METHODS: The suggested methodology presents two examples, one with and one without picture augmentation. A balanced dataset meeting the same criteria in both cases is then generated using augmentative methods. The DenseNet-121-rendered model on the Asia Pacific Tele-Ophthalmology Society (APTOS) and dataset for diabetic retinopathy (DDR) datasets performed exceptionally well when compared to other methods for identifying the five stages of DR.
RESULTS: Our propose model achieved the highest test accuracy of 98.36%, top-2 accuracy of 100%, and top-3 accuracy of 100% for the APTOS dataset, and the highest test accuracy of 79.67%, top-2 accuracy of 92.%76, and top-3 accuracy of 98.94% for the DDR dataset. Additional criteria (precision, recall, and F1-score) for gauging the efficacy of the proposed model were established with the help of APTOS and DDR.
CONCLUSIONS: It was discovered that feeding a model with higher-quality photographs increased its efficiency and ability for learning, as opposed to both state-of-the-art technology and the other, non-enhanced model.
METHODS: Online focus group discussions were conducted among Malaysian parents to gather information about the content, layout and presentation of oral health information parents sought for the provision of oral health care for their children. Video recordings were transcribed verbatim and thematic analysis was performed using an inductive approach.
RESULTS: In total, 24 parents participated in the discussions and 4 main themes were uncovered. The first theme was perceived information needs related to dental caries, oral health care and the importance of deciduous teeth. The second theme was parents' preferred information resources which were social media, dentists, mobile phone applications and medical personnel. Thirdly, information delivery format and specific characteristics were recommended. The final theme was challenges and barriers faced in maintaining oral health due to parental constraints, child behaviour and external factors.
CONCLUSION: Parents' profound feedback and experiential standpoint stipulate the need for the development and delivery of a comprehensible and visually engaging oral health education module by healthcare professionals via social media to enable access to evidence-based information consistently.
METHODS: A retrospective observational study was conducted using a telehealth services database in Malaysian community pharmacies. Consultation records from January 2019 to December 2021 were extracted using a data collection form. The study identified the service usage over time, demographic profiles of users and the most common diagnoses and prescribed medications. Diagnoses were classified using the International Classification of Disease, 10th Revision (ICD-10), and medications were classified using the Anatomical Therapeutic Chemical (ATC) system.
RESULTS: The study included 835,826 telehealth service records, with 88.8% being assisted consultations with e-prescriptions and 11.2% direct consultations. The user population consisted of primarily Malaysians (96.9%), with a mean age of 50 ± 21 years. Both telehealth services saw an increase in unique users over the 3-year study period. There was a moderate correlation between active COVID-19 cases and monthly user count. Assisted consultations were more widely used than direct consultations.
CONCLUSION: This study found an increased usage of telehealth services and its potential to remain as a healthcare system feature in community pharmacies. Further investigation into the impact on medication safety, quality and healthcare delivery is warranted.
OBJECTIVE: Therefore, this study outlines the most recent developments and the current state of the art regarding evaluating the quality of online health information on web pages and specifies the direction of future research.
METHODS: In this article, a systematic literature is conducted according to the PRISMA statement in eight online databases PubMed, Science Direct, Scopus, ACM, Springer Link, Wiley Online Library, Emerald Insight, and Web of Science to identify all empirical studies that use machine and deep learning models for evaluating the online health information quality. Furthermore, the selected techniques are compared based on their characteristics, such as health quality criteria, quality measurement tools, algorithm type, and achieved performance.
RESULTS: The included papers evaluate health information on web pages using over 100 quality criteria. The results show no universal quality dimensions used by health professionals and machine or deep learning practitioners while evaluating health information quality. In addition, the metrics used to assess the model performance are not the same as those used to evaluate human performance.
CONCLUSIONS: This systemic review offers a novel perspective in approaching the health information quality in web pages that can be used by machine and deep learning practitioners to tackle the problem more effectively.
OBJECTIVES: This research aims to find the ideal number of best-hidden layers for the neural network and different activation function variations. The article also thoroughly analyzes how various frameworks can be used to create a comparison or fast neural networks. The final goal of the article is to investigate all such innovative techniques that allow us to speed up the training of neural networks without losing accuracy.
METHODS: A sample data Set from 2001 was collected by www.Kaggle.com. We can reduce the total number of layers in the deep learning model. This will enable us to use our time. To perform the ReLU activation, we will make use of two layers that are completely connected. If the value being supplied is larger than zero, the ReLU activation will return 0, and else it will output the value being input directly.
RESULTS: We use multiple parameters to determine the most effective method to test how well our method works. In the next paragraph, we'll discuss how the calculation changes secret-shared Values. By adopting 19 train set features, we train our reliable model to predict healthcare cost's (numerical) target feature. We found that 0.89503 was the best choice because it gave us a good fit (R2) and let us set enough coefficients to 0. To develop our stable model with this Set of parameters, we require 26 iterations. We use an R2 of 0.89503, an MSE of 0.01094, an RMSE of 0.10458, a mean residual deviance of 0.01094, a mean absolute error of 0.07452, and a root mean squared log error of 0.07207. After training the model on the train set, we applied the same parameters to the test set and obtained an R2 of 0.90707, MSE of 0.01045, RMSE of 0.10224, mean residual deviation of 0.01045, MAE of 0.06954, and RMSE of 0.07051, validating our solution approach. The objective value of our secured model is higher than that of the scikit-learn model, although the former performs better on goodness-of-fit criteria. As a result, our protected model performs quite well, marginally outperforming the (very optimized) scikit-learn model. Using a backpropagation algorithm and stochastic gradient descent, deep Learning develops artificial neural systems with several interconnected layers. There may be hidden layers of neurons in the network that have the tanh, rectification, and max-out hyperparameters. Modern features like momentum training, dropout, active learning rate, rate annealed, and L1 or L2 regularization provide exceptional prediction performance. The worldwide model's parameters are multi-threadedly (asynchronously) trained on the data from that node, and the model-based data is then gradually augmented by model averaging over the entire network. The method is executed on a single-node, direct H2O cluster initiated by the operator. The operation is parallel despite there just being a single node involved. The number of threads may be adjusted in the settings menu under Preferences and General. The optimal number of threads for the system is used automatically. Successful predictions in the healthcare data sets are made using the H2O Deep Learning operator. There will be a classification done since its label is binomial. The Splitting Validation operator creates test and training datasets to evaluate the model. By default, the settings of the Deep Learning activator are used. To put it another way, we'll construct two hidden layers, each containing 50 neurons. The Accuracy measure is computed by linking the annotated Sample Set with a Performer (Binominal Classification) operator. Table 3 displays the Deep Learning Model, the labeled data, and the Performance Vector that resulted from the technique.
CONCLUSIONS: Deep learning algorithms can be used to design systems that report data on patients and deliver warnings to medical applications or electronic health information if there are changes in the patient's health. These systems could be created using deep Learning. This helps verify that patients get the proper effective care at the proper time for each specific patient. A healthcare decision support system was presented using the Internet of Things and deep learning methods. In the proposed system, we examined the capability of integrating deep learning technology into automatic diagnosis and IoT capabilities for faster message exchange over the Internet. We have selected the suitable Neural Network structure (number of best-hidden layers and activation function classes) to construct the e-health system. In addition, the e-health system relied on data from doctors to understand the Neural Network. In the validation method, the total evaluation of the proposed healthcare system for diagnostics provides dependability under various patient conditions. Based on evaluation and simulation findings, a dual hidden layer of feed-forward NN and its neurons store the tanh function more effectively than other NN. To overcome challenges, this study will integrate artificial intelligence with IoT. This study aims to determine the NN's optimal layer counts and activation function variations.
METHODS: An integration of fuzzy logic and decision-making trial and evaluation laboratory (DEMATEL) is utilized, and data was collected from a panel of professional experts in Malaysia. Using a cause-effect relationship diagram, the fuzzy DEMATEL method evaluates the causal relationships between factors.
RESULTS: Findings showed that environmental factors play the most significant roles in preventing COVID-19 infection, followed by technology, individual, and social factors. Getting vaccinated is the most crucial factor in the environmental dimension in cutting the spread of COVID-19. Telehealth, the use of personal protective equipments (PPEs), and the adoption of social distancing are the most important measures in technology, individual and social dimensions, respectively.
CONCLUSIONS: This study offered valuable insights for policymakers and healthcare professionals in designing and implementing effective strategies to prevent pandemic disease transmission. Findings can be practically applied to optimize and prioritize infection prevention measures, assign resources more effectively, and guide evidence-based decision-making in the face of evolving pandemic situations. This process involves the active commitment of all parties, including governments, medical health executives, and citizens.
OBJECTIVES: The aims of this study were to (a) assess the feasibility and acceptability of measuring unified theory of acceptance and use of technology (UTAUT) constructs for DITETM with the Deaf community and Malaysian sign language (BIM) interpreters and (b) seek input from Deaf people and BIM interpreters on DITETM to improve its design.
METHODS: Two versions of the UTAUT questionnaire were adapted for BIM interpreters and the Deaf community. Participants were recruited from both groups and asked to test the DITE app features over a 2-week period. They then completed the questionnaire and participated in focus group discussions to share their feedback on the app.
RESULTS: A total of 18 participants completed the questionnaire and participated in the focus group discussions. Ratings of performance expectancy, effort expectancy, facilitating conditions and behavioural intention were high across both groups, and suggestions were provided to improve the app. High levels of engagement suggest that measurement of UTAUT constructs with these groups (through a modified questionnaire) is feasible and acceptable.
CONCLUSIONS: The process of engaging end users in the design process provided valuable insights and will help to ensure that the DITETM app continues to address the needs of both the Deaf community and BIM interpreters in Malaysia.
METHODS: This review protocol is registered in PROSPERO (CRD42017070194). Scientific databases including CINAHL Complete, MEDLINE, PsycINFO, SPORTDiscus, Cochrane Library and Scopus will be searched for relevant studies published from 1 January 2012 to the date the searches are conducted. Studies will be included if they incorporated adults who used an app or wearable for monitoring physical activity and/or sedentary behaviour; explored the barriers and/or facilitators of using an app and/or wearable; and were published in English. Following duplicate screening of titles and abstracts, full texts of potentially eligible papers will be screened to identify studies using qualitative approaches to explore barriers and facilitators of using apps and/or wearables for monitoring physical activity and/or sedentary behaviour. Discrepancies will be resolved through consensus or by consulting a third screener. Relevant excerpts (quotes and text) from the included papers will be extracted and analysed thematically. The Critical Appraisal Skills Programme Qualitative Research Checklist will be used to appraise included studies.
CONCLUSION: The results of this work will be useful for those intending to monitor physical activity and/or sedentary behaviour using these technologies.
OBJECTIVE: This study aims to design and develop a smartphone app called OASapp to improve medication adherence among older adult stroke survivors and evaluate its usability.
METHODS: OASapp was developed in a three-phase development process. Phase 1 is the exploration phase (including a cross-sectional survey, a systematic review, a search for stroke apps on the app stores of Apple App Store and Google Play Store, and a nominal group technique). In phase 2, a prototype was designed based on the Health Belief Model and Technology Acceptance Model. In phase 3, Alpha and Beta testing was conducted to validate the app.
RESULTS: Twenty-five features for inclusion in the app were collected in round one, and 14 features remained and were ranked by the participants during nominal group technique. OASapp included five core components (medication management, risk factor management, health information, communication, and stroke map). Users of OASapp were satisfied based on reports from Alpha and Beta testing. The mean Usability Metric for User Experience (UMUX) score was 71.4 points (SD 14.6 points).
CONCLUSION: OASapp was successfully developed using comprehensive, robust, and theory-based methods and was found to be highly accepted by users. Further research is needed to establish the clinical efficacy of the app so that it can be utilized to improve clinically relevant outcomes.
METHODS: A prospective cohort study involving 233 patients with high cardiovascular risk was conducted at a primary care clinic in Malaysia. Participants used a digital information diary tool to record online health information they encountered for 2 months and completed a questionnaire about statin necessity, concerns and adherence at the end of the observation period. Data were analysed using structural equation modelling.
RESULTS: The results showed that 55.8% (130 of 233 patients) encountered online health information. Patients who actively sought online health information (91 of 233 patients) had higher concerns about statin use (β = 0.323, p = 0.023). Participants with higher concern about statin use were also more likely to be non-adherent (β = -0.337, p
METHODOLOGY: This was a cross-sectional study, conducted among patients aged ≥ 18 years with cardiovascular risk factors attending a university primary care clinic. Patients were given the app to use for at least three months. Those who fulfilled the eligibility criteria were recruited. Data gathered were on sociodemographic, clinical characteristics, self-management support by doctors, utilisation of the app at home and social support in using the app. The previously translated and validated Malay version of the mHealth App Usability Questionnaire was used to measure usability. The mean usability score was calculated and linear regressions analysis was conducted to determine the factors associated with the usability of the app.
RESULTS: A total of 247 patients with at least one cardiovascular risk factor(s) were recruited. The mean age was 60.2 (±8.2). The majority were Malays (86.2%) and half of them were males (52.2%). The total mean (±SD) usability score was 5.26 (±0.67) indicating a high usability of the app. Usability of the app declined with increasing age in the simple linear regressions analysis. The multiple linear regressions yielded that being Malay (b = 0.31, 95% CI 0.08,0.54), using the app at home to understand their medications (b = 0.33, 95% CI 0.12,0.53) and having social support from family members and friends (b = 0.28, 95% CI 0.07,0.49) were significantly associated with higher usability of the app.
CONCLUSION: The usability of the EMPOWER-SUSTAIN Self-Management Mobile App© was high among patients with cardiovascular risk factors in our primary care clinic. This finding supports the widespread use of this app among our patients. Involvement of family members and friends should be encouraged to improve the usability of the app.
METHODS: The website development involved three stages: content analysis, web development, and validation. The model of Internet Intervention was used to guide the development of the website, in addition to other learning and multimedia theories. The content was developed based on literature reviews and clinical guidelines on hypertension. Then, thirteen experts evaluated the website using Fuzzy Delphi Technique.
RESULTS: The website was successfully developed and contains six learning units. Thirteen experts rated the website based on content themes, presentation, interactivity, and instructional strategies. All experts reached a consensus that the web is acceptable to be used for nutrition education intervention.
CONCLUSION: D-PATH is a valid web-based educational tool ready to be used to help disseminate information on dietary and physical activity to manage hypertension. This web application was suitable for sharing information on dietary and physical activity recommendations for hypertension patients.
METHODS: The study activities and timelines differ by site, with an extensive longitudinal evaluation being conducted at two sites and a basic evaluation being conducted at five sites. The learnings from the more comprehensive evaluations inform the iterative research and development processes while also ensuring ongoing evaluation of usability, acceptability and effectiveness of the app and its content across varying contexts. The study evaluates: (1) the impact of the Thrive by Five content on caregiver knowledge, behaviours, attitudes and confidence; (2) how the content changes relationships at the familial, community and system level; (3) how cultural and contextual factors influence content engagement and effectiveness and (4) the processes that facilitate or disrupt the success of the implementation and dissemination.
RESULTS: All in-country partners have been identified and data collection has been completed in Indonesia, Malaysia, Afghanistan, Kyrgyzstan, Uzbekistan, Namibia and Cameroon.
CONCLUSIONS: Very few digital health solutions have been trialled for usability and effectiveness in diverse cultural contexts. By combining quantitative, qualitative, process and ethnographic methodologies, this innovative study informs the iterative and ongoing optimisation of the cultural and contextual sensitivity of the Thrive by Five content and the processes supporting implementation and dissemination.
METHODS: A two-arm pilot randomized controlled trial was conducted. Twenty-four mothers of adolescent aged 10 to 14 years from a non-clinical sample were recruited online and randomly allocated into two groups (intervention [DaPI] and waitlist-control [WLC]). Eight weekly sessions were delivered online via technological devices. Feasibility outcomes were based on the participants' engagement in DaPI and study retention. Primary (parental behaviors and self-efficacy) and secondary (adolescent mental health) outcomes were assessed using an online survey at baseline (T0), post-intervention (T1), and 1-month follow-up (T2). Data were analyzed using descriptive and inferential statistics and an intention-to-treat approach.
RESULTS: The DaPI was well received by the mothers. Retention was high (81.8%) in both groups and intervention adherence was excellent (91.6%). Within-group analyses showed a significant decrease in physical control at T2 and an increase in parental self-efficacy at T1 and T2 among the DaPI mothers. No significant differences were observed in adolescents' mental health at any time point. As for the WLC group, there were no significant differences in all the outcome variables across the three assessment moments. Between groups analyses revealed DaPI mothers had significant differences in proactive parenting at T1, and in positive reinforcement and lax control at T2. There were no significant differences in adolescents' mental health between the groups at any time point.
DISCUSSION: The DaPI is feasible and acceptable in the Malaysian context. Findings show promise regarding the initial effects of the DaPI. However, a larger RCT is needed to determine its effectiveness in promoting mental health of adolescents.
TRIAL REGISTRATION: https://www.irct.ir/; identifier: IRCT20211129053207N1.
METHODS: This study employed an exploratory qualitative methodology to gather the perceptions of government-employed physiotherapists in Malaysia regarding the benefits, barriers, and recommendations for telerehabilitation in treating musculoskeletal disorders. The researchers conducted semistructured focus group discussions (FGDs) via Google Meet, which were recorded, transcribed, and analyzed using thematic analysis.
RESULTS: Five FGDs were conducted with 24 participants, 37.5% of whom had prior experience with telerehabilitation. The data analysis returned three main themes: (1) perceived benefits, (2) barriers, and (3) recommendations. Four subthemes were derived from perceived benefits: (1a) saving time and money, (1b) convenience, (1c) clients responsible for their treatment, and (1d) alternatives for infectious diseases. Perceived barriers revealed three subthemes: (2a) technology, (2b) organization, and (2c) personal barriers. Finally, participants provided recommendations for improving telerehabilitation services, including training programs to facilitate greater acceptance of this modality.
CONCLUSION: The findings of this study offer crucial insights into the evolving landscape of telerehabilitation in Malaysia. These findings revealed a greater prevalence of barriers to enablers among Malaysian physiotherapists, potentially influenced by varying experience levels. Despite the prevailing lack of experience among participants, this research underscores the significance of identifying barriers and enablers in implementing telerehabilitation with participants offering recommendations for integrating telerehabilitation into their practices. This study provides clear insights and a roadmap for stakeholders aiming to shape the future of telerehabilitation among physiotherapists in Malaysia.
METHODOLOGY: Digital photographs of 188 participants were taken using standardized parameters. The buccal gingival pigmentation was evaluated using three methods (a) a clinical evaluation by two independent assessors using the DOPI, (b) the CIELAB values using the Adobe Photoshop® software (Version 23.1.1) and (c) the CIV calculated using the ImageJ software (Version 1.53k). A hierarchical clustering analysis was used to identify colour groups that clustered together. Agreement between the clinical and digital categorization of the pigmentation was carried out using weighted kappa analysis. Agreements between CIELAB and CIV were compared using intra-class correlation coefficient.
RESULTS: There was a statistically significant difference in the DOPI, the L*, a*, and b* coordinates, and the CIV between the different ethnic groups of the participants. Cluster analysis for the CIELAB and CIV both identified four clusters. The gingival pigmentation categorization using the L*, a*, and b* values moderately agreed with the clinical evaluation using the DOPI index while the categorization with the CIV was in slight agreement with the clinical evaluations.
CONCLUSION: This study identified four clusters of gingival pigmentation in 188 multi-ethnic participants. The clusters, determined by CIELAB values, align with the clinical assessment of gingival pigmentation. Digital measurements derived from clinical photographs can serve as an effective means of pigmentation measurement in dental clinics.