Jun 2026· ACM International Conference on Bioinformatics, Computational Biology and Biomedicine· pp. 1-2· 0 citations· 2 references
Computer Science
TL;DR
The results confirm the technical feasibility of unified multimodal biomedical data integration within a standards-compliant clinical AI platform and establish SupportAI as a solid foundation for AI-powered decision support targeting personalized, accessible healthcare.
Abstract
Healthcare digitalization demands platforms capable of integrating heterogeneous biomedical data and delivering AI-powered clinical decision support at scale [1, 2]. The SupportAI project, funded by the Italian National Recovery and Resilience Plan (PNRR) through the Tech4You program (grant ECS 00000009, Spoke 6, Line B, Project 2.1), addresses this challenge through a comprehensive modular platform that unifies real-time clinical data integration, advanced 3D imaging, generative AI assistance, and collaborative tools within a standards-compliant architecture. The platform’s core architectural innovation is the embedding of Retrieval-Augmented Generation (RAG) directly within live HL7 FHIR R4 and DICOM infrastructure. Unlike offline RAG systems that operate on pre-exported or periodically updated datasets, SupportAI queries the Microsoft FHIR Server and Orthanc PACS at inference time, ensuring AI-generated responses always reflect the patient’s current multimodal record with full traceability to FHIR resource IDs and DICOM study UIDs. Additional contributions include Fourier frequency-domain imaging for anomaly detection beyond the spatial domain, an AI-assisted 3D bioprinting pipeline converting DICOM studies into patient-specific STL anatomical models, and a Jitsi-based telecollaboration infrastructure for multidisciplinary remote consultation. The platform comprises seven interconnected microservices: a Multimodal Data Source Module with automated ETL pipelines and GDPR pseudonymization; a FHIR Server organizing clinical information via RESTful APIs; an Orthanc PACS for DICOMweb-compatible image access; an Advanced 3D Imaging Module with GPU-accelerated Cornerstone3D rendering; a 3D Bioprinting Module supporting AI-assisted MONAI segmentation and manual VTK workflows for STL generation; an AI Module implementing LangChain-orchestrated generative assistance grounded in live clinical records; and a Collaboration Module providing Jitsi-based video consultation in virtual 3D laboratory environments. Security is enforced through TLS 1.3, OAuth 2.0/OpenID Connect, and role-based access control. Technical validation at Technology Readiness Level 6 was conducted using Apache JMeter with 10 concurrent virtual users. Results demonstrate platform robustness across all modules. 3D imaging visualization, AI segmentation, and FHIR patient reads achieved sub-15 ms mean response times; the 3D bioprinting pipeline completed end-to-end within minutes even for complex studies; and the generative chat pipeline averaged 23,629 ms, consistent with GPT-4-class deployments under equivalent context complexity. All eight tested configurations achieved a 0.00% error rate under concurrent load. Static code analysis (SonarQube) on the full 1,214k-line codebase achieved Quality Gate: Passed with zero open security issues. Functional correctness was confirmed through complete User Acceptance Testing across all modules and four representative clinical scenarios: 3D imaging and bioprinting, generative AI clinical assistance, AI-assisted image segmentation, and telecollaboration. Physician evaluators reported substantially increased trust in AI outputs due to transparent source attribution, and post-segmentation manual correction requirements were markedly reduced compared to purely manual workflows. These results confirm the technical feasibility of unified multimodal biomedical data integration within a standards-compliant clinical AI platform and establish SupportAI as a solid foundation for AI-powered decision support targeting personalized, accessible healthcare.
Breast Cancer Multidisciplinary Team (MDT) meetings manage increasingly complex cases under considerable time pressure, and documentation requirements can reduce clinical efficiency and decision quality. Existing AI based MDT workflows rely on cloud-based processing, limiting their use because patient discussions contain identifiable information. We developed a fully on-device AI pipeline using open-source Automatic Speech Recognition (ASR) and Large Language Models (LLMs) that transcribes breast cancer MDT discussions, structures clinical information, and generates treatment recommendations using retrieval-augmented generation (RAG) grounded in National Institute for Health and Care Excellence (NICE) guidance. The pipeline runs on a single NVIDIA Jetson AGX Orin, ensuring that patient audio, transcripts, and outputs remain within institutional infrastructure. Evaluation included two recorded simulated MDT discussions, ten clinically validated synthetic discussions, and 1,270 acoustically augmented recordings. Optimisation of Whisper large-v3 reduced word error rate by 20.7% and 24.4% on the recorded discussions and achieved performance within 0.58% WER and 1.58% word information lost of a commercial clinical ASR benchmark on augmented audio. MedGemma-RAG identified 2.3 times more MDT-concordant interventions than a proprietary cloud comparator (p = 0.020), with no significant difference in overall accuracy. Stakeholders identified automated documentation, treatment recommendation support, and case triage as the most credible near-term applications while highlighting workflow integration, governance, and clinician trust as key implementation challenges. These findings demonstrate the feasibility of privacy-preserving, fully on-device AI for MDT documentation and guideline-informed decision support, providing a foundation for prospective clinical evaluation.
Aarzoo Dhiman, F. Haque, K. Grover et al.· 0 citations
The digital healthcare field is expanding fast, and now it requires platforms that use advanced technology and maintain robust data security and compliance practices. In this paper, we present the main architecture, key methodologies, and compliance strategies of iHelpCare, a digital healthcare system designed to meet HIPAA and GDPR standards while improving healthcare accessibility, efficiency, and inclusivity. The platform uses AI-based features to deliver personalized care solutions, focuses on preventive health management, and offers adaptive tools for people with disabilities. iHelpCare enables real-time patient monitoring, ensures secure medical data management, and facilitates convenient communication among patients, caregivers, and healthcare providers. Moreover, special attention is provided to people with Alzheimer’s through memory aid tools, cognitive exercises, caregiver resources, and AI-based detection analytics. All these characteristics are intended to detect cognitive decline at an early stage, enabling prompt interventions and improving patients’ quality of life. Also, iHelpCare considers unique health needs by providing access to culturally appropriate healthcare materials, telehealth consultations in multiple languages, and community support networks, making healthcare easier and more effective for this community. The platform’s AI analytics provide predictive insights that help medical professionals anticipate health conditions, optimize treatment plans, and reduce caregiver burden. To improve accessibility, features such as voice command, screen reader, and gesture recognition are included to help users with cognitive and physical disabilities. iHelpCare is envisioned to evolve through advanced sensor integration, personalized and inclusive care models, enhanced security, smart home support, and clinically validated tools for patients and caregivers.
Soarov Chakra Borty, Trisha Bhowmick, Mehedi Hasan et al.· 2026 International Conferenc...· 0 citations
KiMeKo (KI-Med-Kollaborationsplattform) is a publically funded collaborative research project that develops a sustainable AI-Med ecosystem for AI-based medical device development. The project runs from July 2024 to December 2027 and joins seven Northern German research institutions. KiMeKo addresses the complete development trajectory, from concept and data acquisition to validation, regulatory evidence generation, and approval-oriented documentation. The project contributes a practical toolchain and platform capabilities for non-experts and experts, including structured innovation support, uncertaintyaware sensor-data fusion, hybrid expert-system modeling, and workflow-guided data acquisition and anonymization. This paper summarizes project objectives, expected outputs, relevance to IEEE COMPSAC 2026 themes, and current progress. In particular, KiMeKo aligns with Applied AI and Smart & Connected Health by combining AI engineering, privacy-conscious data processing, and regulation-aware medical software development.
Serge Autexier, N. Ay, Stefan Fischer et al.· Annual International Compute...· 0 citations
Healthcare systems are rapidly embedding adaptive and generative AI into core clinical processes. The integration of Artificial Intelligence into Clinical Decision Support Systems (AI-CDSS) highlights a fundamental transformation within healthcare delivery. This transformation enables advanced predictive analytics, multimodal data integration, and real-time augmentation of clinical decisions. However, AI introduces systemic, ethical, operational, and governance risks that challenge traditional healthcare audit and assurance frameworks. In fact, assurance methodologies designed for static software are becoming insufficient for patient safety and regulatory compliance. This review discusses the evolving landscape of AI-CDSS audit, highlighting its transition from a technically focused lifecycle validation to an integrated paradigm centered on socio-technical resilience. Early audit approaches adapted conventional medical device and software validation models to machine learning–enabled clinical tools. While these models established baseline safety oversight, they were not designed to address adaptive algorithms functioning in complex, evolving clinical environments. Modern governance frameworks emphasize continuous performance monitoring, lifecycle surveillance, and structured human oversight. However, a persistent implementation gap remains. Many audit models poorly capture the interactions among algorithmic behavior, clinical workflows, organizational culture, and shifting patient populations. In fact, three major paradigm shifts are reforming the AI-CDSS audit. First, the audit scope is expanding from model-centric evaluation to ecosystem-level assurance that incorporates workflow integration, human-machine collaboration quality, and organizational learning capacity. Second, the field is moving from post-hoc explainability toward reasoning traceability and synergistic clinical sense-making. Third, audit philosophy is shifting from static compliance verification toward resilience-oriented monitoring that focuses on adaptive capacity, graceful degradation, and safe performance evolution. Thus, we propose the STRAICS framework (Socio-Technical Resilience Assurance for Intelligent Clinical Systems), which integrates technical robustness, human-machine interaction safeguards, adaptive governance, and transparency-by-design infrastructure. These components are vital for building trustworthy, effective, and impartial clinical AI ecosystems that can safely manage the increasing complexity of healthcare.
Rami A. Al‐Horani, Amanuel F. Tadesse· Frontiers in Artificial Inte...· 0 citations
Medicine is inherently multimodal, requiring clinicians to synthesize information across diverse data streams. Yet the development of multimodal foundation models is constrained by limited access to large-scale, high-quality clinical data. Although PubMed Central (PMC) offers a complementary source of expert-authored image-text data, existing PMC-derived resources remain limited in fidelity, reproducibility, and clinical validation. We introduce MedPMC, an automated, continuously updatable framework that transforms permissively licensed literature into high-fidelity infrastructure for medical multimodal models. Applied to 6.1 million PMC articles, MedPMC curated 11 million medical image-text pairs. Component evaluations showed strong performance for initial screening (F1 = 93.2), multi-panel figure detection (F1 = 96.5), figure separation (mAP = 89.8), caption separation and alignment (F1 = 81.4; ROUGE-L = 85.3), and medical figure classification (F1 = 96.5). Manual review by five annotators, three with medical training, found 95.3% of MedPMC images medically relevant, versus 19.7% in a prior PMC-derived dataset. Across 26 benchmarks spanning 11 specialties, a MedPMC-trained CLIP-style model improved average zero-shot AUC by 7.1 percentage points over the strongest architecture-matched biomedical CLIP baseline despite using fewer than half as many image-text pairs. As the vision encoder in a multimodal large language model, it improved medical visual question-answering by 1.9 and 16.9 percentage points across two benchmarks. In 10,524 Yale New Haven Health System dermatology photographs, it improved morphology-to-image retrieval Recall@5 by 11.7 percentage points. These findings show that high-fidelity literature curation strengthens medical multimodal foundation models across benchmark and clinical settings. We publicly release the framework, corpus, benchmarks, and pretrained models.
Hyunjae Kim, Dain Kim, Pan Xiao et al.· 0 citations
Personalized medicine aims to tailor prevention, diagnosis, and treatment strategies to individual patients by leveraging heterogeneous data sources such as electronic health records, medical imaging, genomic profiles, and real-time physiological signals. Although artificial intelligence has demonstrated remarkable predictive performance in this domain, the widespread clinical adoption of such models remains constrained by their black-box nature, limited transparency, and lack of trust among clinicians and patients. Explainable Artificial Intelligence (XAI) has emerged as a critical paradigm to address these limitations by providing interpretable, transparent, and clinically meaningful insights into model behavior and decision logic. This paper presents a comprehensive examination of explainable AI in personalized medicine, focusing on its role in bridging complex patient data with actionable clinical decisions. The study discusses major XAI methodologies, including intrinsic interpretability models and post hoc explanation techniques, and evaluates their applicability across key medical use cases such as disease risk prediction, treatment response modeling, and clinical decision support systems. Furthermore, challenges related to data heterogeneity, model generalization, ethical compliance, and regulatory acceptance are analyzed. By integrating explainability with predictive accuracy, XAI-driven frameworks have the potential to enhance clinical confidence, support evidence-based decision-making, and improve patientcentric outcomes. The paper concludes by outlining future research directions toward scalable, trustworthy, and regulation-compliant explainable AI systems for next-generation personalized healthcare.
Anurag Shrivastava, Neeraj Gupta, A. Madhavi et al.· 2026 International Conferenc...· 0 citations
