Papers by InfoScience Trends InfoScience Trends
InfoScience Trends, 2024
The COVID-19 pandemic had a profound impact on individuals, organizations, and society. One conse... more The COVID-19 pandemic had a profound impact on individuals, organizations, and society. One consequence of this crisis was an increase in death anxiety among both the general population and patients. Anxiety disorders, including death anxiety, can be effectively treated using exposure therapy, which is a well-established method. To further support medical practitioners, developing mobilebased applications and content that focus on this treatment approach would be beneficial. With this in mind, the objective of the present study was to develop a prototype for a proposed mobile application aimed at alleviating the burden of COVID-19 death anxiety. Our research adopts a structured approach grounded in the five essential phases of high-fidelity prototype design. These methodological steps are as follows: 1) Goal Definition, involving meticulous planning and explicit delineation of the primary purpose, alongside identifying materials utilized in the prototype. 2) User Interface Design, entailing the creation of diverse interface designs to discern and select the optimal design. 3) Adding Interactions, incorporating interactive elements such as clicking, dragging, scrolling, and user input into the prototype. 4) Testing and Evaluation, comprising prototype evaluation and systematic feedback collection. 5) Iteration and Improvement, where the prototype undergoes refinement based on the conclusive feedback garnered during the evaluation phase, aiming to attain the desired prototype. This methodological framework ensures a comprehensive and systematic approach to the development and enhancement of our high-fidelity prototype. In the current challenging time of the COVID-19 pandemic, the identified criteria for content production and the proposed prototype serve as valuable resources for software designers aiming to design and develop suitable applications to alleviate anxiety related to COVID-19 death. By following these criteria and utilizing the prototype as a guide, software designers can create applications that effectively address the anxieties and concerns of individuals during this difficult period.
InfoScience Trends, 2025
Artificial intelligence (AI) has emerged as a transformative technology in healthcare, offering i... more Artificial intelligence (AI) has emerged as a transformative technology in healthcare, offering innovative solutions for blood pressure management and control. This article explores the potential impact of AIenhanced health tools in revolutionizing the approach to managing blood pressure. Various AI models, including machine learning algorithms, deep learning techniques, natural language processing, reinforcement learning, and Bayesian networks, are utilized to analyze data, predict outcomes, and provide personalized recommendations for individuals. These AI models have the capability to extract insights from complex datasets, identify patterns, and tailor interventions based on individual needs and preferences. Despite the promising potential of AI in blood pressure management, several challenges must be addressed. Data quality and privacy concerns, interpretability and transparency of AI algorithms, bias and fairness in decision-making, regulatory and ethical considerations, integration and adoption issues, as well as validation and performance assessments pose significant hurdles in the implementation of AI-enhanced health tools. Overcoming these challenges requires a collaborative effort among healthcare providers, data scientists, ethicists, regulators, and policymakers to ensure the safe, effective, and ethical use of AI in healthcare settings. By addressing these challenges proactively and leveraging the power of AI, healthcare providers can optimize treatment strategies, improve patient outcomes, and empower individuals to take control of their health. The integration of AI in blood pressure management has the potential to transform healthcare delivery, enhance personalized care, and ultimately contribute to better health outcomes for individuals.
InfoScience Trends, 2024
This study examines the efficiency of health-oriented APIs in Iran, analyzing their performance a... more This study examines the efficiency of health-oriented APIs in Iran, analyzing their performance across various categories. Using a combined approach of Data Envelopment Analysis (DEA), machine learning techniques, and statistical analysis, we evaluated 149 APIs to determine their efficiency scores and identify areas for improvement.
The DEA analysis revealed that many APIs, particularly those in the "Health and Wellness" and "Genetic Data" categories, operate at high-efficiency levels. The scores were calculated using an input variable derived from Principal Component Analysis (PCA) and Exploratory Factor Analysis (EFA), while the output was determined using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The Kruskal-Wallis test showed significant differences in efficiency scores among the macro-categories, with "Clinical and Patient Management" demonstrating notable superiority. Pairwise comparisons confirmed these differences, indicating the need for improvement in some categories.We applied a k-means clustering algorithm to classify the APIs into efficiency gradients. Validation through logistic regression confirmed the significant influence of categories on efficiency, supported by SHAP analysis. The results suggest that "Patient Management" APIs are the most efficient.
Future implications include optimizing less efficient APIs and adopting more advanced techniques. These findings provide valuable guidance for improving technological performance and optimizing efficiency in the healthcare sector, contributing to a more innovative and responsive system.
InfoScience Trends, 2025
Information science and data science are two distinct yet interrelated academic fields that often... more Information science and data science are two distinct yet interrelated academic fields that often lead to confusion regarding their scope and application. While they share a common goal of deriving meaningful insights from data, they differ significantly in their methodologies and theoretical foundations. Information science primarily focuses on the management, organization, and retrieval of information. It has various topics, including information architecture, knowledge management, and the design of information retrieval systems. Professionals in this field work to ensure that information is accessible, accurate, and secure, playing a crucial role in the maintenance of databases and digital archives. In contrast, data science is centered around extracting insights from large and complex datasets using statistical analysis, machine learning algorithms, and data visualization techniques. This interdisciplinary field draws from mathematics, computer science, and domain-specific knowledge to study diverse problems and generate innovative solutions. Data scientists analyze patterns within data to inform decision-making processes across various industries. This article explores the nuanced differences between information science and data science while also highlighting their interconnectedness. Ultimately, it posits that data science has a broader scope that includes information science alongside other relevant disciplines. By understanding these distinctions and overlaps, professionals can better navigate their roles within these evolving fields.
InfoScience Trends, 2025
The incidence and costs associated with pressure ulcer (PU) treatment remain high, despite the im... more The incidence and costs associated with pressure ulcer (PU) treatment remain high, despite the implementation of preventive measures. Recent literature suggests that information technology (IT) can enhance the effectiveness of PU prevention strategies. This systematic review aims to identify the current state of IT-based approaches for supporting PU prevention. A comprehensive literature review was conducted to analyze IT approaches in PU prevention. We searched the PubMed and ScienceDirect databases using a predetermined search string to identify relevant primary studies, selecting those that met established inclusion criteria. A total of 22 articles fulfilled the inclusion criteria. Most of these approaches focus on patient monitoring, providing critical information on pressure exposure, temperature, humidity levels, and estimated body position in bed. These systems often generate risk factor intensity charts and mapping visualizations. The predominant methods for preventing pressure ulcers include clinical decision support systems and telemedicine. Additionally, the integration of machine learning technologies has significantly advanced PU prevention efforts. However, no experimental studies have conclusively demonstrated that these approaches effectively reduce PU incidence. While the reviewed technologies show moderate effectiveness in addressing certain mediating factors, there is a need for further development of multifactorial technological solutions akin to those used in managing other chronic conditions. We recommend integrating monitoring, support, and feedback mechanisms to enhance situational awareness, adherence to preventive protocols, and access to professional resources in the fight against pressure ulcers.
InfoScience Trends, 2025
As artificial intelligence (AI) systems become increasingly integrated into healthcare workflows,... more As artificial intelligence (AI) systems become increasingly integrated into healthcare workflows, debates have emerged about whether these tools will serve merely to augment physician capacities or whether they may eventually replace elements of human clinical decision-making. While international studies have investigated these questions, little is known about how physicians in Middle Eastern contexts, particularly Iran, view this evolving role of AI. This study aimed to explore Iranian physicians’ perspectives on artificial intelligence in medicine, focusing on whether AI is perceived as a supportive augmentation tool or a potential substitute for core clinical functions. A qualitative design was employed using semi-structured interviews with 20 Iranian physicians across multiple specialties. Interviews were thematically analyzed to identify patterns in perceptions, concerns, and expectations surrounding AI adoption. Five major themes emerged: (1) AI as a clinical augmentation tool, (2) skepticism toward full replacement, (3) shifting professional identity, (4) challenges of trust and explainability, and (5) unmet infrastructural and educational needs. While AI was generally viewed as beneficial for diagnostic support and efficiency, concerns persisted about the loss of clinical autonomy, deskilling, and the lack of interpretability in AI systems. Iranian physicians largely view AI as a complement rather than a replacement for human expertise. Successful integration will require attention to ethical, cultural, and infrastructural contexts, as well as targeted training and regulatory frameworks.
InfoScience Trends, 2025
Artificial intelligence (AI) is transforming cardiac care by enhancing diagnosis, risk stratifica... more Artificial intelligence (AI) is transforming cardiac care by enhancing diagnosis, risk stratification, and patient monitoring. This systematic review synthesizes evidence from 14 studies (2020–2025) on the economic impact, clinical outcomes, and implementation challenges of AI in cardiology. Findings demonstrate that AI-driven interventions—including machine learning-guided atrial fibrillation screening, AI-enhanced cardiac imaging, and remote monitoring—improve disease detection rates and reduce adverse events (e.g., strokes, hospitalizations) while proving cost-effective. For instance, targeted AI screening identified 27–45% more atrial fibrillation cases versus standard care, with incremental cost-effectiveness ratios favoring AI adoption. However, real-world implementation faces barriers such as electronic health record integration costs, clinician adoption resistance, and workflow disruptions. Facilitators like phased rollouts, embedded decision support tools, and real-world performance tracking mitigated these challenges. Methodological limitations include study heterogeneity, reliance on model-based economic analyses, and underreporting of long-term implementation costs. The review highlights a critical gap between AI's theoretical benefits and its practical deployment, emphasizing the need for pragmatic trials, standardized outcome reporting, and stakeholder collaboration. By addressing these challenges, AI can realize its potential to enhance cardiac care efficiency and patient outcomes.
InfoScience Trends, 2025
Endometriosis diagnosis via laparoscopy remains challenging due to subtle lesion appearances and ... more Endometriosis diagnosis via laparoscopy remains challenging due to subtle lesion appearances and inter-observer variability. While artificial intelligence shows promise for surgical video analysis, the potential of Vision Transformers (ViTs) specifically for endometriosis detection remains unexplored. This study applied a SWOT framework to evaluate ViTs for automated endometriosis diagnosis in laparoscopic videos. Analysis of 10 studies from PubMed, IEEE Xplore, and Scopus identified key findings: Strengths included (1) global attention for lesion detection, (2) outperforming CNNs/RNNs in surgical tasks (91-97% accuracy), and (3) multimodal data integration. Weaknesses were (1) dependence on unavailable annotated datasets, (2) high computational needs, (3) limited local feature sensitivity, and (4) annotation variability issues. Opportunities involved (1) self-supervised learning from unlabeled videos and (2) explainable attention maps. Threats comprised (1) performance variability across surgical settings, (2) lacking regulatory standards, and (3) data privacy concerns. Crucially, no studies directly tested ViTs for endometriosis diagnosis despite their potential. For clinical implementation, three requirements emerged: (1) collaborative dataset creation, (2) optimized hybrid architectures, and (3) ethical guidelines for surgical AI. This
InfoScience Trends, 2025
Postoperative length of stay serves as a crucial metric for managing medical resources and offers... more Postoperative length of stay serves as a crucial metric for managing medical resources and offers an indi-rect measure of surgical complication rates following cancer surgery. In recent years, machine learning techniques have been increasingly utilized for complex medical outcome prediction by leveraging large volumes of clinical data. In this study, we employed machine learning models to forecast prolonged length of stay after cancer surgery using de-identified clinical data from The Cancer Genome Atlas (TCGA), a publicly accessible database. We analyzed records from approximately 3,500 patients who underwent primary surgeries for five types of cancer-stomach, breast, colon, thyroid, and lung. The available clinical variables included surgical details, cancer characteristics, comorbidities, and laboratory assessments. Support vector machine, random forest, and logistic regression models were developed to predict pro-longed postoperative length of stay. The results show that machine learning models can predict prolonged hospital stay for breast and stomach cancers (AUC >0.85) but perform poorly for colon and lung cancers (AUC <0.8), highlighting limitations in generalizability. These findings highlight both the potential and the limitations of using public genomic and clinical datasets for perioperative risk stratification and resource allocation in cancer care.
InfoScience Trends, 2025
Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into clinical ... more Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into clinical workflows, yet evidence comparing their real-world effectiveness remains fragmented. This review systematically evaluates AI/ML methods deployed in healthcare, focusing on implementation strategies, validation rigor, and performance metrics. To identify the most frequently implemented AI/ML techniques, assess their clinical success rates, and analyze workflow integration challenges across specialties. We reviewed PubMed articles (2019-2024) describing AI/ML clinical applications with quantitative outcomes. Ten studies met inclusion criteria, covering radiology, oncology, and pediatrics. Data were extracted on AI methods, validation types, performance metrics (e.g., sensitivity, AUC), and workflow integration. Descriptive statistics summarized findings. Logistic regression and deep learning (e.g., atlasmatching) were the most specified methods. Logistic regression achieved 71% sensitivity and 77% PPV in epilepsy screening, matching clinician performance. Deep learning models showed >90% retrospective acceptability in radiotherapy planning but lacked prospective metrics. Only 40% of studies reported quantitative outcomes; others emphasized usability or frameworks. Workflow integration (e.g., EHR embedding) was critical but inconsistently detailed. While both traditional and advanced AI methods demonstrate clinical utility, heterogeneous reporting and limited head-to-head comparisons hinder definitive conclusions. Future research should prioritize standardized performance metrics and prospective multi-method evaluations to guide evidence-based adoption.
InfoScience Trends, 2025
Artificial intelligence (AI) and machine learning (ML) have revolutionized diabetes management th... more Artificial intelligence (AI) and machine learning (ML) have revolutionized diabetes management through glucose prediction and decision-support systems. However, concerns persist about algorithmic bias and demographic disparities in these technologies, particularly across racial, ethnic, and socioeconomic subgroups. This systematic review evaluates the equity of AI-based glucose prediction models, focusing on performance disparities, fairness reporting, and physiological label biases. We conducted a systematic review following the PRISMA 2020 guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Searches were performed in PubMed, Scopus, and Google Scholar using keywords related to AI, diabetes management, and health disparities. Inclusion criteria encompassed studies examining AI/ML models for glucose prediction or nutrition recommendations in diabetes, with a focus on racial, ethnic, or socioeconomic disparities. Data extraction and quality assessment were performed independently by two reviewers. Among 1,243 initially identified articles, only 10 met inclusion criteria. The review revealed limited evidence of subgroup performance disparities, with just one study explicitly evaluating racial differences in AI model performance. Fairness reporting was rare, with only 7% of AI diabetes studies documenting ethnoracial data and virtually none conducting fairness audits. Physiological and label biases, such as HbA1c discrepancies between racial groups, were documented but unaddressed in AI model development. AI-based diabetes technologies lack robust equity evaluation, with minimal reporting on subgroup performance and fairness. Without systematic bias mitigation and equitable design, these tools risk exacerbating existing health disparities. Future research must prioritize transparency, representativeness, and fairness to ensure equitable benefits for all populations.
InfoScience Trends, 2025
This paper evaluates the application of Convolutional Neural Networks (CNNs) for detecting sponta... more This paper evaluates the application of Convolutional Neural Networks (CNNs) for detecting spontaneous pneumothorax in chest X-rays. For this study, we randomly selected 2000 chest X-ray images with pneumothorax from the publicly available National Institutes of Health (NIH) database, which were subsequently allocated to training and testing datasets. To enhance the model's generalization, the images were preprocessed by normalizing, resizing, and augmenting the dataset. The proposed CNN model includes multiple convolutional layers that extract low-level features, followed by max-pooling layers for progressive dimensionality reduction. To combine the extracted features and make predictions, fully connected layers are employed, and a hyperbolic tangent (tanh) activation function is used in the output layer for binary classification. The Adam optimizer is utilized to train the model, and its performance is assessed using standard performance metrics, including accuracy, precision, sensitivity, specificity, and specificity. The results show that the developed CNN-based method outperforms conventional approaches based on machine learning, such as Decision Tree, Random Forest, and SVM.
InfoScience Trends, 2025
Generative Artificial Intelligence (AI) has emerged as a potential decision-support tool in traum... more Generative Artificial Intelligence (AI) has emerged as a potential decision-support tool in trauma and fracture surgery, yet its efficacy in real-world clinical contexts remains underexplored. This study systematically identified critical information needs of orthopedic trauma surgeons and evaluated the performance of five leading AI models (ChatGPT-4, Gemini, Claude, Med-PaLM, and BioGPT) in addressing these needs. Through a three-phase approach—literature review, AI querying, and blinded expert evaluation—we assessed accuracy, relevance, hallucination rates, and response times. Results indicated ChatGPT-4 outperformed other models in accuracy (mean score: 1.80/2) and relevance (4.40/5), with lower hallucination rates (10%). Med-PaLM and BioGPT also demonstrated strong performance, while Gemini and Claude lagged. All models provided responses within clinically acceptable timeframes (<30 seconds). Findings suggest that while AI can offer rapid, guideline-concordant support, human oversight remains essential due to inconsistencies and hallucination risks. Future integration should focus on workflow-embedded, validated systems with continuous monitoring.
InfoScience Trends, 2025
Cardiovascular imaging has witnessed transformative advancements through artificial intelligence ... more Cardiovascular imaging has witnessed transformative advancements through artificial intelligence (AI) and machine learning (ML), yet the integration of these technologies across multiple modalities remains underexplored. This systematic review synthesizes evidence on AI/ML approaches for automated interpretation across echocardiography, cardiac computed tomography (CT), and cardiac magnetic resonance imaging (MRI), with a focus on clinically relevant tasks. Following PRISMA guidelines, we conducted a PubMed search (2017–2025), identifying 1,091 records. After rigorous screening, six studies (five reviews, one empirical) met inclusion criteria. Our analysis revealed that while deep learning dominates cardiac CT and MRI applications—particularly for segmentation and disease classification—echocardiography lags in empirical validation. The sole empirical study benchmarked AI performance across CT and MRI but excluded echocardiography, highlighting a critical gap in multi-modality research. Review articles consistently emphasized the potential of AI but noted persistent challenges, including data standardization, external validation, and clinical integration. Key limitations include the predominance of single-modality studies and a lack of head-to-head comparisons between traditional ML and deep learning methods. Future research should prioritize unified frameworks encompassing all three modalities, robust clinical validation, and standardized performance metrics to bridge these gaps. This review underscores AI’s transformative potential in cardiac imaging while advocating for more comprehensive, clinically grounded studies to realize its full impact.
InfoScience Trends, 2025
Breast cancer remains a leading cause of cancer-related mortality among women globally, emphasizi... more Breast cancer remains a leading cause of cancer-related mortality among women globally, emphasizing the need for early and accurate detection. This study combines a systematic review of artificial intelligence (AI) applications in breast cancer imaging with empirical benchmarking of deep learning models across public datasets. The review analyzed 21 studies, highlighting convolutional neural networks (CNNs) as the dominant AI approach, with mammography being the most studied modality. Benchmarking involved evaluating a baseline CNN, ResNet-50, and U-Net on datasets including DDSM (mammography), INbreast (mammography), and BUSI (ultrasound). Results demonstrated that ResNet-50 significantly outperformed the baseline CNN in classification tasks, with a mean AUC improvement of 0.073 (p = 0.023). U-Net achieved robust segmentation performance on BUSI, particularly for malignant lesions (Dice coefficient = 0.87). The study underscores the superiority of transfer learning and deeper architectures in breast cancer imaging while identifying gaps in multimodal integration and explainability. Future directions include expanding to multimodal datasets, incorporating interpretability tools, and validating models in real-world settings. These findings contribute to the growing body of evidence supporting AI's role in enhancing breast cancer diagnostics and pave the way for clinically actionable solutions.
InfoScience Trends, 2025
The effective performance of emergency medical service (EMS) teams is crucial in providing timely... more The effective performance of emergency medical service (EMS) teams is crucial in providing timely and life-saving medical assistance during emergencies. This study aimed to analyze the performance of EMS teams in Iran, specifically focusing on Shahid Beheshti Hospital in Babol, using a social network analysis (SNA) approach. Data was collected by identifying and surveying EMS team members to understand their communication patterns and relationships. Interviews were conducted with key individuals to gain qualitative insights. Observations were made during emergency response simulations and real-life emergency situations. SNA measures, including degree centrality, betweenness centrality, and closeness centrality, were employed to assess the team's social network structure and dynamics. The findings reveal a high disruption between team members and a relatively low network density, with only 24.3% of possible connections established. Notably, the Emergency Medical Dispatch (EMD) group exhibits the highest degree of centrality, followed by Emergency Medical Technicians (EMTs) and Paramedics. These results underscore challenges in communication and cooperation within the teams, with low density contributing to inconsistency and potential performance issues. Using social network analysis (SNA), this study evaluates communication and coordination patterns within EMS teams in Iran, identifying critical gaps in network density and centrality. These findings offer actionable strategies to enhance teamwork and emergency response efficiency. This research contributes to the field of emergency medicine and provides valuable insights for healthcare professionals and policymakers in Iran and beyond.
Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by social communic... more Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by social communication deficits and repetitive behaviors, with early diagnosis being critical for effective intervention. Traditional diagnostic methods rely on subjective behavioral assessments, often delaying identification. Recent advances in neuroimaging, particularly resting-state functional magnetic resonance imaging (rs-fMRI), offer a promising avenue for objective and early ASD detection by capturing atypical functional connectivity patterns in the brain. This study leverages the temporal dynamics of rs-fMRI data to classify ASD in young children aged 5–10 years using Bidirectional Long Short-Term Memory (BiLSTM) neural networks. The model processes rs-fMRI time-series bidirectionally, capturing both past and future contextual information to identify ASD-related connectivity alterations. Training and testing were conducted on the publicly available Autism Brain Imaging Data Exchange (ABIDE) I dataset, with performance evaluated through stratified 10-fold cross-validation. The proposed BiLSTM model achieved an accuracy of 76.91%, outperforming comparable Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Convolutional Neural Network (CNN) architectures. Sensitivity and specificity were 75.31% and 78.35%, respectively, highlighting the model's balanced performance in identifying both ASD and typically developing controls. These results underscore the importance of temporal modeling in rs-fMRI analysis, as spatial methods like CNNs yielded lower accuracy (71.88%). The study demonstrates the potential of BiLSTMs for pediatric ASD classification while acknowledging the need for further refinement to bridge the gap with higher-performing hybrid models reported in the literature.
InfoScience Trends, 2025
AI-driven psychological interventions offer scalable solutions for chronic illness patients in re... more AI-driven psychological interventions offer scalable solutions for chronic illness patients in resource-limited settings, yet evidence gaps persist in low-income countries regarding efficacy, ethics, and implementation. A mixed-methods study (systematic review, survey of 122 Iranian health professionals, and 18 qualitative interviews) evaluated AI’s role in psychological support. The review analyzed 10 studies (2020–2025), while survey and interview data explored practitioner perceptions. The review identified efficacy in middle-income settings (e.g., chatbots reduced distress in Egyptian breast cancer patients [RCT, n=150]), but no studies in low-income countries. Survey results revealed high optimism for AI’s accessibility (90.2%) but significant concerns about privacy (76.2%) and cultural relevance (61.5%). Qualitative themes highlighted demands for dialect-adapted NLP, human escalation protocols, and integration with community health workers. AI tools show promise for expanding mental health care, but their success hinges on contextual adaptation, ethical safeguards, and infrastructure investment. Policymakers must address digital literacy and regulatory gaps to enable equitable deployment.
InfoScience Trends, 2025
Substance abuse poisoning remains a critical global public health challenge, necessitating robust... more Substance abuse poisoning remains a critical global public health challenge, necessitating robust surveillance systems to inform prevention and treatment strategies. This systematic review evaluates the design, implementation, and outcomes of substance abuse poisoning registries worldwide, with a focus on their health informatics infrastructure and technological innovations. We analyzed 8 studies and institutional reports, identifying 8 major registry systems across 6 countries (Canada, USA, Mexico, Norway, Germany, Malaysia). Key findings reveal substantial variability in registry architectures, from Canada's comprehensive Drug and Alcohol Treatment Information System (DATIS) to Malaysia's GIS-based hot-spot mapping. Technological approaches ranged from web-based platforms (Norway's Java/MySQL system) to real-time SMS alerts (Germany's poison center network). Registries demonstrated measurable impacts, including improved treatment tracking (DATIS), enhanced spatial analysis of abuse patterns (Malaysia), and faster emergency response (Germany). However, critical gaps persist, particularly in data interoperability and integration with mental health records. The review highlights how informatics solutions – including standardized data models, geospatial technologies, and mobile health applications – can address surveillance challenges in low-resource settings. For countries lacking robust systems (e.g., Iran), we propose a hybrid framework combining Canada's clinical data standards, Malaysia's GIS capabilities, and Germany's notification protocols. These findings underscore the transformative potential of health informatics in substance abuse surveillance while identifying key priorities for future research, including AI-powered predictive modeling and blockchain-based data sharing.
InfoScience Trends, 2025
Artificial intelligence (AI) has emerged as a transformative tool in pediatric oncology imaging, ... more Artificial intelligence (AI) has emerged as a transformative tool in pediatric oncology imaging, enhancing diagnostic accuracy, prognostic evaluation, and treatment monitoring. This systematic review synthesizes evidence from 22 studies to evaluate AI applications—including machine learning (ML) and deep learning (DL)—in tumor classification, segmentation, radiogenomics, and treatment response assessment. Key findings reveal that convolutional neural networks (CNNs) and radiomics pipelines achieve expert-level performance in classifying pediatric brain tumors (e.g., medulloblastoma, pilocytic astrocytoma) with AUCs >0.95 and Dice scores up to 0.96 for segmentation tasks. AI also shows promise in predicting molecular markers (e.g., MYCN, BRAF) and automating longitudinal tumor volume measurements using frameworks like RAPNO. However, challenges persist, such as data scarcity due to the rarity of pediatric cancers, heterogeneity in imaging protocols, and limited external validation. Ethical concerns regarding data privacy and model interpretability further hinder clinical adoption. Multi-institutional collaborations (e.g., Children’s Brain Tumor Network) and explainable AI (XAI) tools (e.g., Grad-CAM) are proposed to address these limitations. Future research should prioritize large-scale, prospective studies, standardized reporting frameworks (e.g., TRIPOD-AI), and integration of AI into clinical workflows. While AI demonstrates significant potential to revolutionize pediatric oncology imaging, overcoming current barriers is essential for robust, generalizable, and ethically sound implementations.
Uploads
Papers by InfoScience Trends InfoScience Trends
The DEA analysis revealed that many APIs, particularly those in the "Health and Wellness" and "Genetic Data" categories, operate at high-efficiency levels. The scores were calculated using an input variable derived from Principal Component Analysis (PCA) and Exploratory Factor Analysis (EFA), while the output was determined using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The Kruskal-Wallis test showed significant differences in efficiency scores among the macro-categories, with "Clinical and Patient Management" demonstrating notable superiority. Pairwise comparisons confirmed these differences, indicating the need for improvement in some categories.We applied a k-means clustering algorithm to classify the APIs into efficiency gradients. Validation through logistic regression confirmed the significant influence of categories on efficiency, supported by SHAP analysis. The results suggest that "Patient Management" APIs are the most efficient.
Future implications include optimizing less efficient APIs and adopting more advanced techniques. These findings provide valuable guidance for improving technological performance and optimizing efficiency in the healthcare sector, contributing to a more innovative and responsive system.