JMA journal最新文献

The advantages and disadvantages of the use of generative artificial intelligence, such as ChatGPT, in medical writing have been widely discussed; however, two concerns remain largely unexplored. The first involves "human touch," such as personal anecdotes and experiences. This touch often distinguishes human-written papers from those generated by ChatGPT as ChatGPT cannot independently access personal experiences. Although ChatGPT may mimic humanlike behavior, including the incorporation of a human touch, it lacks genuine emotions. With the lack of established guidelines on the acceptable levels of ChatGPT use and imperfect detection tools, many authors fear that their work could be perceived as overly reliant on ChatGPT. I worry that writers may artificially insert forced personal touches simply to assert their own writing. The second concern is the authors' worry and doubt about whether to use ChatGPT and, if so, to what extent, which may disrupt their reflective and quiet writing process. While I acknowledge the lack of empirical data, I offer practical suggestions to balance the benefits of ChatGPT assistance and the preservation of the integrity of human writing in medical publications.

Digital transformation of healthcare is rapidly progressing. Digital transformation improves the quality of services and access to health information for users, reduces the workload and associated costs for healthcare providers, and supports clinical decision-making. Data and artificial intelligence (AI) play a key role in this process. The AI used for this purpose is called medical AI. Medical AI is currently undergoing a shift from task-specific to general-purpose models. Large language models have the potential to systematize existing medical knowledge in a standardized way. The usage of AI in medicine is not limited to digital transformation; it plays a pivotal role in fundamentally changing the state of medical science. This approach, known as "AI for Medical Science," focuses on pioneering a form of medical science that predicts the onset and progression of disease based on the underlying causes of disease. The key to such predictive medicine is the concept of "states," which can be sought through machine learning. Using states instead of symptoms not only dramatically improves the accuracy of identification (diagnosis) and prediction (prognosis) but also potentially pioneers P4 medicine by integrating it with empirical knowledge and theories based on natural principles.

Introduction: Patients with Behçet's disease (BD) have a variety of symptoms, and the exacerbation of these symptoms affects their daily life and social participation and reduces their quality of life (QOL). This study aimed to clarify the relationship between social participation and QOL in BD patients.

Methods: The BD-checklist 92 and 36-item Short Form Survey (SF-36) questionnaires were mailed to 10 affiliates. A total of 174 patients with BD completed the questionnaire. The patients were divided into two groups according to the presence or absence of problems in each "participation" category of the BD-checklist 92, and the SF-36 scores were compared. Subsequently, a correlational analysis was used to examine the relationship between the number of problem categories extracted from "participation" and scores on the eight subscales of the SF-36. Multiple regression analyses were performed to identify factors associated with SF-36 scores.

Results: The SF-36 subscale scores were significantly lower in patients with problems in the participation category, particularly in those with difficulties in shopping, housework, relationships with friends and family, and community activities. A multiple regression analysis revealed that "basic interpersonal relationships" and "community life" were associated with the SF-36 subscales "role physical", "social functioning", "role emotional", and "mental health".

Conclusions: This study showed that despite excluding the effects of BD-specific primary and secondary symptoms, problems with basic interpersonal relationships, such as those with friends and family, and restricted community activities were associated with reduced QOL.

Introduction: This study examined the association of trunk function evaluated using Functional Assessment for Control of Trunk (FACT) with independent walking. It aimed to determine the effectiveness of the FACT cutoff score in predicting independent walking at hospital discharge.

Methods: This retrospective observational study included patients with cerebral infarction. The patients were categorized into the independent (Functional Independence Measure [FIM] locomotion walking score of the patient was ≥6; n = 102) and dependent (≤5; n = 111) groups based on the FIM locomotion scale at discharge. Multivariate logistic regression analysis was employed to determine the significant independent variables on admission for predicting independent walking at discharge. Furthermore, the receiver operating characteristic was used to calculate the cutoff value for admission status.

Results: A total of 213 patients (122 men and 91 women) were included in this study. The independent group had higher scores in FACT (15.0 [12.0-20.0] vs. 6.0 [2.0-12.0], P < 0.001) on admission than the dependent group. The results of the multivariate logistic regression analysis indicated that the factors associated with independent walking were the FACT and Mini-Mental State Examination-Japanese (MMSE-J) on admission. The optimal cutoff score for the FACT on admission was 8, and the area under the curve for the FACT scores on admission when discriminating between independent walking at discharge was 0.82.

Conclusions: The results of this study can facilitate the optimization of patient rehabilitation as early as possible. The effects of improved trunk function require further validation through prospective observational studies.

Recent advancements in artificial intelligence (AI) have markedly affected various fields, with notable progress in surgery. This study explores the integration of AI in surgery, particularly focusing on minimally invasive surgery (MIS), where high-quality surgical videos provide fertile ground for computer vision (CV) technology applications. CV plays an important role in enhancing intraoperative decision-making through real-time image recognition. This study considers the challenges in clinical applications and future perspectives by reviewing the current state of AI in navigation during surgery, postoperative analysis, and automated surgical skill assessment.

Introduction: The prognosis for activities of daily living (ADL) ability after stroke is negatively influenced by undernutrition and impaired balance. However, the association between undernutrition and balance improvement has not yet been elucidated. This study aimed to investigate the influence of undernutrition on balance function improvement in patients with stroke.

Methods: This retrospective observational study included patients with cerebral infarction aged ≥65 years. The study period was from May 2018 to May 2022. The patients were divided into undernutrition and intact nutrition groups according to the Global Leadership Initiative on Malnutrition criteria. The primary outcome was the change in the Berg Balance Scale (BBS) score (BBS score at discharge - BBS score at admission).

Results: This study included 304 patients (mean age, 79.2 ± 8.1 years; 173 men and 131 women). These patients were divided into the undernutrition (N = 114) and intact nutrition (N = 190) groups. The undernutrition group demonstrated lower BBS scores at admission (16.0 ± 17.1 vs. 28.3 ± 18.4, p < 0.001) and at discharge (24.2 ± 19.6 vs. 40.0 ± 16.9, p < 0.001) than the intact nutrition group. After adjusting for confounding factors, undernutrition was associated with a smaller change in the BBS score (coefficient = -2.988, 95% confidence interval = -5.481 to -0.495, p = 0.019).

Conclusions: Undernutrition negatively influences balance function recovery in post-stroke patients. A strategy aimed at improving nutritional status could have beneficial effects on patients' balance function.

Ophthalmology is well suited for the integration of artificial intelligence (AI) owing to its reliance on various imaging modalities, such as anterior segment photography, fundus photography, and optical coherence tomography (OCT), which generate large volumes of high-resolution digital images. These images provide rich datasets for training AI algorithms, which enables precise diagnosis and monitoring of various ocular conditions. Retinal disease management heavily relies on image recognition. Limited access to ophthalmologists in underdeveloped areas and high image volumes in developed countries make AI a promising, cost-effective solution for screening and diagnosis. In corneal diseases, differential diagnosis is critical yet challenging because of the wide range of potential etiologies. AI and diagnostic technologies offer promise for improving the accuracy and speed of these diagnoses, including the differentiation between infectious and noninfectious conditions. Smartphone imaging coupled with AI technology can advance the diagnosis of anterior segment diseases, democratizing access to eye care and providing rapid and reliable diagnostic results. Other potential areas for AI applications include cataract and vitreous surgeries as well as the use of generative AI in training ophthalmologists. While AI offers substantial benefits, challenges remain, including the need for high-quality images, accurate manual annotations, patient heterogeneity considerations, and the "black-box phenomenon". Addressing these issues is crucial for enhancing the effectiveness of AI and ensuring its successful integration into clinical practice. AI is poised to transform ophthalmology by increasing diagnostic accuracy, optimizing treatment strategies, and improving patient care, particularly in high-risk or underserved populations.