- COVID-19 epidemiological studies
- Pregnancy and preeclampsia studies
- Birth, Development, and Health
- Artificial Intelligence in Healthcare
- Machine Learning in Healthcare
- Data-Driven Disease Surveillance
- RNA and protein synthesis mechanisms
- Protein Structure and Dynamics
- Retinal Imaging and Analysis
- Machine Learning in Bioinformatics
- Gestational Diabetes Research and Management
- Neonatal Respiratory Health Research
- Retinal and Optic Conditions
- Explainable Artificial Intelligence (XAI)
- Congenital Diaphragmatic Hernia Studies
- vaccines and immunoinformatics approaches
- Genomics and Phylogenetic Studies
- COVID-19 Pandemic Impacts
- HIV Research and Treatment
- Stroke Rehabilitation and Recovery
- HIV/AIDS drug development and treatment
- Vaccine Coverage and Hesitancy
- Misinformation and Its Impacts
- Biomedical Text Mining and Ontologies
- Machine Learning and Data Classification
Taipei Medical University Hospital
2018-2025
National Yang Ming Chiao Tung University
2007-2025
Taipei Medical University
2016-2025
Albany Medical Center Hospital
2024
Artificial Intelligence in Medicine (Canada)
2015-2024
University of Colorado Anschutz Medical Campus
2023
New York Eye and Ear Infirmary
2012-2019
Chung Shan Medical University
2019
Chung Shan Medical University Hospital
2019
Oracle (United States)
2008-2018
To investigate a relationship between the inner segment-outer segment (IS-OS) junctional layer integrity and overlying retinal sensitivity assessed by Spectral OCT/SLO (spectral-domain optical coherence tomography) microperimetry testing in patients with dry wet forms of age-related macular degeneration (AMD).Spectral-domain tomography examination were performed 55 eyes 43 consecutive AMD. Microperimetry maps registered onto three-dimensional topography maps, point-to-point analysis...
The risk factors of diabetic retinopathy (DR) were investigated extensively in the past studies, but it remains unknown which more associated with DR than others. If we can detect related accurately, then exercise early prevention strategies for most high-risk population. purpose this study is to build a prediction model type 2 diabetes mellitus using data mining techniques including support vector machines, decision trees, artificial neural networks, and logistic regressions.Experimental...
To determine whether periodontitis is a modifiable risk factor for dementia.Prospective cohort study.National Health Insurance Research Database in Taiwan.Individuals aged 65 and older with (n = 3,028) an age- sex-matched control group 3,028).Individuals were compared controls incidence density hazard ratio (HR) of new-onset dementia. Periodontitis was defined according to International Classification Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 523.3-5 diagnosed by...
RNA-protein interaction plays an essential role in several biological processes, such as protein synthesis, gene expression, posttranscriptional regulation and viral infectivity. Identification of RNA-binding sites proteins provides valuable insights for biologists. However, experimental determination remains time-consuming labor-intensive. Thus, computational approaches prediction have become highly desirable. Extensive studies site led to the development methods. they could yield low...
We developed and validated an artificial intelligence (AI)-assisted prediction of preeclampsia applied to a nationwide health insurance dataset in Indonesia.The BPJS Kesehatan have been preprocessed using nested case-control design into preeclampsia/eclampsia (n = 3318) normotensive pregnant women 19,883) from all with one pregnancy. The provided 95 features consisting demographic variables medical histories started 24 months event ended by delivery as the event. Six algorithms were compared...
Introduction Assisted reproductive technology has been proposed for women with infertility. Moreover, in vitro fertilization (IVF) cycles are increasing. Factors contributing to successful pregnancy have widely explored. In this study, we used machine learning algorithms construct prediction models clinical pregnancies IVF. Materials and methods A total of 24,730 patients entered IVF intracytoplasmic sperm injection outcomes at Taipei Medical University Hospital. Data included patient...
An all-inclusive and accurate prediction of outcomes for patients with acute ischemic stroke (AIS) is crucial clinical decision-making. This study developed extreme gradient boosting (XGBoost)-based models using three simple factors-age, fasting glucose, National Institutes Health Stroke Scale (NIHSS) scores-to predict the three-month functional after AIS. We retrieved medical records 1848 diagnosed AIS managed at a single center between 2016 2020. validated predictions ranked importance...
Sedentary behaviors and dietary intake are independently associated with obesity risk. In the literature, only a few studies have investigated gender differences for such associations. The present study aims to assess association of sedentary unhealthy foods in men women comparative manner. analysis presented this was based on data from population-based, cross-sectional, nationally representative survey (Indonesian Basic Health Research 2013/RISKESDAS 2013). total, 222,650 248,590 aged 19–55...
Abstract Background Dengue fever is a well-studied vector-borne disease in tropical and subtropical areas of the world. Several methods for predicting occurrence dengue Taiwan have been proposed. However, to best our knowledge, no study has investigated relationship between air quality indices (AQIs) Taiwan. Results This aimed develop prediction model which meteorological factors, vector index, AQIs were incorporated into different machine learning algorithms. A total 805 records from 2013...
Abstract Background Protein subcellular localization is crucial for genome annotation, protein function prediction, and drug discovery. Determination of using experimental approaches time-consuming; thus, computational become highly desirable. Extensive studies prediction have led to the development several methods including composition-based homology-based methods. However, their performance might be significantly degraded if homologous sequences are not detected. Moreover, that integrate...
Abstract Background Development of computational tools that can accurately predict presence and location B-cell epitopes on pathogenic proteins has a valuable application to the field vaccinology. Because highly variable yet enigmatic nature epitopes, their prediction presents great challenge immunologists. Methods We propose method, BEEPro ( B -cell e pitope by volutionary information pro pensity scales), which adapts linear averaging scheme 16 properties using support vector machine model...
Background Preeclampsia and intrauterine growth restriction are placental dysfunction–related disorders (PDDs) that require a referral decision be made within certain time period. An appropriate prediction model should developed for these diseases. However, previous models did not demonstrate robust performances and/or they were from datasets with highly imbalanced classes. Objective In this study, we predictive of PDDs by machine learning uses features at 24-37 weeks’ gestation, including...
South Korea is among the best-performing countries in tackling coronavirus pandemic by using mass drive-through testing, face mask use, and extensive social distancing. However, understanding patterns of risk perception could also facilitate effective communication to minimize impacts disease spread during this crisis.We attempt explore community health perceptions COVID-19 internet search data.Google Trends (GT) NAVER relative volumes (RSVs) data were collected COVID-19-related terms Korean...
The prevalence of nonalcoholic fatty liver disease is increasing over time worldwide, with similar trends to those diabetes and obesity. A biopsy, the gold standard diagnosis, not favored due its invasiveness. Meanwhile, noninvasive evaluation methods are still either very expensive or demonstrate poor diagnostic performances, thus, limiting their applications. We developed neural network-based models assess classify severity using B-mode ultrasound (US) images.We followed standards for...
Maternal nutrition intake during pregnancy may affect the mother-to-child transmission of bacteria, resulting in gut microflora changes offspring, with long-term health consequences later life. Longitudinal human studies are lacking, as only a small amount showing effect on microbiome infants have been performed, and these mainly conducted animals. This pilot study explores effects high or low fruit vegetable gestational infant microbiome. We enrolled pregnant women complete 3-day dietary...
The human immunodeficiency virus type 1 (HIV-1) aspartic protease is an important enzyme owing to its imperative part in viral development and a causative agent of deadliest disease known as acquired immune deficiency syndrome (AIDS). Development HIV-1 inhibitors can help understand the specificity substrates which restrain replication HIV-1, thus antagonize AIDS. However, experimental methods identification cleavage sites are generally time-consuming labor-intensive. Therefore, using...
Acute hepatopancreatic necrosis disease (AHPND) in shrimp is caused by Vibrio strains that harbor a pVA1-like plasmid containing the pirA and pirB genes. It also known production of PirA PirB proteins, which are key factors drive observed symptoms AHPND, can be influenced environmental conditions this leads to changes virulence bacteria. However, our knowledge, mechanisms involved regulating expression pirA/pirB genes have not previously been investigated. In study, we show AHPND-causing...
Background: Current nomogram can only be created for regression algorithm. Providing any machine learning (ML) algorithms may accelerate model deployment in clinical settings or improve availability. We developed an R package and web application to construct with explainability of ML algorithms. Methods: formulated a function transform prediction into nomogram, requiring datasets with: (1) all possible combinations predictor values; (2) the corresponding outputs model; (3) values each...