The landscape of Artificial Intelligence Driven Digital Platforms in Healthcare and Life Sciences: A Scoping Review Towards Universal Integration (Preprint) DOI

Rahela Penovski,

Stanko Srčič,

Reinhold Scherer

и другие.

Опубликована: Апрель 7, 2025

BACKGROUND Digital platforms are transforming healthcare and life sciences by enhancing operational efficiency, improving patient outcomes, accelerating product development market access. These facilitate telemedicine, remote monitoring, data-driven care delivery, while in sciences, they streamline research, optimize manufacturing, support regulatory compliance. However, current Artificial Intelligence (AI) based solutions remain fragmented domain-specific, lacking the integration necessary to address complex challenges within across these industries. While AI excels specialized areas—such as radiology image analysis, sepsis-detection predictive analytics, AI-driven drug discovery—these often operate isolation. OBJECTIVE This scoping review aims critically analyze landscape of digital identify existing gaps opportunities, explore feasibility developing a comprehensive, universal AI-based platform. METHODS A was conducted following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews Meta-Analyses extension Scoping Reviews) guidelines. comprehensive search peer-reviewed literature, industry reports, documents performed databases including PubMed, Embase, CINAHL, PsycINFO, Scopus, IEEE Xplore, Web Science. Inclusion criteria focused on studies, published from 2014 2024, addressing platforms, integration, science applications, regulations, professional ethics. Data were charted synthesized themes related platform capabilities, gaps, potential. RESULTS The identified seven common types utilized alongside growing trend implementation AI, generative AI. Significant found interoperability, cross-sector collaboration, utilization. technologies showed promise discovery diagnostic accuracy reducing costs. Despite advances, differences data standards, constraints, proprietary formats hinder seamless with electronic health record systems workflows. Moreover, applications neglect personalized treatment plans that integrate genomics, lifestyle factors, comorbidities, social determinants health. Organizations like World Health Organization, Massachusetts Institute Technology, Microsoft exploring integrated solutions, but standardization, privacy, regulatory, ethical considerations remain. CONCLUSIONS could revolutionize research innovation. By fragmentation fostering sectors, such enable continuous improvement innovation delivery development. overcoming compliance, standards is critical. Collaborative efforts sectors essential develop connected, responsive, effective national global systems. Future should focus defining platform's structure scope, design strategies, aligning ethics ensure equitable integration.

Язык: Английский

An Efficient Method for Lung Lesions Classification Using Automatic Vascularization Evaluation on Color Doppler Ultrasound DOI Creative Commons
Roxana Both,

Adrian Satmari,

Romeo Chira

и другие.

Applied Sciences, Год журнала: 2025, Номер 15(5), С. 2851 - 2851

Опубликована: Март 6, 2025

Lung cancer still represents one of the main causes cancer-related mortality, highlighting necessity for precise, effective, and minimally intrusive diagnostic methods. This research presents an innovative approach to classifying lung lesions using Doppler ultrasound imagery combined with a feed-forward neural network (FNN). study integrates mode vascularization features—blood vessel area, tortuosity index, orientation—into FNN classify as benign or malignant. A dataset 565 pictures was extended augmentation techniques enhance robustness, yielding training 3390 images. The architecture trained utilizing Levenberg–Marquardt algorithm, achieving classification accuracy 98%, demonstrating its potential aid. results indicate that integrating all three factors significantly improves diagnosis compared individual modules. method offers non-invasive cost-effective complementary tool conventional such CT scans, improve early detection treatment planning patients.

Язык: Английский

Процитировано

0
Atrous spatial pyramid pooling with swin transformer model for classification of gastrointestinal tract diseases from videos with enhanced explainability DOI Creative Commons

Arefin Ittesafun Abian,

Mohaimenul Azam Khan Raiaan, Mirjam Jonkman

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 150, С. 110656 - 110656

Опубликована: Март 25, 2025

Язык: Английский

Процитировано

0
The landscape of Artificial Intelligence Driven Digital Platforms in Healthcare and Life Sciences: A Scoping Review Towards Universal Integration (Preprint) DOI

Rahela Penovski,

Stanko Srčič,

Reinhold Scherer

и другие.

Опубликована: Апрель 7, 2025

BACKGROUND Digital platforms are transforming healthcare and life sciences by enhancing operational efficiency, improving patient outcomes, accelerating product development market access. These facilitate telemedicine, remote monitoring, data-driven care delivery, while in sciences, they streamline research, optimize manufacturing, support regulatory compliance. However, current Artificial Intelligence (AI) based solutions remain fragmented domain-specific, lacking the integration necessary to address complex challenges within across these industries. While AI excels specialized areas—such as radiology image analysis, sepsis-detection predictive analytics, AI-driven drug discovery—these often operate isolation. OBJECTIVE This scoping review aims critically analyze landscape of digital identify existing gaps opportunities, explore feasibility developing a comprehensive, universal AI-based platform. METHODS A was conducted following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews Meta-Analyses extension Scoping Reviews) guidelines. comprehensive search peer-reviewed literature, industry reports, documents performed databases including PubMed, Embase, CINAHL, PsycINFO, Scopus, IEEE Xplore, Web Science. Inclusion criteria focused on studies, published from 2014 2024, addressing platforms, integration, science applications, regulations, professional ethics. Data were charted synthesized themes related platform capabilities, gaps, potential. RESULTS The identified seven common types utilized alongside growing trend implementation AI, generative AI. Significant found interoperability, cross-sector collaboration, utilization. technologies showed promise discovery diagnostic accuracy reducing costs. Despite advances, differences data standards, constraints, proprietary formats hinder seamless with electronic health record systems workflows. Moreover, applications neglect personalized treatment plans that integrate genomics, lifestyle factors, comorbidities, social determinants health. Organizations like World Health Organization, Massachusetts Institute Technology, Microsoft exploring integrated solutions, but standardization, privacy, regulatory, ethical considerations remain. CONCLUSIONS could revolutionize research innovation. By fragmentation fostering sectors, such enable continuous improvement innovation delivery development. overcoming compliance, standards is critical. Collaborative efforts sectors essential develop connected, responsive, effective national global systems. Future should focus defining platform's structure scope, design strategies, aligning ethics ensure equitable integration.

Язык: Английский

Процитировано

0