differences in the clinical background, long COVID symptoms, inflammatory markers, and clinical scores of patients with normal and high CH50. Logistic regression model investigated the association between high CH50 levels and these factors., Results: Of 659 patients who visited our hospital, 478 patients were included. Of these, 284 (59.4%) patients had high CH50 levels. Poor concentration was significantly more frequent in the high CH50 group (7.2% vs. 13.7%), whereas no differences were observed in other subjective symptoms (fatigue, headache, insomnia, dyspnea, tiredness, and brain fog). Multivariate analysis was performed on factors that could be associated with poor concentration, suggesting a significant relationship to high CH50 levels (adjusted odds ratio [aOR], 2.70; 95% confidence interval [CI], 1.33-5.49). Also, high CH50 was significantly associated with brain fog (aOR, 1.66; 95% CI, 1.04-2.66)., Conclusions: High CH50 levels were frequently reported in individuals with long COVID, indicating a relationship with brain fog. Future in-depth research should examine the pathological role and causal link between complement immunity and the development of long COVID., Competing Interests: Declaration of competing interest The authors declare no conflict of interest., (Copyright © 2023 Japanese Society of Chemotherapy, Japanese Association for Infectious Diseases, and Japanese Society for Infection Prevention and Control. Published by Elsevier Ltd. All rights reserved.)

Background: The utility of a clinical decision support system using a machine learning (ML) model for simultaneous cardiac and pulmonary auscultation is unknown., Objective: This study aimed to develop and evaluate an ML system's utility for cardiopulmonary auscultation., Methods: First, we developed an ML system for cardiopulmonary auscultation, using cardiopulmonary sound files from our previous study. The technique involved pre-processing, feature extraction, and classification through several neural network layers. After integration, the output class was categorized as "normal," "abnormal," or "undetermined." Second, we evaluated the ML system with 24 junior residents in an open-label randomized controlled trial at a university hospital. Participants were randomly assigned to the ML system group (intervention) or conventional auscultation group (control). During training, participants listened to four cardiac and four pulmonary sounds, all of which were correctly classified. Then, participants classified a series of 16 simultaneous cardiopulmonary sounds. The control group auscultated the sounds using noise-cancelling headphones, while the intervention group did so by watching recommendations from the ML system., Results: The total scores for correctly identified normal or abnormal cardiopulmonary sounds in the intervention group were significantly higher than those in the control group (366/384 [95.3%] vs. 343/384 [89.3%], P  = 0.003). The cardiac test score in the intervention group was better (111/192 [57.8%] vs. 90/192 [46.9%], P  = 0.04); there was no significant difference in pulmonary auscultation., Conclusions: The ML-based system improved the accuracy of cardiopulmonary auscultation for junior residents. This result suggests that the system can assist early-career physicians in accurate screening., Competing Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article., (© The Author(s) 2024.)

Background: Functional hyperthermia (FH) is characterized by hyperthermia resulting from sympathetic hyperactivity rather than inflammation, and it is frequently overlooked by medical practitioners due to the absence of abnormalities in a medical examination. Although FH is an important differential diagnosis for fever of unknown origin (FUO), the literature on FUO cases in Japan lacks information on FH. In this study, we aimed to uncover the population of FH patients hidden in FUO cases. Methods: An outpatient clinic for FUO was established at Okayama University Hospital, and 132 patients were examined during the period from May 2019 to February 2022. Results: A diagnosis of FH was made in 31.1% of the FUO cases, and FH predominantly affected individuals in their third and fourth decades of life with a higher incidence in females (68.3%). The frequency of a history of psychiatric illness was higher in patients with FH than in patients with other febrile illnesses. Although the C-reactive protein (CRP) is generally negative in FH cases, some obese patients, with a body mass index ≥ 25 had slightly elevated levels of CRP but were diagnosed with FH. Conclusions: The results showed the importance of identifying FH when encountering patients with FUO without any organic etiology.

Background: Although many adverse reactions after SARS-CoV-2 vaccination have been reported, there have been few comprehensive studies on persistent symptoms after SARS-CoV-2 vaccination. The aim of this study was to determine the clinical characteristics of patients with various persistent symptoms after SARS-CoV-2 vaccination. Methods: A retrospective descriptive study was performed for patients who visited a specialized clinic established at Okayama University Hospital to evaluate adverse events after SARS-CoV-2 vaccination during the period from April 2021 to March 2023. Results: Descriptive analysis was performed for 121 of 127 patients who visited the clinic during the study period, and separate analysis was performed for the other 6 patients who had serious complications, who required treatment with prednisolone, and who had persistent symptoms. The median [interquartile range] age of the patients was 48 years [31-64 years], and the patients included 44 males (36.4%) and 77 females (63.6%). The most frequent symptoms were sensory impairment (34 patients, 28.1%), general fatigue (30 patients, 24.8%), fever/low-grade fever (21 patients, 17.4%), and headache (21 patients, 17.4%). Serious complications included myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), sarcoidosis, aseptic meningitis, neuromyelitis optica spectrum disorders (NMOSDs), tendon adhesions, and idiopathic thrombocytopenia. Conclusions: Although causal relationships were not determined, 15 persistent symptoms after SARS-CoV-2 vaccination were characterized. All of the symptoms had onset from 12 hours to one week after vaccination, with 10 symptoms persisting for 6 months or longer. The most frequent symptom was sensory impairment.

Background: The diagnostic accuracy of differential diagnoses generated by artificial intelligence chatbots, including ChatGPT models, for complex clinical vignettes derived from general internal medicine (GIM) department case reports is unknown., Objective: This study aims to evaluate the accuracy of the differential diagnosis lists generated by both third-generation ChatGPT (ChatGPT-3.5) and fourth-generation ChatGPT (ChatGPT-4) by using case vignettes from case reports published by the Department of GIM of Dokkyo Medical University Hospital, Japan., Methods: We searched PubMed for case reports. Upon identification, physicians selected diagnostic cases, determined the final diagnosis, and displayed them into clinical vignettes. Physicians typed the determined text with the clinical vignettes in the ChatGPT-3.5 and ChatGPT-4 prompts to generate the top 10 differential diagnoses. The ChatGPT models were not specially trained or further reinforced for this task. Three GIM physicians from other medical institutions created differential diagnosis lists by reading the same clinical vignettes. We measured the rate of correct diagnosis within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and the top diagnosis., Results: In total, 52 case reports were analyzed. The rates of correct diagnosis by ChatGPT-4 within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and top diagnosis were 83% (43/52), 81% (42/52), and 60% (31/52), respectively. The rates of correct diagnosis by ChatGPT-3.5 within the top 10 differential diagnosis lists, top 5 differential diagnosis lists, and top diagnosis were 73% (38/52), 65% (34/52), and 42% (22/52), respectively. The rates of correct diagnosis by ChatGPT-4 were comparable to those by physicians within the top 10 (43/52, 83% vs 39/52, 75%, respectively; P=.47) and within the top 5 (42/52, 81% vs 35/52, 67%, respectively; P=.18) differential diagnosis lists and top diagnosis (31/52, 60% vs 26/52, 50%, respectively; P=.43) although the difference was not significant. The ChatGPT models' diagnostic accuracy did not significantly vary based on open access status or the publication date (before 2011 vs 2022)., Conclusions: This study demonstrates the potential diagnostic accuracy of differential diagnosis lists generated using ChatGPT-3.5 and ChatGPT-4 for complex clinical vignettes from case reports published by the GIM department. The rate of correct diagnoses within the top 10 and top 5 differential diagnosis lists generated by ChatGPT-4 exceeds 80%. Although derived from a limited data set of case reports from a single department, our findings highlight the potential utility of ChatGPT-4 as a supplementary tool for physicians, particularly for those affiliated with the GIM department. Further investigations should explore the diagnostic accuracy of ChatGPT by using distinct case materials beyond its training data. Such efforts will provide a comprehensive insight into the role of artificial intelligence in enhancing clinical decision-making., (©Takanobu Hirosawa, Ren Kawamura, Yukinori Harada, Kazuya Mizuta, Kazuki Tokumasu, Yuki Kaji, Tomoharu Suzuki, Taro Shimizu. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 09.10.2023.)

Objective: The most common symptom of post-acute coronavirus disease 2019 (COVID-19) is fatigue, and it potentially leads to myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS); however, a specific prognosticator is lacking. We aimed to elucidate the clinical characteristics of patients who developed ME/CFS after COVID-19., Methods: In this retrospective observational study, patients who visited Okayama University Hospital for long COVID between February 2021 and March 2022 were investigated., Results: Of the 234 patients, 139 (59.4%) had fatigue symptoms. Fifty patients with fatigue symptoms (21.4%) met the criteria for ME/CFS (ME/CFS group), while the other 89 patients did not (non-ME/CFS group); 95 patients had no fatigue complaints (no-fatigue group). Although the patients' backgrounds were not significantly different between the three groups, the ME/CFS group presented the highest scores on the self-rating symptom scales, including the Fatigue Assessment Scale (FAS), EuroQol, and the Self-Rating Depression Scale (SDS). Furthermore, serum ferritin levels, which were correlated with FAS and SDS scores, were significantly higher in the ME/CFS group (193.0 μg/L, interquartile range (IQR): 58.8-353.8) than in the non-ME/CFS group (98.2 μg/L, 40.4-251.5) and no-fatigue group (86.7 μg/L, 37.5-209.0), and a high serum ferritin level was prominent in female patients. Endocrine workup further showed that the ME/CFS group had higher thyrotropin levels but lower growth hormone levels in serum and that insulin-like growth factor-I levels were inversely correlated with ferritin levels (R = -0.328, p < 0.05)., Conclusions: Serum ferritin level is a possible predictor of the development of ME/CFS related to long COVID, especially in female patients.