Browsing by Author "Lin C.Y."

Filter results by typing the first few letters
Now showing 1 - 20 of 34
  • No Thumbnail Available
    ItemMetadata only
    A Decision Machine Learning Support System for Human Skin Disease Classifier
    (2022-01-01) Banditsingha P.; Thaipisutikul T.; Shih T.K.; Lin C.Y.; Mahidol University
    For the past decades, the prevalence of dermatological disorders, especially human skin diseases, has been rising. The majority of these diseases are contagious and are also based on visual perceptions. Although many works have shown promising results on the image classification problem, only a few studies compare traditional machine learning models and the recent deep learning models with various metrics on human skin diseases classification. Therefore, in this paper, we propose A Decision Machine Learning Support System for Human Skin Disease Classifier (DSSC) to classify five skin disease classes, including 750 images gained from the Dermnet dataset. In particular, we perform image pre-processing, image resizing, image interpolation, and image augmentation to adjust the input images into the proper format for all models. Through the extensive experiments, RestNet50 outperforms other deep learning and traditional machine learning methods in all metrics, including accuracy, precision, recall, and F-measure by a large margin.
  • No Thumbnail Available
    ItemMetadata only
    A deep feature-level fusion model for masked face identity recommendation system
    (2022-01-01) Thaipisutikul T.; Tatiyamaneekul P.; Lin C.Y.; Tuarob S.; Mahidol University
    The widespread occurrences of airborne outbreaks (e.g., COVID-19) and pollution (e.g., PM2.5) have urged people in the affected regions to protect themselves by wearing face masks. In certain areas, wearing masks amidst such health-endangering times is even enforced by law. While most people wear masks to guard themselves against airborne substances, some exploit such excuses and use face masks to conceal their identity for criminal purposes such as shoplifting, robbery, drug transport, and assault. While automatic face recognition models have been proposed, most of these models aim to identify clear, unobstructed faces for authentication purposes and cannot effectively handle cases where masks cover most facial areas. To mitigate such a problem, this paper proposes a deep-learning-based feature-fusion framework, FIREC, that combines additional demographic-estimated features such as age, gender, and race into the underlying facial representation to compensate for the information lost due to mask obstruction. Given an image of a masked face, our system recommends a ranked list of potential identities of the person behind the mask. Empirical results show that the best configuration of our proposed framework can recognize bare faces and masked faces with the accuracy of 99.34% and 97.65% in terms of Hit@10, respectively. The proposed framework could greatly benefit high-recall facial identity recognition applications such as identifying potential suspects from CCTV or passers-by’s cameras, especially during crisis times when people commonly cover their faces with protective masks.
  • No Thumbnail Available
    ItemMetadata only
    A Large Dengue Outbreak in Taiwan, 2023: Driven by Imported Cases, Serotype Cocirculation, and Climate Variability
    (2026-03-01) Huang J.Y.; Weng S.F.; Yang Z.S.; Tung Y.W.; Wang W.H.; Assavalapsakul W.; Thitithanyanont A.; Chao D.Y.; Lin C.Y.; Chen Y.H.; Wang S.F.; Huang J.Y.; Mahidol University
    Background Taiwan, a region traditionally considered non-endemic for dengue, experienced an unexpected and large-scale outbreak in 2023. We investigated the multifactorial drivers of this outbreak, including cross-border viral importation, serotype cocirculation, vector ecology, and climate variability. Methods We analyzed national dengue surveillance data (2013-2023), meteorological records, and Breteau Index (BI) values, alongside molecular serotyping and whole-genome sequencing of clinical isolates. Time-lagged Poisson regression was used to identify predictors of indigenous dengue transmission in Kaohsiung and Tainan. Full-genome comparisons were conducted between 2023 strains and historical epidemic isolates. Results A total of 26 706 laboratory-confirmed cases were reported, primarily in Tainan (80.7%) and Kaohsiung (11.9%). Real-time RT-PCR identified cocirculating DENV-1 and DENV-2 strains. Phylogenetic analysis confirmed the 2023 DENV-1 and DENV-2 strains were genetically linked to contemporary strains from Southeast Asian countries. Whole-genome sequencing identified several nonsynonymous mutations in the NS2A, NS3, and NS5 regions when compared with historical outbreak isolates. Time-lagged regression showed that imported cases, precipitation, and the BI were associated with incidence in univariate models. In Kaohsiung, the best-fitting multivariable model included the BI, but temperature and precipitation were the independent predictors. In Tainan, precipitation and, at longer lags, imported cases were more influential, while the BI lost significance after adjustment. Conclusions The 2023 dengue outbreak in Taiwan was driven by a complex interplay between viral introductions, climatic conditions, and vector dynamics. The differing transmission drivers observed between cities highlight the need for region-specific vector surveillance, climate-informed early warning systems, and sustained genomic monitoring to prevent future re-emergence of dengue in this non-endemic setting.
  • No Thumbnail Available
    ItemMetadata only
    Acute-on-chronic liver failure (ACLF): the 'Kyoto Consensus'-steps from Asia
    (2025-02-01) Choudhury A.; Kulkarni A.V.; Arora V.; Soin A.S.; Dokmeci A.K.; Chowdhury A.; Koshy A.; Duseja A.; Kumar A.; Mishra A.K.; Patwa A.K.; Sood A.; Roy A.; Shukla A.; Chan A.; Krag A.; Mukund A.; Mandot A.; Goel A.; Butt A.S.; Sahney A.; Shrestha A.; Cárdenas A.; Di Giorgio A.; Arora A.; Anand A.C.; Dhawan A.; Jindal A.; Saraya A.; Srivastava A.; Kumar A.; Kaewdech A.; Pande A.; Rastogi A.; Valsan A.; Goel A.; Kumar A.; Singal A.K.; Tanaka A.; Coilly A.; Singh A.; Meena B.L.; Jagadisan B.; Sharma B.C.; Lal B.B.; Eapen C.E.; Yaghi C.; Kedarisetty C.K.; Kim C.W.; Panackel C.; Yu C.; Kalal C.R.; Bihari C.; Huang C.H.; Vasishtha C.; Jansen C.; Strassburg C.; Lin C.Y.; Karvellas C.J.; Lesmana C.R.A.; Philips C.A.; Shawcross D.; Kapoor D.; Agrawal D.; Payawal D.A.; Praharaj D.L.; Jothimani D.; Song D.S.; Kim D.J.; Kim D.S.; Zhongping D.; Karim F.; Durand F.; Shiha G.E.; D'Amico G.; Lau G.K.; Pati G.K.; Narro G.E.C.; Lee G.H.; Adali G.; Dhakal G.P.; Szabo G.; Lin H.C.; Li H.; Nair H.K.; Devarbhavi H.; Tevethia H.; Ghazinian H.; Ilango H.; Yu H.L.; Hasan I.; Fernandez J.; George J.; Behari J.; Fung J.; Bajaj J.; Benjamin J.; Lai J.C.; Jia J.; Hu J.H.; Choudhury A.; Mahidol University
    Acute-on-chronic liver failure (ACLF) is a condition associated with high mortality in the absence of liver transplantation. There have been various definitions proposed worldwide. The first consensus report of the working party of the Asian Pacific Association for the Study of the Liver (APASL) set in 2004 on ACLF was published in 2009, and the "APASL ACLF Research Consortium (AARC)" was formed in 2012. The AARC database has prospectively collected nearly 10,500 cases of ACLF from various countries in the Asia-Pacific region. This database has been instrumental in developing the AARC score and grade of ACLF, the concept of the 'Golden Therapeutic Window', the 'transplant window', and plasmapheresis as a treatment modality. Also, the data has been key to identifying pediatric ACLF. The European Association for the Study of Liver-Chronic Liver Failure (EASL CLIF) and the North American Association for the Study of the End Stage Liver Disease (NACSELD) from the West added the concepts of organ failure and infection as precipitants for the development of ACLF and CLIF-Sequential Organ Failure Assessment (SOFA) and NACSELD scores for prognostication. The Chinese Group on the Study of Severe Hepatitis B (COSSH) added COSSH-ACLF criteria to manage hepatitis b virus-ACLF with and without cirrhosis. The literature supports these definitions to be equally effective in their respective cohorts in identifying patients with high mortality. To overcome the differences and to develop a global consensus, APASL took the initiative and invited the global stakeholders, including opinion leaders from Asia, EASL and AASLD, and other researchers in the field of ACLF to identify the key issues and develop an evidence-based consensus document. The consensus document was presented in a hybrid format at the APASL annual meeting in Kyoto in March 2024. The 'Kyoto APASL Consensus' presented below carries the final recommendations along with the relevant background information and areas requiring future studies.
  • No Thumbnail Available
    ItemMetadata only
    An Improved Face Mask-aware Recognition System Based on Deep Learning
    (2022-01-01) Lin C.Y.; Rojanasarit A.; Thaipisutikul T.; Lung C.W.; Akhyar F.; Mahidol University
    Face mask detection and recognition have been incorporated into many applications in daily life, especially during the current COVID-19 pandemic. To mitigate the spread of coronavirus, wearing face masks has become commonplace. However, traditional face detection and recognition systems utilize main facial features such as the mouth, nose, and eyes to determine a person’s identity. Masks make facial detection and recognition tasks more challenging since certain parts of the face are concealed. Yet, how to improve the performance of existing systems with a face mask overlaid on the original face input images remains an open area of inquiry. In this study, we propose an improved face mask-aware recognition system named ‘MAR’ based on deep learning, which can tackle challenges in face mask detection and recognition. MAR consists of five main modules to handle various kinds of input images. We re-train the CenterNet model with our augmented face mask inputs to perform face mask detection and propose four variations on face mask recognition models based on the pre-trained ArcFace to handle facial recognition. Finally, we demonstrate the effectiveness of our proposed models on the VGGFACE2 dataset and achieve a high accuracy score on both detection and recognition tasks.
  • No Thumbnail Available
    ItemMetadata only
    An Improved Speed Estimation Using Deep Homography Transformation Regression Network on Monocular Videos
    (2023-01-01) Yohannes E.; Lin C.Y.; Shih T.K.; Thaipisutikul T.; Enkhbat A.; Utaminingrum F.; Mahidol University
    Vehicle speed estimation is one of the most critical issues in intelligent transportation system (ITS) research, while defining distance and identifying direction have become an inseparable part of vehicle speed estimation. Despite the success of traditional and deep learning approaches in estimating vehicle speed, the high cost of deploying hardware devices to get all related sensor data, such as infrared/ultrasonic devices, Global Positioning Systems (GPS), Light Detection and Ranging (LiDAR systems), and magnetic devices, has become the key barrier to improvement in previous studies. In this paper, our proposed model consists of two main components: 1) a vehicle detection and tracking component - this module is designed for creating reliable detection and tracking every specific object without doing calibration; 2) homography transformation regression network - this module has a function to solve occlusion issues and estimate vehicle speed accurately and efficiently. Experimental results on two datasets show that the proposed method outperforms the state-of-the-art methods by reducing the mean square error (MSE) metric from 14.02 to 6.56 based on deep learning approaches. We have announced our test code and model on GitHub with https://github.com/ervinyo/Speed-Estimation-Using-Homography-Transformation-and-Regression-Network.
  • No Thumbnail Available
    ItemMetadata only
    An unexpected dengue outbreak in Taiwan, 2023: A retrospective analysis of potential risk factors
    (2025-11-01) Tung Y.W.; Yang Z.S.; Wang W.H.; Hsu Y.T.; Tsui C.I.; Assavalapsakul W.; Thitithanyanont A.; Lin C.Y.; Chao D.Y.; Chen Y.H.; Wang S.F.; Tung Y.W.; Mahidol University
    Background: Taiwan experienced a major dengue outbreak in 2023 following the relaxation of COVID-19 border controls. The contributing factors remained unclear. This study investigated potential virological, immunological, and clinical drivers. Methods: We retrospectively analyzed laboratory-confirmed dengue virus (DENV) infections at a tertiary care hospital in southern Taiwan. Serotypes were identified by qRT-PCR. Viral origins were assessed through phylogenetic and envelope (E) gene amino acid analyses. Clinical features of DENV-1 and DENV-2 cases were compared. Neutralization and antibody-dependent enhancement (ADE) were evaluated using PRNT and ADE assays. Results: DENV-1 and DENV-2 were identified as the predominant circulating serotypes. Clinical analysis revealed that DENV-2 infection was significantly associated with older age, diabetes mellitus, hypertension, and elevated hematocrit levels (p < 0.05), and these associations remained statistically significant in multivariate analysis. Phylogenetic analysis showed that DENV-1 isolates belonged to genotypes I and IV, while DENV-2 strains were of the cosmopolitan genotype. These viruses clustered closely with strains from Southeast Asia. Amino acid analysis indicated that DENV-1 strains exhibited 2–10 substitutions relative to 2014 isolates, while DENV-2 strains closely matched those from 2015. Sera from the 2014–2015 outbreaks demonstrated potent homotypic but limited heterotypic neutralization. ADE was observed in heterotypic infection contexts. Conclusions: The 2023 dengue outbreak in Taiwan was driven by co-circulation of DENV-1 and DENV-2, limited heterotypic immunity, and ADE. These findings highlight the importance of integrated virological surveillance, genotype monitoring, and immunological assessment to inform dengue control strategies in non-endemic regions experiencing imported viral threats.
  • No Thumbnail Available
    ItemMetadata only
    Boosting the detection performance of severe acute respiratory syndrome coronavirus 2 test through a sensitive optical biosensor with new superior antibody
    (2022-01-01) Lin C.Y.; Wang W.H.; Li M.C.; Lin Y.T.; Yang Z.S.; Urbina A.N.; Assavalapsakul W.; Thitithanyanont A.; Chen K.R.; Kuo C.C.; Lin Y.X.; Hsiao H.H.; Lin K.D.; Lin S.Y.; Chen Y.H.; Yu M.L.; Su L.C.; Wang S.F.; Mahidol University
    The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus emerged in late 2019 leading to the COVID-19 disease pandemic that triggered socioeconomic turmoil worldwide. A precise, prompt, and affordable diagnostic assay is essential for the detection of SARS-CoV-2 as well as its variants. Antibody against SARS-CoV-2 spike (S) protein was reported as a suitable strategy for therapy and diagnosis of COVID-19. We, therefore, developed a quick and precise phase-sensitive surface plasmon resonance (PS-SPR) biosensor integrated with a novel generated anti-S monoclonal antibody (S-mAb). Our results indicated that the newly generated S-mAb could detect the original SARS-CoV-2 strain along with its variants. In addition, a SARS-CoV-2 pseudovirus, which could be processed in BSL-2 facility was generated for evaluation of sensitivity and specificity of the assays including PS-SPR, homemade target-captured ELISA, spike rapid antigen test (SRAT), and quantitative reverse transcription polymerase chain reaction (qRT-PCR). Experimentally, PS-SPR exerted high sensitivity to detect SARS-CoV-2 pseudovirus at 589 copies/ml, with 7-fold and 70-fold increase in sensitivity when compared with the two conventional immunoassays, including homemade target-captured ELISA (4 × 103 copies/ml) and SRAT (4 × 104 copies/ml), using the identical antibody. Moreover, the PS-SPR was applied in the measurement of mimic clinical samples containing the SARS-CoV-2 pseudovirus mixed with nasal mucosa. The detection limit of PS-SPR is calculated to be 1725 copies/ml, which has higher accuracy than homemade target-captured ELISA (4 × 104 copies/ml) and SRAT (4 × 105 copies/ml) and is comparable with qRT-PCR (1250 copies/ml). Finally, the ability of PS-SPR to detect SARS-CoV-2 in real clinical specimens was further demonstrated, and the assay time was less than 10 min. Taken together, our results indicate that this novel S-mAb integrated into PS-SPR biosensor demonstrates high sensitivity and is time-saving in SARS-CoV-2 virus detection. This study suggests that incorporation of a high specific recognizer in SPR biosensor is an alternative strategy that could be applied in developing other emerging or re-emerging pathogenic detection platforms.
  • No Thumbnail Available
    ItemMetadata only
    Correction: Evaluation of two weight stigma scales in Malaysian university students: weight self-stigma questionnaire and perceived weight stigma scale (Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, (2022), 27, 7, (2595-2604), 10.1007/s40519-022-01398-3)
    (2023-12-01) Gan W.Y.; Tung S.E.H.; Ruckwongpatr K.; Ghavifekr S.; Paratthakonkun C.; Nurmala I.; Chang Y.L.; Latner J.D.; Huang R.Y.; Lin C.Y.; Mahidol University
    In this article, the given and family names of the following two authors were incorrect. The correct names are given below. Given name is Kamolthip and family name is Ruckwongpatr. Given name is Chirawat and family name is Paratthakonkun. Given name is Kamolthip and family name is Ruckwongpatr. Given name is Chirawat and family name is Paratthakonkun. The original article [1] has been corrected.
  • No Thumbnail Available
    ItemMetadata only
    DC-SIGN and Galectin-3 individually and collaboratively regulate H5N1 and H7N9 avian influenza A virus infection via interaction with viral envelope hemagglutinin protein
    (2023-05-17) Yang Z.S.; Wang W.H.; Lin Y.T.; Lin C.Y.; Urbina A.N.; Thitithanyanont A.; Lu P.L.; Chen Y.H.; Wang S.F.; Mahidol University
    DC-SIGN and Galectin-3 are two different lectins and have been reported to participate in regulation of several virus infections. WHO has pointed that H5N1 and H7N9 avian influenza viruses (AIVs) play continuous threats to global health. AIV hemagglutinin (HA) protein-a highly glycosylated protein-mediates influenza infection and was proposed to have DC-SIGN and Gal3 interactive domains. This study aims to address the individual and collaborative roles of DC-SIGN and Gal3 toward AIVs infection. Firstly, A549 cells with DC-SIGN expression or Gal3-knockdown, via lentiviral vector-mediated CD209 gene expression or LGALS-3 gene knockdown, respectively were generated. Quantitative reverse transcription PCR (qRT-PCR) results indicated that DC-SIGN expression and Gal3 knockdown in A549 cells significantly promoted and ameliorated HA or NP gene expression, respectively after H5N1 and H7N9-reverse genetics (RG) virus postinfections (P < 0.05). Similar results observed in immunoblotting, indicating that DC-SIGN expression significantly facilitated H5N1-RG and H7N9-RG infections (P < 0.05), whereas Gal3 knockdown significantly reduced both viral infections (P < 0.05). Furthermore, we found that DC-SIGN and Gal3 co-expression significantly enhanced infectivity of both H5N1-RG and H7N9-RG viruses (P < 0.01) and higher regulatory capabilities by DC-SIGN and Gal3 in H5N1-RG than H7N9-RG were noted. The promoting effect mainly relied on exogenous Gal3 and DC-SIGN directly interacting with the HA protein of H5N1 or H7N9 AIVs, subsequently enhancing virus infection. This study sheds light on two different lectins individually and collaboratively regulating H5N1 and H7N9 AIVs infection and suggests that inhibitors against DC-SIGN and Gal3 interacting with HA could be utilized as alternative antiviral strategies.
  • No Thumbnail Available
    ItemMetadata only
    Evaluation of two weight stigma scales in Malaysian university students: weight self-stigma questionnaire and perceived weight stigma scale
    (2022-10-01) Gan W.Y.; Tung S.E.H.; Kamolthip R.; Ghavifekr S.; Chirawat P.; Nurmala I.; Chang Y.L.; Latner J.D.; Huang R.Y.; Lin C.Y.; Mahidol University
    Background: This study aimed to examine the psychometric properties of the Weight Self-Stigma Questionnaire (WSSQ) and Perceived Weight Stigma Scale (PWS) among Malaysian university students. Methods: University students who were studying in a Malaysia university with a mean age of 24.0 years (n = 380; females 71.6%) were recruited through convenience sampling between 19 August and 30 September 2021. They completed a Google Form consisting of information on sociodemographic background, weight stigma, psychological distress and self-reported body weight and height. Psychometric testing was conducted using the classical test theory (including confirmatory factor analysis) and Rasch models to confirm the two-factor structure of WSSQ and the unidimensional structure of the PWS using the various fit indices. Concurrent validity of the total scores of WSSQ and PWS with psychological distress and body mass index (BMI) was also investigated. Internal consistency using Cronbach’s alpha was conducted. Results: The confirmatory factor analyses and Rasch analyses verified the two-factor structure for the WSSQ and the single-factor structure for the PWS. Both the WSSQ and PWS showed good internal consistency and good concurrent validity as demonstrated by their significant correlations with psychological distress and BMI. Conclusion: The WSSQ and PWS have strong validity and reliability, and they can both be used to assess weight stigma among Malaysian university students. Level of evidence: V: Descriptive study.
  • No Thumbnail Available
    ItemMetadata only
    Facile and Unplugged Surface Plasmon Resonance Biosensor with NIR-Emitting Perovskite Nanocomposites for Fast Detection of SARS-CoV-2
    (2023-01-01) Chen L.C.; Li M.C.; Chen K.R.; Cheng Y.J.; Wu X.Y.; Chen S.A.; Youh M.J.; Kuo C.C.; Lin Y.X.; Lin C.Y.; Wang C.F.; Huang C.F.; Lin S.Y.; Wang W.H.; Chen Y.H.; Yu M.L.; Thitithanyanont A.; Wang S.F.; Su L.C.; Mahidol University
    The emergence of the coronavirus disease 2019 (COVID-19) pandemic prompted researchers to develop portable biosensing platforms, anticipating to detect the analyte in a label-free, direct, and simple manner, for deploying on site to prevent the spread of the infectious disease. Herein, we developed a facile wavelength-based SPR sensor built with the aid of a 3D printing technology and synthesized air-stable NIR-emitting perovskite nanocomposites as the light source. The simple synthesis processes for the perovskite quantum dots enabled low-cost and large-area production and good emission stability. The integration of the two technologies enabled the proposed SPR sensor to exhibit the characteristics of lightweight, compactness, and being without a plug, just fitting the requirements of on-site detection. Experimentally, the detection limit of the proposed NIR SPR biosensor for refractive index change reached the 10-6 RIU level, comparable with that of state-of-the-art portable SPR sensors. In addition, the bio-applicability of the platform was validated by incorporating a homemade high-affinity polyclonal antibody toward the SARS-CoV-2 spike protein. The results demonstrated that the proposed system was capable of discriminating between clinical swab samples collected from COVID-19 patients and healthy subjects because the used polyclonal antibody exhibited high specificity against SARS-CoV-2. Most importantly, the whole measurement process not only took less than 15 min but also needed no complex procedures or multiple reagents. We believe that the findings disclosed in this work can open an avenue in the field of on-site detection for highly pathogenic viruses.
  • No Thumbnail Available
    ItemMetadata only
    Generative AI Shanshui animation enhancement using Perlin noise and diffusion models
    (2026-12-01) Wattanachote K.; Lin C.Y.; Hsu S.E.; Shih T.K.; Wattanachote K.; Mahidol University
    Deep learning models have achieved remarkable advancements in image generation but face persistent challenges in synthesizing traditional Shanshui (mountain-water) landscape paintings due to limited domain-specific training data and the complexity of aesthetic principles. This study integrated Perlin Noise, Stable Diffusion, ControlNet, and AnimateDiff to enhance Shanshui landscape generation and animation. Perlin Noise constructs naturalistic skeletal structures, which are further refined using ControlNet for precise structural control. Advanced prompt engineering with GPT-4 and Textual Inversion improved prompt descriptiveness and mitigated low-quality outputs. Furthermore, LoRA fine-tuning improved the adaptability of our Shanshui landscapes model. Integrating I2V Encoders and AnimateDiff enabled the seamless transformation of static landscape images into dynamic animations, preserving artistic authenticity while introducing motion consistency. The experimental results demonstrated significant improvements in realism, stylistic fidelity, and diversity, addressing key limitations in existing generative approaches. This framework not only advances the field of generative AI in digital art but also offers new opportunities for the creation of multimedia content and cultural preservation through the synthesis of computational Shanshui animation.
  • No Thumbnail Available
    ItemMetadata only
    Healthcare providers' experiences in providing sexual health care to breast cancer survivors: A mixed-methods systematic review
    (2023-01-01) Pimsen A.; Lin W.H.; Lin C.Y.; Kuo Y.L.; Shu B.C.; Pimsen A.; Mahidol University
    Aims: To analyse healthcare providers' (HCPs) experiences in sexual health care through the mixed-methods systematic review (MMSR). Background: Sexual health for breast cancer survivors (BCSs) is becoming increasingly important as survivors live longer. HCPs are critical in providing sexual health care. Design: A mixed-methods systematic review. Methods: Literature searches were conducted in databases MEDLINE, CINAHL, Psychology & Behavioral Sciences Collection, Web of Science, Cochrane Library, Scopus, ClinicalTrials.gov and reference lists were searched from inception to 30 December 2022. Two independent reviewers extracted and analysed the data using the JBI guidelines for MMSR. Results: After screening for 2849 citations, 19 studies were eligible for MMSR, involving 2068 HCPs. Most HCPs believe that sexual health care is their responsibility. However, sexual health was not adequately addressed. A lack of knowledge was the most significant barrier to providing sexual health care. Moreover, HCPs would like to acquire more knowledge and felt that current sexual healthcare training was inadequate. Conclusions: Findings suggest that HCPs did not frequently address sexual health in BCSs and that lack of knowledge was the most common barrier. Healthcare session administrators should allocate resources for sexual healthcare training that offer multiple formats, accessible content and convenience. They should also be multifaceted and proactive, meet the diverse needs of BCS at different stages and focus on effective communication. Relevance to clinical practice: This study highlights the importance of addressing sexual health in BCSs and the need for HCPs to receive training in this area. Training should be multifaceted, proactive and meet the diverse needs of BCSs at different stages, with a focus on effective communication. By addressing this issue, HCPs will be better equipped to support the sexual health needs of BCSs, ultimately improving their overall well-being and quality of life. PROSPERO Registration Number: CRD42022327018 (https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=327018).
  • No Thumbnail Available
    ItemMetadata only
    Impact of social media addiction on anxiety symptoms as modified by social support and its subscales
    (2025-01-01) Yeh C.R.; Vo H.T.; Lin C.Y.; Lai C.F.; Tran Le T.C.; Yang S.H.; Chao J.C.J.; Tsai P.S.; Duong T.V.; Yeh C.R.; Mahidol University
    Objectives: We aimed to investigate the moderating effect of social support and its subscales on the relationship between social media addiction (SMA) and anxiety symptoms in young Taiwanese adults. Methods: A cross-sectional study was conducted on young adults in universities across regions in Taiwan. 1324 Taiwanese young adults aged 20–40 participated in this survey. Sociodemographics, health-related behaviors, social media use, perceived social support (including family, friends, and significant others), and anxiety symptoms were assessed. Linear regression models were used to examine the associations and interactions. Results: Of the sample, 21.4% exhibited SMA symptoms. The mean anxiety symptoms and social support scores were 6.7 ± 5.4 and 61.8 ± 14.2, respectively. SMA was associated with higher anxiety scores (adjusted coefficient [aB] = 2.02, 95% confidence interval (CI) = 1.35, 2.70; p < .001). Social support was associated with lower anxiety scores (aB = −0.06, 95% CI = −0.08, −0.04; p < .001). Among individuals with SMA, higher overall social support (aB = −0.05, 95% CI = −0.10, −0.01; p = .033), family support (aB = −0.14, 95% CI = −0.26 to −0.02; p = .025), and support from significant others (aB = −0.13, 95% CI = −0.25, −0.01; p = .047) were each associated with lower severity of anxiety symptoms. Conclusions: Overall social support, particularly support from family and significant others, is associated with reduced anxiety symptom levels and serves as a moderating factor in mitigating the adverse effects of social media addiction on anxiety symptom severity. Social support plays an important role in protecting young adults’ mental health from social media use.
  • No Thumbnail Available
    ItemMetadata only
    Multi-hop Video Super Resolution with Long-Term Consistency (MVSRGAN)
    (2023-01-01) Aditya W.; Shih T.K.; Thaipisutikul T.; Lin C.Y.; Mahidol University
    Utilizing deep learning, and especially Generative Adversarial Networks (GANs), for super-resolution images has yielded auspicious results. However, performing super resolutions with a big difference in scaling between input and output will add a certain degree of difficulty. In this paper we propose a super resolution with multiple steps, which means scaling the image gradually to stimulate maximum results. Video super resolution (VSR) needs different treatment from single image super resolution (SISR). It requires a temporal connection in between the frames, but this has not been fully explored by most of the existing studies. This temporal feature is significant to maintain the video consistency, in term of video quality and motion continuity. Using this loss functions, we can avoid the inconsistent failure in the image which accumulate continuously over time. Finally, our method has been shown to generate a super-resolution video that maintains both the video quality and its motion continuity. The quantitative result has higher Peak Signal to Noise Ratio (PSNR) scores for the Vimeo90K, Vid4, and Fireworks datasets with 37.70, 29.91, and 31.28 respectively compared to the state-of-the-art methods. The result shows that our models is better than other state-of-the-art methods using a different dataset.
  • No Thumbnail Available
    ItemMetadata only
    Multiple mediation analyses on exercise addiction and muscularity-oriented eating in young adults
    (2026-01-01) Tsai J.F.; Rudeejaroonrung K.; Chaimano S.; Efendi F.; Lin C.Y.; Lee C.T.; Ng A.K.; Paratthakonkun C.; Strong C.; Tsai M.C.; Tsai J.F.; Mahidol University
    Disordered eating and exercise behaviors may co-occur with muscle dysmorphia. This study investigates potential psychological mediators (psychological distress, weight self-stigma, drive for muscularity, drive for leanness) of the relationships between muscle dysmorphia and exercise addiction and muscularity-oriented eating in young Taiwanese individuals. We also examined whether these mediating effects differed by sex and sexual orientation. A cross-sectional sample of 1500 young adults (Mage = 22.3 years, 38.3 % male) participated in an anonymous online survey. We performed mediation analyses using AMOS to investigate the indirect effects of potential psychological mediators and multi-group analyses to examine the variation between males and females and between heterosexual and non-heterosexual individuals. We found that weight self-stigma, drive for muscularity, and drive for leanness were significant mediators, and these effects were invariant across sex and sexual orientation. Psychological distress, weight self-stigma, drive for muscularity, and drive for leanness mediated the relationship between muscle dysmorphia and muscularity-oriented eating. Males were more likely to report muscularity-oriented eating influenced by weight self-stigma and drive for leanness, and drive for leanness was more likely to facilitate heterosexual participants toward muscularity-oriented eating than non-heterosexual subjects. Successful interventions for disordered eating and exercise require an understanding of the underlying psychological and behavioral drivers.
  • No Thumbnail Available
    ItemMetadata only
    Multivariate time series analysis on variables that influence pandemic expansion
    (2022-01-01) Thaipisutikul T.; Lin C.Y.; Chen S.C.; Mahidol University
    The ongoing COVID-19 pandemic has wreaked havoc on social and economic systems worldwide. The variance in the rapidly increasing number of illnesses and deaths in each country is primarily due to national policies and actions. As a result, governments and institutions need to get insights into the critical factors influencing COVID-19 future case counts to properly manage the adverse effects of pandemics and promptly prepare appropriate measures. Thus, in this paper, we conduct extensive experiments on the real-world covid-19 datasets to examine the important factors influencing in the pandemic growth. In particular, we perform an exploratory data analysis to get the statistic and characteristics of multivariate time-series data on pandemic dynamic. Also, we utilize a statistical measure such as Pearson correlation to compute the relations of the past on the future daily new cases. The experimental results demonstrate that some restrictions have a positive effect on daily new confirmed cases at the early stage of the local pandemic transmission. Also, the results show that the early trend of COVID-19 can be explained well by human mobility in various categories. Thus, our proposed framework can be served as a guideline for future pandemic prevention and control decision-making.
  • No Thumbnail Available
    ItemMetadata only
    Novel Spatio-Temporal Continuous Sign Language Recognition Using an Attentive Multi-Feature Network
    (2022-09-01) Aditya W.; Shih T.K.; Thaipisutikul T.; Fitriajie A.S.; Gochoo M.; Utaminingrum F.; Lin C.Y.; Mahidol University
    Given video streams, we aim to correctly detect unsegmented signs related to continuous sign language recognition (CSLR). Despite the increase in proposed deep learning methods in this area, most of them mainly focus on using only an RGB feature, either the full-frame image or details of hands and face. The scarcity of information for the CSLR training process heavily constrains the capability to learn multiple features using the video input frames. Moreover, exploiting all frames in a video for the CSLR task could lead to suboptimal performance since each frame contains a different level of information, including main features in the inferencing of noise. Therefore, we propose novel spatio-temporal continuous sign language recognition using the attentive multi-feature network to enhance CSLR by providing extra keypoint features. In addition, we exploit the attention layer in the spatial and temporal modules to simultaneously emphasize multiple important features. Experimental results from both CSLR datasets demonstrate that the proposed method achieves superior performance in comparison with current state-of-the-art methods by 0.76 and 20.56 for the WER score on CSL and PHOENIX datasets, respectively.
  • No Thumbnail Available
    ItemMetadata only
    Nurses' intention and attitude to participate in advance care planning: An extended theory of planned behaviour using structural equation modelling–A cross-sectional study
    (2024-01-01) Apiradee A.; Lin C.Y.; Wirojratana V.; Lin P.C.; Shu B.C.; Apiradee A.; Mahidol University
    Aims: This study aimed to investigate the factors influencing nurses' intentions to participate in advance care planning (ACP) by examining the mediating roles of attitude, subjective norm, and perceived behavioural control in the relationship between knowledge and intention, using an extended theory of planned behaviour and structural equation modelling. Methods: A descriptive cross-sectional survey was conducted between January and April 2023, involving 515 registered nurses, selected through two-stage sampling. Data were collected using a self-administered online survey distributed via the internal communication system of hospital. Structural equation Modelling was employed to analyse the relationships among knowledge, attitude, subjective norm, perceived behavioural control and intention to participate in ACP. Results: The results supported two hypotheses regarding the relationships between knowledge, attitude, subjective norm, perceived behavioural control, and intention (p < 0.05). While the direct effect of knowledge on intention was not significant (β = 0.087, p = 0.292), the total indirect effect through attitude, subjective norm and perceived behavioural control was significant (β = 0.449, p < 0.001), accounting for approximately 83.77% of the total effect on intention. This underscores the critical role of these mediators in influencing nurses' intention to participate in ACP. Conclusions: This study highlights the significant indirect influence of knowledge on nurses' intentions to participate in ACP through attitude, subjective norms and perceived behavioural control. These findings suggest that targeted educational is needed to enhance ACP participation among nurses. Implications for the Profession and/or Patient Care: Understanding the role of attitude, subjective norm and perceived behavioural control can enhance nursing practice. Creating supportive environments and promoting interdisciplinary collaboration are crucial. Professional development through training, mentorship and role modelling can empower nurses in ACP. Comprehensive programs that increase knowledge and foster positive attitudes are essential for advancing ACP practice among nurses. Impact: Educational programs aimed at nurses should include components designed to strengthen knowledge and the identified mediators, equipping nurses with the necessary ACP skills. Organizational support through appropriate policy frameworks can facilitate these educational endeavours and ensure a sustainable impact on practice. Reporting Method: The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for cross-sectional studies.