Purpose The purpose of this study was to evaluate the changes in capabilities between the Generative Pre-trained Transformer (GPT)-3.5 and GPT-4 versions of the large-scale language model ChatGPT within a Japanese medical context. Methods The study involved ChatGPT versions 3.5 and 4 responding to questions from the 112th Japanese National Nursing Examination (JNNE). The study comprised three analyses: correct answer rate and score rate calculations, comparisons between GPT-3.5 and GPT-4, and comparisons of correct answer rates for conversation questions. Results ChatGPT versions 3.5 and 4 responded to 237 out of 238 Japanese questions from the 112th JNNE. While GPT-3.5 achieved an overall accuracy rate of 59.9%, failing to meet the passing standards in compulsory and general/scenario-based questions, scoring 58.0% and 58.3%, respectively, GPT-4 had an accuracy rate of 79.7%, satisfying the passing standards by scoring 90.0% and 77.7%, respectively. For each problem type, GPT-4 showed a higher accuracy rate than GPT-3.5. Specifically, the accuracy rates for compulsory questions improved from 58.0% with GPT-3.5 to 90.0% with GPT-4. For general questions, the rates went from 64.6% with GPT-3.5 to 75.6% with GPT-4. In scenario-based questions, the accuracy rates improved substantially from 51.7% with GPT-3.5 to 80.0% with GPT-4. For conversation questions, GPT-3.5 had an accuracy rate of 73.3% and GPT-4 had an accuracy rate of 93.3%. Conclusions The GPT-4 version of ChatGPT displayed performance sufficient to pass the JNNE, significantly improving from GPT-3.5. This suggests specialized medical training could make such models beneficial in Japanese clinical settings, aiding decision-making. However, user awareness and training are crucial, given potential inaccuracies in ChatGPT's responses. Hence, responsible usage with an understanding of its capabilities and limitations is vital to best support healthcare professionals and patients.

Cervical cancer is prevalent among women, with a reported 604,127 cases in 2020 worldwide. The incidence of cervical cancer has been mitigated in most high-income countries by promoting the human papilloma virus (HPV) vaccine. However, in Japan, cervical cancer is still a leading cause of mortality and the most prevalent cancer among women aged between 15 and 39. This can be attributed to the 7-year suspension of HPV vaccination recommendations by the Japanese government. A decline in vaccination coverage followed this suspension, caused by a small number of reported adverse events, resulting in a steep decline in vaccination coverage from over 70% to less than 1%. However, there have been indications of a change in trend in Japan. In 2020, a group of volunteer doctors initiated awareness-raising activities through social networking services and other platforms, and the target population that received at least one dose of the vaccine in 2020 increased to 15.9%. Additionally, in July 2020, the Japanese government approved the updated 9-valent HPV vaccine and resumed recommendations in November 2021. As a result, 30.1% of those eligible for routine HPV vaccination received at least one dose of the vaccine from April to September, 2022. However, the HPV vaccine coverage in Japan is still far from the 90% recommended by the World Health Organization, and continued communication and education on the vaccine’s benefits are necessary to achieve optimal coverage.

Artificial intelligence (AI), particularly ChatGPT, developed by OpenAI (San Francisco, CA, USA), is making significant strides in the medical field. In a simulated case study, a 66-year-old Japanese female patient's dialogue with a physician was transcribed and inputted into ChatGPT to assess its efficacy in drafting medical records, formulating differential diagnoses, and establishing treatment plans. The results showed a high similarity between the medical summaries generated by ChatGPT and those of the attending physician. This suggests that ChatGPT has the potential to assist physicians in clinical reasoning and reduce the administrative burden, allowing them to spend more time with patients. However, there are limitations, such as the system's reliance on linguistic data and occasional inaccuracies. Despite its potential, the ethical implications of using patient data and the risk of AI replacing clinicians emphasize the need for continuous evaluation, rigorous oversight, and the establishment of comprehensive guidelines. As AI continues to integrate into healthcare, it is crucial for physicians to ensure that technology complements, rather than replaces, human expertise, with the primary focus remaining on delivering high-quality patient care.

The coronavirus disease 2019 (COVID-19) pandemic impacted people’s mental health negatively worldwide, including in non-WEIRD (Western, Educated, Industrialised, Rich and Democratic) countries. Self-compassion, kindness and understanding towards oneself in difficult times have received increasing attention in the field of mental health. Self-compassion is strongly associated with good mental health in various populations. This narrative review aimed to synthesise the evidence on self-compassion and mental health in non-WEIRD countries during the COVID-19 pandemic. MEDLINE and PsycINFO were searched for empirical studies. Self-compassion was consistently associated with positive mental health in non-WEIRD countries too. However, how, and to what degree, each component of self-compassion impacts mental health remains to be evaluated across different cultures. Future research such as multi-national intervention studies, or component network meta-analysis, is needed to advance our understanding of how self-compassion improves mental health in different populations.

ChatGPT is gaining widespread acceptance for its ability to generate natural language sentences in response to various inputs and is expected to become a supplementary tool for diagnosing and determining treatment policies in clinical settings. ChatGPT was used to evaluate its ability to perform clinical inference and accuracy in answering questions on the 117th Japanese National Medical Licensing Examination held in February 2023. The exam questions were manually inputted into ChatGPT's window, and the accuracy of ChatGPT's responses was determined based on answers provided by a preparatory school. ChatGPT provided answers for 389 out of 400 questions, and its overall correct answer rate was 55.0%. The correct answer rate for 5-choice-1, 5-choice-2, and 5-choice-3 were 57.8%, 42.9%, and 41.2%, respectively. The highest correct answer rate was for the compulsory exam (67.0%), followed by the specific knowledge exam (54.1%) and the cross category exam (47.9%). The correct answer rate for non-image questions was 56.2% and for image questions, it was 51.5%. The study suggests that ChatGPT has potential to support healthcare professionals in clinical decision-making in Japanese clinical settings, but caution should be exercised in interpreting and using the answers generated by ChatGPT due to room for improvement in performance.

Japan's conditional/time‐limited early approval program, initiated in 2014, aimed to advance regenerative medicine by expediting market access. However, the withdrawal of autologous skeletal myoblast sheets (Heartsheet) due to ineffectiveness raises concerns about the balance between rapid approval and scientific integrity. While the program seeks to boost innovation, it risks endorsing costly, unclear treatments under national health care. This case highlights the need to refine regulatory approaches, ensuring clinical efficacy and fiscal responsibility in regenerative therapies.