Journal of Health Informatics in Developing Countries

Factors Affecting the Usage of Sehhaty Telehealth Services Among Young Adults in Riyadh

2025-08-31T09:58:16-04:00

Introduction:

Telehealth platforms, particularly during the COVID-19 pandemic, have significantly enhanced healthcare access worldwide. In Saudi Arabia, Sehhaty has emerged as a key digital health tool; however, user engagement among young adults remains inconsistent, potentially due to trust issues, usability challenges, and interoperability barriers.

Objectives:

This study aims to assess the influence of trust, perceived usefulness, ease of use, satisfaction, and system interoperability on young adults’ behavioral intention to use Sehhaty in Riyadh, using the Technology Acceptance Model (TAM) as a theoretical framework.

Method:

A cross-sectional survey was distributed online to Saudi residents aged 18–35 in Riyadh. A total of 314 responses were collected using convenience sampling. The survey included 30 Likert-scale items validated through previous TAM-based studies. Data were analyzed using descriptive statistics, Cronbach’s Alpha, Pearson correlation, and multiple linear regression.

Results:

All constructs showed good internal consistency (α > 0.75), and perceived usefulness (β = 0.75, p < 0.001) was the strongest predictor of behavioral intention, followed by ease of use and trust. Interoperability showed a moderate to low correlation with satisfaction (r = 0.40), and open-ended responses highlighted lack of integration with other public and private hospital systems as a common concern.

Conclusion:

While young users view Sehhaty positively overall, improving system interoperability and maintaining ease of use may significantly boost adoption. Findings highlight the importance of user-centered and integrated digital health strategies.

Enhancing Public Sector Decision-Making through Artificial Intelligence Models: A Comparative Study

2025-05-08T04:06:19-04:00

As governments worldwide embrace digital transformation, the role of artificial intelligence (AI) in public policy formulation and analysis has gained unprecedented relevance. This study explores the capabilities and limitations of two advanced AI models (customized ChatGPT and DeepSeek) as decision-support tools. Briefing notes were generated using three different approaches: one by human policy analyst and two by AI models. The aim was to evaluate whether contemporary natural language processing (NLP) technologies can produce briefing notes that are relevant and useful for public policy decision-making. The AI-generated content was tested through simulated policy scenarios to assess performance in tasks such as information retrieval, stakeholder-specific communication, policy brief generation, and scenario analysis. To ensure a robust evaluation, a panel of subject-matter experts assessed the quality of all briefing notes using a structured heuristic evaluation rubric. Results indicate that AI model can enhance analytical capacity, improve policy document drafting, and foster more responsive decision-making. However, the study also identifies critical challenges, including model bias, explainability deficits, and the need for sustained human oversight. Drawing the importance of hybrid governance frameworks that combine AI tools with institutional safeguards. The findings contribute to ongoing discussions on ethical AI integration and provide actionable recommendations for responsibly incorporating large language models into public sector workflows, especially in digitally transforming nations.

The Use of Explainable Deep Learning Models for Diabetic Retinopathy Management: From Screening to Severity Grading

2025-05-28T10:59:47-04:00

Background: Diabetic Retinopathy (DR), a leading cause of blindness, is rising globally due to increasing diabetes prevalence, especially in low-resource settings lacking diagnostic tools and specialists. Early DR is often asymptomatic; timely and scalable screening is essential. This study introduces a cost-effective two-stage deep learning framework for DR detection: Stage 1 performs binary classification on Optical Coherence Tomography (OCT) images, while Stage 2 grades DR severity use multi-class classification. Shapley Additive exPlanations (SHAP) enhance model transparency and clinical trust. Trained on public datasets and validated by a CHUK ophthalmologist, the approach provides a cost-effective solution for DR management in the underserved diabetic population.

Methods: The study followed five phases: (1) image dataset preparation using a four-step preprocessing pipeline: blurring, denoising, augmentation, and vector transformation for CNN input; (2) training and evaluating five pre-trained CNNs (MobileNetV3, DenseNet121, InceptionV3, Xception, VGG19) with standard metrics; (3) applying SHAP to interpret predictions of a multi-class model; (4) validating performance on 270 CHUK specialist-annotated images and 270 from the Brazil dataset; and (5) deploying the top models in a web application.

Results: The proposed model achieved strong performance in both DR screening and severity grading. DenseNet121 performed best, reaching 97% accuracy for referable DR detection and class accuracy of 97% (healthy), 92% (mild), 89% (moderate), 94% (severe), and 92% (proliferative). Validation datasets yielded Cohen’s Kappa scores of 0.771 and 0.680, demonstrating substantial agreement with expert grading. SHAP explanations enhanced interpretability across DR stages.

Conclusion: The applied preprocessing techniques improved image clarity and boosted model accuracy in grading DR severity. The proposed CNN framework enables end-to-end DR management. Combining expert validation, quality imaging, and interpretable deep learning offers a promising solution for early detection of DR and vision preservation in diabetic populations.