OBJECTIVE: To develop a user-friendly visualization application for the automatic annotation of Human Phenotype Ontology (HPO) terms based on large language models and retrieval-augmented generation (RAG) technology, and to validate its performance in an authoritative case dataset. METHODS: By integrating the domestic open-source large language model DeepSeek-V3 with RAG technology, an interactive web application was deployed on the Streamlit cloud platform. Using only the latest official HPO dataset as the data source, the lightweight sentence-embedding model BAAI/bge-small-en-v1.5 was employed to construct a FAISS vector index. During the online phase, a four-step closed-loop process is automatically completed: multilingual translation, phenotype phrase extraction, RAG candidate retrieval, term mapping, and official database validation. 121 English case reports publicly released by BMJ Case Reports and Oxford Medical Case Reports (with a gold-standard HPO set of 1 794 terms) were selected for application validation. Precision, recall, and F1 score were calculated and compared horizontally with traditional dictionary tools, standalone large language models, and the similar application "RAG-HPO". Finally, replace the model with the more advanced ChatGPT-5 and evaluate its performance on the newly extracted dataset. RESULTS: An HPO term automatic annotation visualization application named PhenoRAG, based on large language models and RAG technology, was successfully developed. Users can access it directly via a web link. Across the 112 cases, a total of 2 150 HPO terms were generated; 2,064 (96.0%) were fully validated by the official database, with a hallucination rate of 1.3% and an HPO ID-name mismatch rate of 2.7%. After deduplication, 1,906 terms remained for testing. The overall precision was 63.65%, recall was 67.34%, and F1 was 65.44%, significantly outperforming traditional annotation tools (F1: 0.45-0.49, P < 0.001). Although PhenoRAG's F1 was lower than that of RAG-HPO (F1 = 0.78, P < 0.001), which relies on a manually constructed synonym database of 54 000 entries plus the HPO dataset, it requires no additional dictionary maintenance and can be used without any background in computer programming. Moreover, after switching to the GPT-5 model, PhenoRAG exhibited no hallucination rate on the new dataset, and its F1 score significantly increased (P = 0.038). CONCLUSION Without constructing a synonym database, the PhenoRAG achieved high-accuracy automatic mapping from clinical text to standard HPO terms. It features a low usage threshold, free access, and a Chinese-language interface, and can directly serve rare disease diagnosis, genetic counseling, and research scenarios in China and worldwide, warranting further clinical promotion and multicenter validation.
Humans ; Phenotype ; Biological Ontologies ; Language ; Software ; Large Language Models

OBJECTIVE: To explore the efficacy of a Thalassemia risk assessment software for the screening of thalassemia mutation carriers and distribution of thalassemia genotypes detected by screening. METHODS: A total of 6 040 individuals were evaluated at Leshan Maternal and Child Health Care Hospital between 2022 and 2024 using the commonly used clinical thalassemia risk assessment method and the thalassemia screening software, respectively, and the performance indicators of the two methods were compared and analyzed against the result of thalassemia gene testing. This study was approved by the Ethics Committee of our hospital (Ethics No.: LfyLL[2022]005). RESULTS: The high-risk rate by the thalassemia screening software was 11.19%, with a sensitivity of 95.12%, specificity of 93.28%, positive predictive value of 43.20%, negative predictive value of 99.72%, and the area under the ROC curve (AUC) was 0.942. The thalassemia gene detection rate of the high-risk samples screened was 4.83%. The high-risk screening rate of the conventional method was 2.50%, with a sensitivity of 51.22%, specificity of 93.28%, positive predictive value of 80.79%, negative predictive value of 97.40%, and the AUC was 0.754. The thalassemia gene detection rate of the high-risk samples was 2.02%. CONCLUSION The software can effectively detect thalassemia carriers and significantly reduce the missed detection compared with conventional method, thereby significantly improve the efficacy of screening.
Humans ; Thalassemia/diagnosis* ; Software ; Female ; Genetic Testing/methods* ; Male ; Mutation ; Adult ; Genotype ; ROC Curve ; Risk Assessment

OBJECTIVES

The increasing prevalence of Kawasaki disease in Southeast Asia (SEA) and its potential relation with Coronavirus Disease 2019 (COVID-19) infection resulted in heightened interest in KD in the region, thus, this paper aimed to determine the trend and the socioeconomic facilitators of scientific productivity of KD research within the region. Specifically, this article determined the number of publication and citations related to KD per country, institution, and journal. We also explored the networks of countries within the region to the rest of the world and the keywords mostly associated with KD research in the region. Lastly, correlation of these bibliometric indices with socioeconomic factors in the region was analyzed.

METHODS

A literature search of KD papers in SEA was performed using Scopus database. We obtained bibliographic data from the available literature and visualized network of existing collaborations and keywords using VOSviewer software.

RESULTS

A total of 196 papers were included in the study. Bibliometric analysis showed a rising trend in publication within the region, most of which were from institutions in Singapore and Thailand. The most common topics on KD studies included clinical features, complications, treatment, and comorbidities. 

Country characteristics such as gross domestic product (GDP) per capita, research and development (R&D) expenditure (% GDP), and number of physician and R&D researchers were positively correlated with bibliometric indices of KD research in SEA. Moreover, number of international linkages was significantly associated with KD research productivity in the region. 

CONCLUSION

In summary, we showed an increasing trend of KD research in SEA. Funding allocation and capacity building are necessary to strengthen research productivity within the region.

Asia ; Asia, Southeastern ; Bibliometrics ; Capacity Building ; Coronavirus ; Covid-19 ; Database ; Disease ; Efficiency ; Gross Domestic Product ; Guanosine Diphosphate ; Infection ; Infections ; Literature ; Mucocutaneous Lymph Node Syndrome ; Paper ; Physicians ; Prevalence ; Publications ; Research ; Research Personnel ; Rest ; Singapore ; Socioeconomic Factors ; Software ; Thailand ; Therapeutics

OBJECTIVES

This study aims to create a study protocol that will explore UST administrators’ perceptions of the benefits and risks of AI use in higher education learning environments.

METHODS

A qualitative descriptive design will be employed, using semi-structured interviews with at least fifteen administrators selected through purposive sampling. Audio-recorded interviews will be transcribed verbatim and subjected to thematic analysis using NVivo software

RESULTS

Administrators from different college-level fields perceive and engage with AI across various academic contexts. Exploring these perceptions will allow guidance in the development of coherent, contextually grounded institutional policies that promote responsible GenAI use and support digital leadership in Philippine higher education.

Human ; Artificial Intelligence ; Universities ; Software ; Administrative Personnel ; Intelligence ; Risk ; Policy

OBJECTIVE

This study assessed the correlation of macular pigment optical density (MPOD) and varying levels of exposure to video display terminals (VDT) among young adult Filipinos.

METHODS

This cross-sectional, analytical, single-center study compared the MPOD, measured using the Zeiss VISUCAM 500, between two groups of individuals aged 20 to 35 years old with differing VDT exposure. The more exposed group consisted of individuals who spent at least 8 hours per day on VDTs for the past 1 year, while the less exposed group spent less than 6 hours per day. Student’s t-test and chi-square test were used to compare the two groups, while Pearson's r coefficient was utilized to determine the relationship between MPOD and VDT exposure.

RESULTS

A total of 80 individuals (40 in each group) were included in the study. Both groups had similar profiles, except for refractive errors, which were significantly higher in the prolonged VDT exposure group (p = 0.02). The prolonged exposure group averaged 10 hours of VDT use per day, compared to 3 hours in the low VDT exposure group. The MPOD level, particularly the maximum optical density (Max OD), was significantly lower in the prolonged VDT exposure group (0.2034 DU) compared to the low VDT exposure group (0.2467 DU) (p = 0.0051). A negative weak correlation was observed between the VDT exposure hours and MPOD levels (Max OD r = –0.387, p = 0.0005).

CONCLUSION

This study found a weak but significant negative correlation between prolonged VDT exposure and lower MPOD levels, suggesting that extended screen time may contribute to reduced macular pigment density. While the correlation was weak (r = –0.387), these findings underscore the potential risk of diminished macular health with increased VDT use. The results highlight the importance of promoting protective strategies, such as reducing screen time and encouraging dietary or lifestyle changes that support eye health, especially among individuals with high VDT exposure.

Human ; Video Display Terminals (vdt) ; Computer Terminals

Hospitals worldwide recognise the importance of data and digital transformation in healthcare. We traced a smart hospital's data-driven journey to build an artificial intelligence and digital ecosystem (AIDE) to achieve healthcare excellence. We measured the impact of data and digital transformation on patient care and hospital operations, identifying key success factors, challenges, and opportunities. The use of data analytics and data science, robotic process automation, AI, cloud computing, Medical Internet of Things and robotics were stand-out areas for a hospital's data-driven journey. In the future, the adoption of a robust AI governance framework, enterprise risk management system, AI assurance and AI literacy are critical for success. Hospitals must adopt a digital-ready, digital-first strategy to build a thriving healthcare system and innovate care for tomorrow.
Artificial Intelligence ; Humans ; Delivery of Health Care ; Hospitals ; Cloud Computing ; Robotics ; Internet of Things ; Data Science

INTRODUCTION: Interpretation and analysis of magnetic resonance imaging (MRI) scans in clinical settings comprise time-consuming visual ratings and complex neuroimage processing that require trained professionals. To combat these challenges, artificial intelligence (AI) techniques can aid clinicians in interpreting brain MRI for accurate diagnosis of neurodegenerative diseases but they require extensive validation. Thus, the aim of this study was to validate the use of AI-based AQUA (Neurophet Inc., Seoul, Republic of Korea) segmentation software in a Southeast Asian community-based cohort with normal cognition, mild cognitive impairment (MCI) and dementia. METHOD: Study participants belonged to the community-based Biomarker and Cognition Study in Singapore. Participants aged between 30 and 95 years, having cognitive concerns, with no diagnosis of major psychiatric, neurological or systemic disorders who were recruited consecutively between April 2022 and July 2023 were included. Participants underwent neuropsychological assessments and structural MRI, and were classified as cognitively normal, with MCI or with dementia. MRI pre-processing using automated pipelines, along with human-based visual ratings, were compared against AI-based automated AQUA output. Default mode network grey matter (GM) volumes were compared between cognitively normal, MCI and dementia groups. RESULTS: A total of 90 participants (mean age at visit was 63.32±10.96 years) were included in the study (30 cognitively normal, 40 MCI and 20 dementia). Non-parametric Spearman correlation analysis indicated that AQUA-based and human-based visual ratings were correlated with total (ρ=0.66; P<0.0001), periventricular (ρ=0.50; P<0.0001) and deep (ρ=0.57; P<0.0001) white matter hyperintensities (WMH). Additionally, volumetric WMH obtained from AQUA and automated pipelines was also strongly correlated (ρ=0.84; P<0.0001) and these correlations remained after controlling for age at visit, sex and diagnosis. Linear regression analyses illustrated significantly different AQUA-derived default mode network GM volumes between cognitively normal, MCI and dementia groups. Dementia participants had significant atrophy in the posterior cingulate cortex compared to cognitively normal participants (P=0.021; 95% confidence interval [CI] -1.25 to -0.08) and in the hippocampus compared to cognitively normal (P=0.0049; 95% CI -1.05 to -0.16) and MCI participants (P=0.0036; 95% CI -1.02 to -0.17). CONCLUSION Our findings demonstrate high concordance between human-based visual ratings and AQUA-based ratings of WMH. Additionally, the AQUA GM segmentation pipeline showed good differentiation in key regions between cognitively normal, MCI and dementia participants. Based on these findings, the automated AQUA software could aid clinicians in examining MRI scans of patients with cognitive impairment.
Humans ; Cognitive Dysfunction/pathology* ; Magnetic Resonance Imaging/methods* ; Male ; Middle Aged ; Female ; Aged ; Artificial Intelligence ; Software ; Dementia/diagnostic imaging* ; Aged, 80 and over ; Adult ; Singapore ; Neuropsychological Tests ; Brain/pathology* ; Cohort Studies ; Gray Matter/pathology* ; Southeast Asian People

BACKGROUND: Retinal ganglion cell (RGC) death caused by acute ocular hypertension is an important characteristic of acute glaucoma. Receptor-interacting protein kinase 3 (RIPK3) that mediates necroptosis is a potential therapeutic target for RGC death. However, the current understanding of the targeting agents and mechanisms of RIPK3 in the treatment of glaucoma remains limited. Notably, artificial intelligence (AI) technologies have significantly advanced drug discovery. This study aimed to discover RIPK3 inhibitor with AI assistance. METHODS: An acute ocular hypertension model was used to simulate pathological ocular hypertension in vivo . We employed a series of AI methods, including large language and graph neural network models, to identify the target compounds of RIPK3. Subsequently, these target candidates were validated using molecular simulations (molecular docking, absorption, distribution, metabolism, excretion, and toxicity [ADMET] prediction, and molecular dynamics simulations) and biological experiments (Western blotting and fluorescence staining) in vitro and in vivo . RESULTS: AI-driven drug screening techniques have the potential to greatly accelerate drug development. A compound called HG9-91-01, identified using AI methods, exerted neuroprotective effects in acute glaucoma. Our research indicates that all five candidates recommended by AI were able to protect the morphological integrity of RGC cells when exposed to hypoxia and glucose deficiency, and HG9-91-01 showed a higher cell survival rate compared to the other candidates. Furthermore, HG9-91-01 was found to protect the retinal structure and reduce the loss of retinal layers in an acute glaucoma model. It was also observed that the neuroprotective effects of HG9-91-01 were highly correlated with the inhibition of PANoptosis (apoptosis, pyroptosis, and necroptosis). Finally, we found that HG9-91-01 can regulate key proteins related to PANoptosis, indicating that this compound exerts neuroprotective effects in the retina by inhibiting the expression of proteins related to apoptosis, pyroptosis, and necroptosis. CONCLUSION AI-enabled drug discovery revealed that HG9-91-01 could serve as a potential treatment for acute glaucoma.
Animals ; Glaucoma/metabolism* ; Neuroprotective Agents/pharmacology* ; Mice ; Receptor-Interacting Protein Serine-Threonine Kinases/metabolism* ; Artificial Intelligence ; Retinal Ganglion Cells/metabolism* ; Disease Models, Animal ; Molecular Docking Simulation ; Mice, Inbred C57BL ; Male

The Baihe Dihuang Decoction(BDD) is a representative traditional Chinese medicine formula that has been used to treat depression. This study employed metabolomics and network pharmacology to investigate the mechanism of BDD in the treatment of depression. Fifty male Sprague-Dawley(SD) rats were randomly assigned to the normal control group, model group, fluoxetine group, and high-and low-dose BDD groups. A rat model of depression was established through chronic unpredictable mild stress(CUMS), and the behavioral changes were detected by forced swimming test and open field test. Metabolomics technology was used to analyze the metabolic profiles of serum and hippocampal tissue to screen differential metabolites and related metabolic pathways. Additionally, network pharmacology and molecular docking techniques were used to investigate the key targets and core active ingredients of BDD in improving metabolic abnormalities of depression. A "component-target-metabolite-pathway" regulatory network was constructed. BDD could significantly improve depressive-like behavior in CUMS rats and regulate 12 differential metabolites in serum and 27 differential metabolites in the hippocampus, involving tryptophan metabolism, phenylalanine, tyrosine, and tryptophan biosynthesis, alanine, aspartate, and glutamate metabolism, tyrosine metabolism, and purine metabolism. Verbascoside, isorbascoside, and regaloside B were the key active ingredients for improving metabolic abnormalities in depression. Epidermal growth factor receptor(EGFR), protooncogene tyrosine-protein kinase(SRC), glycogen synthase kinase 3β(GSK3β), and androgen receptor(AR) were the key core targets for improving metabolic abnormalities of depression. This study offered a preliminary insight into the mechanism of BDD in alleviating metabolic abnormalities of depression through network regulation, providing valuable guidance for its clinical use and subsequent research.
Animals ; Drugs, Chinese Herbal/administration & dosage* ; Male ; Rats, Sprague-Dawley ; Rats ; Metabolomics ; Depression/genetics* ; Antidepressive Agents/chemistry* ; Network Pharmacology ; Hippocampus/drug effects* ; Humans ; Molecular Docking Simulation ; Behavior, Animal/drug effects* ; Disease Models, Animal

Network pharmacology and molecular docking were employed to predict the mechanism of Zhizhu Decoction in regulating macrophage polarization to reduce adipose tissue inflammation in obese children, and animal experiments were then carried out to validate the prediction results. The active ingredients and targets of Zhizhu Decoction were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform(TCMSP). The inflammation related targets in the adipose tissue of obese children were searched against GeneCards, OMIM, and DisGeNET, and a drug-disease-target network was established. STRING was used to construct a protein-protein interaction(PPI) network and screen for core targets. R language was used to carry out Gene Ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway enrichment analyses. AutoDock was used for the molecular docking between core targets and active ingredients. 24 SPF grade 6-week C57B/6J male mice were adaptively fed for 1 week, and 8 mice were randomly selected as the blank group. The remaining 16 mice were fed with high-fat diet for 8 weeks to onstruct a high-fat diet induced mouse obesity model. After successful modeling, the 16 mice were randomly divided into model group and Zhizhu Decoction group, with 8 mice in each group. Zhizhu Decoction group was intervened by gavage for 14 days, once a day. Blank group and model group were given an equal amount of sterile double distilled water(ddH_2O) by gavage daily. After the last gavage, serum and inguinal adipose tissue were collected from mice for testing. The morphology of inguinal adipose tissue was observed by hematoxylin-eosin(HE) staining, the levels of inflammatory factors interleukin-6(IL-6) and tumor necrosis factor-α(TNF-α)were detected by enzyme-linked immunosorbent assay(ELISA), and the protein expression of macrophage marker molecule nitric oxide synthase(iNOS) and epidermal growth factor like hormone receptor 1(F4/80) was detected by immunofluorescence staining. Network pharmacology predicted luteolin, naringenin, and nobiletin as the main active ingredients in Zhizhu Decoction and 15 core targets. KEGG pathway enrichment analysis revealed involvement in the key signaling pathway of nuclear factor κB(NF-κB). Molecular docking showed that the active ingredients of Zhizhu Decoction bound well to the core targets. Animal experiment showed that compared with the model group, Zhizhu Decoction reduced the distribution of inflammatory cytokines in the inguinal adipose tissue of mice, lowered the levels of TNF-α and IL-6 in the serum(P<0.05, P<0.01), and down-regulated the expression of iNOS and F4/80(P<0.05). The results showed that the active ingredients in Zhizhu Decoction, such as luteolin, naringenin, and nobiletin, inhibit the aggregation of macrophages in adipose tissue, downregulate their classic activated macrophage(M1) polarization, reduce the expression of inflammatory factors IL-6 and TNF-α, and thus improve adipose tissue inflammation in obese mice.
Animals ; Drugs, Chinese Herbal/pharmacology* ; Molecular Docking Simulation ; Adipose Tissue/immunology* ; Mice ; Male ; Humans ; Network Pharmacology ; Macrophages/immunology* ; Mice, Inbred C57BL ; Child ; Protein Interaction Maps/drug effects* ; Obesity/genetics* ; Inflammation/drug therapy*