Survey and perspective on verification, validation, and uncertainty quantification of digital twins for precision medicine

npj Digital Medicine, Год журнала: 2025, Номер 8(1)

Опубликована: Янв. 17, 2025

Digital twins in precision medicine provide tailored health recommendations by simulating patient-specific trajectories and interventions. We examine the critical role of Verification, Validation, Uncertainty Quantification (VVUQ) for digital ensuring safety efficacy, with examples cardiology oncology. highlight challenges opportunities developing personalized trial methodologies, validation metrics, standardizing VVUQ processes. frameworks are essential integrating into clinical practice.

Язык: Английский

---

Integrating large language model and digital twins in the context of industry 5.0: Framework, challenges and opportunities

Robotics and Computer-Integrated Manufacturing, Год журнала: 2025, Номер 94, С. 102982 - 102982

Опубликована: Фев. 10, 2025

Язык: Английский

---

From screens to scenes: A survey of embodied AI in healthcare

Information Fusion, Год журнала: 2025, Номер unknown, С. 103033 - 103033

Опубликована: Фев. 1, 2025

Язык: Английский

---

A comprehensive survey of large language models and multimodal large language models in medicine

Information Fusion, Год журнала: 2024, Номер unknown, С. 102888 - 102888

Опубликована: Дек. 1, 2024

Язык: Английский

---

Medical Digital Twins for Personalized Chronic Care

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Янв. 16, 2025

This work introduces the concept of Patient Medical Digital Twins (PMDTs) to simulate treatment outcomes, optimize drug dosages, and deliver personalized chronic care. The PMDT model, supported by an interconnected ecosystem, is validated iteratively medical institutions ensure its efficacy applicability. At core, leverages expressive knowledge structures capture a patient’s psychosomatic, cognitive, biometric, genetic data, creating comprehensive personal digital footprint. enables professionals run simulations predicting health issues over time proactively implement preventive interventions. ecosystem integrates big data analytics, continuous monitoring, cognitive simulation, AI technologies. By connecting stakeholders across care continuum, it provides deeper insights into history supports informed, shared decision-making. Validated in pilot study through EU-funded healthcare initiative, demonstrates transformative potential at intersection Big Data AI, positioning itself as critical tool for advancing

Язык: Английский

---

A Scoping Review of Artificial Intelligence Applications in Clinical Trial Risk Assessment

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2025, Номер unknown

Опубликована: Янв. 22, 2025

Abstract Artificial intelligence (AI) is increasingly applied to clinical trial risk assessment, aiming improve safety and efficiency. This scoping review analyzes 142 studies published between 2013 2024, focusing on (n=55), efficacy (n=46), operational (n=45) prediction. AI techniques, including traditional machine learning, deep learning (e.g., graph neural networks, transformers), causal are used for tasks like adverse drug event prediction, treatment effect estimation, phase transition These methods utilize diverse data sources, from molecular structures protocols patient scientific publications. Recently, large language models (LLMs) have seen a surge in applications, representing over 20% of 2023. While some achieve high performance (AUROC up 96%), challenges remain, selection bias, limited prospective studies, quality issues. Despite these limitations, AI-based assessment holds substantial promise transforming trials, particularly through improved risk-based monitoring frameworks.

Язык: Английский

---

Beyond digital twins: the role of foundation models in enhancing the interpretability of multiomics modalities in precision medicine

FEBS Open Bio, Год журнала: 2025, Номер unknown

Опубликована: Фев. 24, 2025

Medical digital twins (MDTs) are virtual representations of patients that simulate the biological, physiological, and clinical processes individuals to enable personalized medicine. With increasing complexity omics data, particularly multiomics, there is a growing need for advanced computational frameworks interpret these data effectively. Foundation models (FMs), large‐scale machine learning pretrained on diverse types, have recently emerged as powerful tools improving interpretability decision‐making in precision This review discusses integration FMs into MDT systems, their role enhancing multiomics data. We examine current challenges, recent advancements, future opportunities leveraging analysis MDTs, with focus application

Язык: Английский

---

The comprehensive clinical benefits of digital phenotyping: from broad adoption to full impact

npj Digital Medicine, Год журнала: 2025, Номер 8(1)

Опубликована: Апрель 8, 2025

Digital phenotyping collects health data digitally, supporting early disease diagnosis and management. This paper systematically reviews the diversity of research methods in digital its clinical benefits, while also focusing on importance within P4 medicine paradigm core role advancing application biobanks. Furthermore, envisions continued benefits phenotyping, driven by technological innovation, global collaboration, policy support.

Язык: Английский

---

Clinical Laboratory Parameter–Driven Machine Learning for Participant Selection in Bioequivalence Studies Among Patients With Gastric Cancer: Framework Development and Validation Study

JMIR AI, Год журнала: 2025, Номер 4, С. e64845 - e64845

Опубликована: Май 5, 2025

Abstract Background Insufficient participant enrollment is a major factor responsible for clinical trial failure. Objective We formulated machine learning (ML)–based framework using laboratory parameters to identify participants eligible in bioequivalence study. Methods acquired records of 11,592 patients with gastric cancer from the electronic medical Kyungpook National University Hospital Korea. The ML model was developed 8 parameters, including complete blood count and liver kidney function tests, along dates acquisition. Two datasets were collected: (1) training dataset design an ML-based candidate selection method (2) test evaluate performance proposed method. generalization confirmed F 1 - score area under curve (AUC). compared random its efficacy recruiting participants. Results weighted ensemble achieved strong -score above 0.8 AUC value exceeding 0.8, demonstrating ability accurately valid candidates while minimizing misclassification. Its high sensitivity further enhanced model’s efficiency prioritizing screening. In case study, reduced workload by 57%, efficiently identifying 150 pool 209, 485 required selection. Conclusions can be used trial, enabling faster enrollment.

Язык: Английский

---

Digital twins, synthetic patient data, and in-silico trials: can they empower paediatric clinical trials?

The Lancet Digital Health, Год журнала: 2025, Номер unknown, С. 100851 - 100851

Опубликована: Май 1, 2025

Randomised controlled trials are the gold standard to assess effectiveness and safety of clinical interventions; however, many paediatric discontinued early due challenges in patient enrolment. Hence, most suffer from lack adequate power. Additionally, expensive might expose patients unproven therapies. Alternatives overcome these issues using virtual data-namely, digital twins, synthetic data, in-silico trials-are now possible rapid advances health-care tools interventions. However, such innovations have been rarely used trials. In this Viewpoint, we propose data empower The use has advantages decreased exposure children potentially ineffective or risky interventions, shorter trial durations leading more ascertainment faster drug approvals. Use could lead personalised treatment options with low costs result implementation interventions children. ethical regulatory concerns, including replacing humans privacy, security should be addressed sustainability innovation ensured before adopted widely.

Язык: Английский

---