Large Language Models (LLMs), such as ChatGPT (OpenAI, CA, US), have revolutionized scientific writing and research processes across academic disciplines, providing comprehensive support throughout the entire research lifecycle.Generative artificial intelligence (GAI) tools enhance every aspect of scientific writing, from hypothesis genera‑ tion and methodology design to data analysis and manuscript preparation. This review examines the applications of LLMs in hematological research, with particular emphasis on advanced techniques, including prompt engineering and retrieval augmented generation (RAG) frameworks. Prompt engineering methods, including zero-shot and fewshot learning along with a chain-of-thought approach, enable researchers to generate more precise context-specific content, especially in scientific writing. Integrating RAG frameworks with the current medical literature and clinical guidelines significantly reduces the risk of misinformation while ensuring alignment with contemporary medical standards. Even though these GAI tools offer remarkable potential for streamlining research writing and enhancing documentation quality, the study also addresses the critical importance of maintaining scientific integrity, ethical considerations, and privacy concerns in hematological research.

Large Language Models (LLMs), such as ChatGPT (OpenAI, CA, US), have revolutionized scientific writing and research processes across academic disciplines, providing comprehensive support throughout the entire research lifecycle.Generative artificial intelligence (GAI) tools enhance every aspect of scientific writing, from hypothesis genera‑ tion and methodology design to data analysis and manuscript preparation. This review examines the applications of LLMs in hematological research, with particular emphasis on advanced techniques, including prompt engineering and retrieval augmented generation (RAG) frameworks. Prompt engineering methods, including zero-shot and fewshot learning along with a chain-of-thought approach, enable researchers to generate more precise context-specific content, especially in scientific writing. Integrating RAG frameworks with the current medical literature and clinical guidelines significantly reduces the risk of misinformation while ensuring alignment with contemporary medical standards. Even though these GAI tools offer remarkable potential for streamlining research writing and enhancing documentation quality, the study also addresses the critical importance of maintaining scientific integrity, ethical considerations, and privacy concerns in hematological research.

Large Language Models (LLMs), such as ChatGPT (OpenAI, CA, US), have revolutionized scientific writing and research processes across academic disciplines, providing comprehensive support throughout the entire research lifecycle.Generative artificial intelligence (GAI) tools enhance every aspect of scientific writing, from hypothesis genera‑ tion and methodology design to data analysis and manuscript preparation. This review examines the applications of LLMs in hematological research, with particular emphasis on advanced techniques, including prompt engineering and retrieval augmented generation (RAG) frameworks. Prompt engineering methods, including zero-shot and fewshot learning along with a chain-of-thought approach, enable researchers to generate more precise context-specific content, especially in scientific writing. Integrating RAG frameworks with the current medical literature and clinical guidelines significantly reduces the risk of misinformation while ensuring alignment with contemporary medical standards. Even though these GAI tools offer remarkable potential for streamlining research writing and enhancing documentation quality, the study also addresses the critical importance of maintaining scientific integrity, ethical considerations, and privacy concerns in hematological research.

INTRODUCTION: Adolescent depression is prevalent, and teen suicide rates are on the rise locally. A systemic review to understand associated risk and protective factors is important to strengthen measures for the prevention and early detection of adolescent depression and suicide in Singapore. This systematic review aims to identify the factors associated with adolescent depression in Singapore. METHODS: A systematic search on the following databases was performed on 21 May 2020: PubMed, EMBASE and PsycINFO. Full texts were reviewed for eligibility, and the included studies were appraised for quality using the Newcastle Ottawa Scale. Narrative synthesis of the finalised articles was performed through thematic analysis. RESULTS: In total, eight studies were included in this review. The four factors associated with adolescent depression identified were: (1) sociodemographic factors (gender, ethnicity); (2) psychological factors, including childhood maltreatment exposure and psychological constructs (hope, optimism); (3) coexisting chronic medical conditions (asthma); and (4) lifestyle factors (sleep inadequacy, excessive internet use and pathological gaming). CONCLUSION The identified factors were largely similar to those reported in the global literature, except for sleep inadequacy along with conspicuously absent factors such as academic stress and strict parenting, which should prompt further research in these areas. Further research should focus on current and prospective interventions to improve mental health literacy, targeting sleep duration, internet use and gaming, and mitigating the risk of depression in patients with chronic disease in the primary care and community setting.
Humans ; Singapore/epidemiology* ; Adolescent ; Risk Factors ; Depression/etiology* ; Protective Factors ; Male ; Female ; Life Style ; Suicide

The aging microenvironment, as a key driver of tumorigenesis and progression, plays a critical role in tumor immune regulation through one of its core features-the senescence-associated secretory phenotype (SASP). SASP consists of a variety of interleukins, chemokines, proteases, and growth factors. It initially induces surrounding cells to enter a state of senescence through paracrine mechanisms, thereby creating a sustained inflammatory stimulus and signal amplification effect within the tissue microenvironment. Furthermore, these secreted factors activate key signaling pathways such as NF-κB, cGAS-STING, and mTOR, which regulate the expression of immune-related molecules (such as PD-L1) and promote the recruitment of immunosuppressive cells, including regulatory T cells and myeloid-derived suppressor cells. This process ultimately contributes to the formation of an immunosuppressive tumor microenvironment. Furthermore, the article explores potential anti-tumor immunotherapy strategies targeting SASP and its associated molecular mechanisms, including approaches to inhibit SASP secretion or eliminate senescent cells. Although these strategies have shown promise in certain tumor models, the high heterogeneity among tumor types may result in varied responses to SASP-targeted therapies. This highlights the need for further research into adaptive stratification and personalized treatment approaches. Targeting immune regulatory mechanisms in the aging microenvironment-particularly SASP-holds great potential for advancing future anti-tumor therapies.

Glutamate-ammonia ligase (GLUL, also known as glutamine synthetase) is a crucial enzyme that catalyzes ammonium and glutamate into glutamine in the ATP-dependent condensation. Although GLUL plays a critical role in multiple cancers, the expression and function of GLUL in gastric cancer remain unclear. In the present study, we have found that the expression level of GLUL was significantly lower in gastric cancer tissues compared with adjacent normal tissues, and correlated with N stage and TNM stage, and low GLUL expression predicted poor survival for gastric cancer patients. Knockdown of GLUL promoted the growth, migration, invasion and metastasis of gastric cancer cells in vitro and in vivo, and vice versa, which was independent of its enzyme activity. Mechanistically, GLUL competed with β-Catenin to bind to N-Cadherin, increased the stability of N-Cadherin and decreased the stability of β-Catenin by alerting their ubiquitination. Furthermore, there were lower N-Cadherin and higher β-Catenin expression levels in gastric cancer tissues compared with adjacent normal tissues. GLUL protein expression was correlated with that of N-Cadherin, and could be the independent prognostic factor in gastric cancer. Our findings reveal that GLUL stabilizes N-Cadherin by antagonizing β-Catenin to inhibit the progress of gastric cancer.

Human immunodeficiency virus (HIV) infection presents a major global health challenge. According to the World Health Organization, approximately 39.0 million people have HIV, as of 2022, with 1.3 million new infections annually. In Korea, the first HIV cases were reported in 1985, and the number of cases has increased continuously since 1988. In 2022, 1,066 cases were newly diagnosed, increasing the total number of cases to approximately 15,880.Current Concepts: Diagnostics play a crucial role in the management of HIV, for early diagnosis and initiation of antiretroviral therapy. Approval of the first HIV antibody test in 1985 by the US Food and Drug Administration marked a notable advance in the ability to diagnose HIV infection. The screening test for HIV infection typically consists of a two-step approach, in which a fourth-generation combination HIV-1/2 immunoassay is followed by a confirmatory Western blot. Despite its high accuracy for chronic infections, the window period and the early-stage of HIV infections are problematic, and populations with low HIV prevalence show a low positive predictive value. A false positive result can occur from blood transfusions, vaccinations, and autoantibodies. Novel diagnostic tools, such as nucleic acid testing for viral RNA/DNA and antigen testing for p24 antigen have reduced the window period, allowing for earlier detection and differentiation between HIV-1 and HIV-2. Pointof-care tests are beneficial in resource-limited settings although their lower sensitivity and specificity require additional confirmatory follow-up testing.Discussion and Conclusion: In this review, we provide a historical perspective on the evolution of HIV diagnostics (serological and molecular) and describe the roles of conventional and point-of-care testing within the laboratory diagnostic network.

Background/Aims: Hypopharyngeal multichannel intraluminal impedance-pH (HMII-pH) technology incorporating 2 trans-upper esophageal sphincter impedance channels has been developed to detect pharyngeal reflux. We used the HMII-pH technique to validate the candidate pharyngeal acid reflux (PAR) episodes based on the dual-pH tracings and determined the interobserver reproducibility. Methods: We conducted a cross-sectional study in tertiary centers in Taiwan. Ninety patients with suspected laryngopharyngeal reflux and 28 healthy volunteers underwent HMII-pH test when off acid suppressants. Candidate PAR episodes were characterized by pharyngeal pH drops of at least 2 units and reaching a nadir pH of 5 within 30 seconds during esophageal acidification. Two experts manually independently identified candidate PAR episodes based on the dual-pH tracings. By reviewing the HMII-pH tracings, HMII-pH-proven PAR episodes were subsequently confirmed. The consensus reviews of HMII-pH-proven PAR episodes were considered to be the reference standard diagnosis. The interobserver reproducibility was assessed. Results: A total of 105 candidate PAR episodes were identified. Among them 84 (80.0%; 95% CI, 71.0-87.0%) were HMII-pH-proven PAR episodes (82 in 16 patients and 2 in 1 healthy subject). Patients tended to have more HMII-pH-proven PAR episodes than healthy controls (median and percentile values [25th, 75th, and 95th percentiles]: 0 [0, 0, 3] vs 0 [0, 0, 0], P = 0.067). The concordance rate in diagnosing HMII-pH-proven PAR episodes between 2 independent observers was 92.2%. Conclusion Our preliminary data showed that 80.0% (71.0-87.0%) of the proposed candidate PAR episodes were HMII-pH-proven PAR episodes, among which the interobserver reproducibility was good.

Purpose: Medical schools have faced various challenges in preparing their clinical students for the frontlines of a pandemic. This study investigated medical students’ satisfaction with their institutions during the coronavirus disease 2019 (COVID-19) pandemic with the intention of guiding educators in future public health crises. Methods: In this cross-sectional study surveying students in clinical rotations, the primary outcome was overall satisfaction regarding medical schools’ responses to the pandemic, and the four secondary outcomes were school communication, exposure to COVID-19, availability of personal protective equipment, and access to COVID-19 testing. Results: The survey was distributed to ten medical schools, of which 430 students responded for a response rate of 13.0%. While most students were satisfied (61.9%, n=266) with their schools’ response, more than one in five (21.9%, n=94) were dissatisfied. Among the four secondary outcomes, communication with students was most predictive of overall satisfaction. Conclusion In future crises, schools can best improve student satisfaction by prioritizing timely communication.

COVID-19/epidemiology* ; China/epidemiology* ; Humans ; Public Facilities ; Tobacco Smoke Pollution ; Workplace