Journal of Personalized Medicine最新文献

{"title":"Personalised Blood Glucose Time Series Forecasting in Type 1 Diabetes: Deep Collaborative Adversarial Learning.","authors":"Heydar Khadem, Hoda Nemat, Jackie Elliott, Mohammed Benaissa","doi":"10.3390/jpm16040210","DOIUrl":"https://doi.org/10.3390/jpm16040210","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Blood glucose prediction (BGP) for individuals with type 1 diabetes (T1D) is a clinically essential yet highly challenging task in time series forecasting (TSF) and an important problem in personalised medicine. Accurate bespoke BGP is crucial for individualised T1D management, reducing complications, and supporting patient-specific glycaemic risk mitigation. However, the pronounced volatility of glycaemic fluctuations in T1D, combined with the need for mathematical rigor and clinical relevance, hampers reliable prediction. This complexity underscores the demand to explore and enhance more advanced techniques. While adversarial learning is adept at modelling intricate data variability, its potential for BGP remains largely untapped. <b>Methods:</b> This work presents a novel approach for BGP by addressing a key limitation in conventional adversarial learning when applied to this task. Typically, these methods optimise prediction accuracy within a set horizon by minimising adversarial loss. This focus overlooks how predictions align with longer-term patterns, which are critical for clinical relevance in BGP, thereby yielding suboptimal results. To overcome this limitation, we introduce collaborative augmented adversarial learning, designed to improve the model's temporal awareness. Incorporating collaborative interaction optimisation, this approach enables the model to reflect extended time dependencies beyond the immediate horizon, thereby improving both the clinical reliability of predictions and overall predictive performance. We develop and evaluate four learning systems for BGP: independent learning, adversarial learning, collaborative learning, and adversarial collaborative learning. The proposed systems were evaluated for two clinically relevant prediction horizons, namely 30 min and 60 min ahead. <b>Results:</b> The interdependent collaboratively augmented learning frameworks, validated using the well-established Ohio T1D datasets, demonstrate statistically significant superior performance in both clinical and mathematical evaluations. <b>Conclusions:</b> Beyond advancing BGP accuracy and clinical reliability, the proposed approach supports personalised medicine by improving subject-specific glucose forecasting from CGM data, with potential relevance for more individualised diabetes monitoring and decision support. The proposed approach also opens new avenues for advancements in other complex TSF domains, as outlined in our future work.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117441/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Extracellular Vesicles in Osteonecrosis of the Femoral Head: An Integrated Review of Experimental and Bioinformatic Evidence.","authors":"Elvira Immacolata Parrotta, Giorgia Lucia Benedetto, Giovanni Cuda, Umile Giuseppe Longo, Arianna Carnevale, Olimpio Galasso, Giorgio Gasparini, Michele Mercurio","doi":"10.3390/jpm16040208","DOIUrl":"https://doi.org/10.3390/jpm16040208","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Osteonecrosis of the femoral head (ONFH) is a progressive condition characterized by bone necrosis, impaired vascularization, and immune dysregulation, often resulting in femoral head collapse. Effective strategies to halt disease progression are limited. Extracellular vesicles (EVs), including exosomes and microvesicles, mediate intercellular communication and influence osteogenesis, angiogenesis, and immune responses. This review summarizes current evidence on EVs in ONFH and their translational potential. <b>Methods:</b> A structured narrative review of PubMed, Scopus, Web of Science, and Cochrane Central databases was conducted, including in vitro, preclinical, and clinical studies on EVs in ONFH. Data on EV sources, molecular cargo, signaling pathways, functional effects, and translational implications were qualitatively synthesized. No pooled statistical analysis was performed because the extracted data were heterogeneous. Bioinformatic analyses such as Gene Ontology, KEGG enrichment, and protein-protein interaction networks were also summarized. <b>Results:</b> In vitro, EVs from bone marrow mesenchymal stem cells, endothelial cells, and M2 macrophages modulate osteogenic differentiation, angiogenesis, and inflammation. Preclinical studies demonstrate that EV administration reduces femoral head necrosis, improves trabecular structure, and enhances neovascularization. Clinical studies have identified EV-associated molecules (SAA1, C4A, RPS8) linked to disease stage and the risk of femoral head collapse. Bioinformatic analyses connect EV cargo to pathways regulating bone formation, vascularization, immunity, and metabolism. <b>Conclusions:</b> EVs appear to play key roles in ONFH pathogenesis and may represent promising candidates for diagnostic and therapeutic applications. However, current clinical evidence remains limited and requires validation in larger studies. Nonetheless, heterogeneity and limited clinical data require standardized, longitudinal studies to validate their translational relevance.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117771/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rationale and Design of a Randomised Proof-of-Concept Trial to Assess the Safety of Early Discharge Using Index Microcirculatory Resistance in Patients with Acute Myocardial Infarction: SECURE Study.","authors":"Muntaser Omari, Mohamed Ali, Luke Spray, Adam McDiarmid, Mohammad Alkhalil","doi":"10.3390/jpm16040207","DOIUrl":"https://doi.org/10.3390/jpm16040207","url":null,"abstract":"<p><p><b>Background</b>: Current guidelines acknowledge that early discharge is not associated with late mortality and that in-hospital length of stay (LOS) of 48-72 h should be considered following successful primary percutaneous coronary intervention (PPCI) in low-risk patients. Recent studies have highlighted the safety of very early discharge after PPCI in highly selected low-risk patients; however, objective tools to guide discharge timing remain limited. The Index of Microcirculatory Resistance (IMR) offers a quantitative assessment of microvascular function and may help identify patients suitable for very early discharge. We aimed to evaluate the feasibility of using IMR to guide very early discharge in patients who underwent uncomplicated PPCI. <b>Study design and objectives</b>: The Safety of Early Discharge Using Index Microcirculatory Resistance in Patients with Acute Myocardial Infarction (SECURE) study is designed to assess the feasibility of using IMR, measured immediately following successful PPCI, to guide early discharge from hospital within 24 h. The SECURE study is a prospective, proof-of-concept, functional non-inferiority, single-centre, randomised, open-label trial to determine if patients with low IMR can be safely discharged when compared to standard discharge policy. The SECURE study will recruit 82 patients with low IMR following successful PPCI. Participants will be 1:1 randomised to either standard discharge timing or very early discharge (within 24 h). The left ventricle ejection fraction will be assessed using cardiac magnetic resonance imaging. A telephone follow-up at 3 months will be arranged. Clinical events are collected as secondary and exploratory safety endpoints. <b>Conclusions</b>: The SECURE study will provide proof-of-concept data about the feasibility of using IMR to guide very early discharge following PPCI. If successful, this study will provide data to plan for a larger study to determine the safety of this personalised approach.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13118203/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In Silico Psycho-Oncology: Understanding Resilience Pathways in Breast Cancer-Determinants of Longitudinal Depression and Quality-of-Life Trajectories.","authors":"Eleni Kolokotroni, Paula Poikonen-Saksela, Ruth Pat-Horenczyk, Berta Sousa, Albino J Oliveira-Maia, Ketti Mazzocco, Haridimos Kondylakis, Georgios S Stamatakos","doi":"10.3390/jpm16040209","DOIUrl":"https://doi.org/10.3390/jpm16040209","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Patients with breast cancer show substantial heterogeneity in terms of psychological adjustment following diagnosis. We aimed to characterize longitudinal trajectories of quality of life (QoL) and depressive symptoms during the first 18 months post-diagnosis and to identify robust clinical, psychosocial, and behavioral predictors associated with distinct adjustment pathways. <b>Methods:</b> Women (<i>N</i> = 538; mean age 55.4 years; range 40-70) with operable breast cancer (stages I-III) were drawn from the multicenter BOUNCE cohort. QoL (Global Health Status/QoL scale of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30) and depressive symptoms (depression subscale of the Hospital Anxiety and Depression Scale) were assessed at baseline and months 3, 6, 9, 12, 15 and 18. Latent class growth analysis and growth mixture modeling identified distinct trajectory classes. Associations between early predictors and trajectory membership were examined using logistic regression combined with elastic net regularization. <b>Results:</b> Depression trajectories demonstrated heterogeneity, with groups characterized by persistent resilience (59.7%), stable moderate/high (25.3%), delayed onset (5.0%), and recovery (10.0%). QoL trajectories ranged from stable excellent (13.2%) and stable high (40.7%) to moderate (31.4%) and persistent low/deteriorating (6.9%), as well as a distinct recovering trajectory (7.8%). Trajectory differentiation was primarily driven by psychological resources, symptom burden, functional status, and coping processes, alongside specific contributions from clinical factors. <b>Conclusions:</b> Distinct subgroups of women with breast cancer follow divergent adjustment pathways. These findings highlight the multidimensional nature of resilience and support the need for tailored interventions that promote long-term well-being beyond simple risk reduction.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117978/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Assessment of GPT-3.5 and GPT-4.0 Responses to Scoliosis FAQs.","authors":"Tu-Lan Vu-Han, Enikö Regényi, Vikram Sunkara, Paul Köhli, Friederike Schömig, Alexander P Hughes, Michael Putzier, Matthias Pumberger, Thilo Khakzad","doi":"10.3390/jpm16040206","DOIUrl":"https://doi.org/10.3390/jpm16040206","url":null,"abstract":"<p><p><b>Background</b>: ChatGPT is a large language model (LLM) online chatbot developed by OpenAI and launched in November 2022. Early adoption studies have shown high readiness to use this technology for health-related questions and self-diagnosis. However, the quality and clinical adequacy of health-related responses remain incompletely characterized. This study aimed to explore responses generated by ChatGPT-3.5 and ChatGPT-4.0 to common patient questions regarding scoliosis. <b>Methods</b>: Ten scoliosis-related frequently asked questions (FAQs) were selected from a larger pool of over 250 patient-facing questions compiled from 17 publicly available FAQ webpages and informed by a Google Trends analysis. Questions were harmonized, grouped by theme, and then reduced by rule-based expert review to a final set intended to represent common patient concerns. <b>Results</b>: The median ratings of ChatGPT-3.5 and ChatGPT-4.0 responses ranged from satisfactory, requiring minimal (2) to moderate clarification (3). Across the ten matched questions, no statistically detectable difference was found between models in this study setting (<i>W</i> = 8.0, <i>p</i> = 0.59; Cliff's δ = -0.12 [95% CI -0.58, 0.40]); however, given the small question set, unblinded rating process, and poor inter-rater reliability, this should not be interpreted as evidence of equivalence, non-inferiority, or comparable model performance. The results apply only to the 10-15 April 2024, online snapshots of ChatGPT-3.5 and ChatGPT-4.0 and should not be generalized to later model iterations. <b>Conclusions</b>: This study should be interpreted as a clinically oriented observational report, intended to inform physician awareness and patient-physician communication rather than validate chatbot accuracy or safety. In this 10-15 April 2024, sample, both model outputs frequently required clinician clarification. Given the small FAQ set, low inter-rater reliability, unblinded design, and single-sample outputs, the findings do not establish equivalence or superiority and apply only to the specific 10-15 April 2024, model snapshots and evaluated questions.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117403/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized Prediction of Postoperative Recurrence in Lung Squamous Cell Carcinoma: Integrating AI-Based Nuclear Morphometry and Clinical Data.","authors":"Tomokazu Omori, Akira Saito, Yoshihisa Shimada, Yujin Kudo, Jun Matsubayashi, Toshitaka Nagao, Masahiko Kuroda, Norihiko Ikeda","doi":"10.3390/jpm16040205","DOIUrl":"https://doi.org/10.3390/jpm16040205","url":null,"abstract":"<p><p><b>Background:</b> This study employed artificial intelligence (AI) to analyze quantitative nuclear morphological features obtained from digital pathology images to predict postoperative recurrence in patients with lung squamous cell carcinoma (LSQCC). We aimed to develop a prediction model that contributes to the realization of 'personalized postoperative management' tailored to individual tumor biology by integrating AI-extracted morphological features with clinical information. <b>Methods:</b> A total of 185 of the 253 surgically resected LSQCC cases were included; 136 were randomly assigned to the training set and 49 to the test set. Nuclear features from manually selected regions of interest were extracted and used to build AI-based prediction models. Three recurrence models were developed: recurrence within 2 years, within 5 years, and a three-category model (≤2 years, 3-5 years, >5 years or no recurrence). Support vector machine (SVM) and random forest (RF) algorithms were applied to each, yielding six predictive models. An ensemble approach was used to calculate AI-based risk scores, and a \"total risk score\" was developed by integrating these with the pathologic stage. <b>Results</b>: All six AI models demonstrated stable predictive performance, with AUC values ranging from 0.76 to 0.91. Kaplan-Meier analysis showed that the total risk score provided the most precise risk stratification (<i>p</i> < 0.005), with clearer separation between risk groups than the AI-based risk score alone. <b>Conclusions:</b> The integration of AI-based nuclear morphology analysis and clinical data provides an objective and practical tool for personalized postoperative management in LSQCC. This approach enables tailored clinical decision-making by identifying patients at high risk for early recurrence and customizing postoperative treatment plans to meet the specific needs of each individual.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117173/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does Provider Identity at Triage Improve Machine Learning Prediction of Hospital Admission? A Comparative Analysis of Ten Supervised Classifiers with SHAP Explainability.","authors":"Adam E Brown, Chance W Marostica, Wayne A Martini","doi":"10.3390/jpm16040204","DOIUrl":"https://doi.org/10.3390/jpm16040204","url":null,"abstract":"<p><p><b>Background/Objectives:</b> Machine learning (ML) models can predict hospital admission from emergency department (ED) triage data with areas under the receiver operating characteristic curve (AUC) exceeding 0.85. Whether incorporating the assigned provider's identity-as a proxy for unmeasured practice variation-improves prediction has not been systematically studied. We aimed to compare 10 supervised ML classifiers for predicting hospital admission at ED triage, with and without provider identity, and to characterize model reasoning using SHapley Additive exPlanations (SHAP). <b>Methods:</b> We conducted a retrospective cohort study of 186,094 ED visits (2020-2023, training) and 58,151 visits (2024, temporal holdout test) at one academic tertiary-care ED. Ten classifiers spanning linear, distance-based, tree-based, ensemble, probabilistic, and neural network families were each trained in two conditions: baseline (23 triage features) and with provider identity appended. SHAP TreeExplainer was applied to the top-performing models (CatBoost and XGBoost). <b>Results:</b> The admission rate was 31.3% (training) and 31.7% (test). CatBoost achieved the highest baseline AUC of 0.8906 (0.8878-0.8933). Adding provider identity produced negligible AUC changes across all models (ΔAUC range: -0.0029 to +0.0015; all DeLong <i>p</i> > 0.05). SHAP analysis identified ESI level, respiratory rate, temperature, complaint category, and age as the dominant predictors, with clinically intuitive directionality. <b>Conclusions:</b> Provider identity does not meaningfully improve ML prediction of hospital admission beyond standard triage variables. The observed 28-percentage-point variation in provider admission rates is explained by patient case-mix differences than with independent practice pattern effects on prediction. SHAP provides transparent, clinically interpretable explanations suitable for bedside decision support.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117648/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Soluble Isoforms of PD-1 and PD-L1 in Non-Small Cell Lung Cancer: Correlation with Tumor Stage, Longitudinal Analysis and Prognostic Implications.","authors":"Konstantinos Vachlas, Dimitra Grapsa, Stylianos Gaitanakis, Anna Papadopoulou, Paraskevi Moutsatsou, Nikolaos Syrigos, Ioannis P Trontzas","doi":"10.3390/jpm16040203","DOIUrl":"https://doi.org/10.3390/jpm16040203","url":null,"abstract":"<p><p><b>Background:</b> Soluble immune checkpoint molecules, including soluble PD-1 (sPD-1) and soluble PD-L1 (sPD-L1), have emerged as potential minimally invasive biomarkers in non-small cell lung cancer (NSCLC). However, their diagnostic, kinetic, and prognostic significance across different disease settings remains unclear. This prospective study evaluated baseline levels, longitudinal fluctuations, and clinical associations of sPD-1 and sPD-L1 in early- and advanced-stage NSCLC. <b>Methods:</b> Three cohorts were prospectively enrolled: early-stage NSCLC patients undergoing curative surgery (n = 25), advanced-stage NSCLC patients receiving pembrolizumab-based immunotherapy (n = 55), and non-oncological controls (n = 16). Serum sPD-1 and sPD-L1 were measured by ELISA at baseline and at four months post-surgery (early stage) or six months post-treatment (advanced stage). Baseline comparisons, longitudinal changes, correlation with tumor PD-L1 expression (TPS), and associations with recurrence (early stage) or 6-month objective response (advanced stage) were assessed. <b>Results:</b> Baseline sPD-1 and sPD-L1 levels did not differ significantly among controls, early-stage, and advanced-stage cohorts. In early-stage patients, sPD-L1 increased post-operatively (<i>p</i> = 0.006) while sPD-1 decreased (<i>p</i> < 0.001). In advanced-stage disease, sPD-1 declined during immunotherapy (<i>p</i> < 0.001), whereas sPD-L1 remained unchanged (<i>p</i> = 0.37). Baseline levels and continuous percent changes were not predictive of most outcomes. However, a ≥20% postoperative increase in sPD-L1 was strongly associated with recurrence in early-stage NSCLC (OR = 10.29; 95% CI: 1.40-215.20; <i>p</i> = 0.019). No sPD-1/PD-L1 metric predicted response in advanced disease. Baseline sPD-L1 showed no correlation with tumor PD-L1 expression (ρ = -0.09, <i>p</i> = 0.53) in the advanced-stage cohort. <b>Conclusions:</b> sPD-1 and sPD-L1 demonstrate distinct kinetic patterns across NSCLC settings. A postoperative >20% surge in sPD-L1 may identify early-stage patients at elevated risk of recurrence, whereas soluble checkpoints were not predictive of treatment response in advanced disease. These findings support further investigation of soluble checkpoint dynamics as complementary biomarkers in NSCLC management in larger cohorts.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117836/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774614","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Role of Cardiovascular Magnetic Resonance in Post-Heart Transplant Surveillance: Integrating Evidence with Prospective Cohort Data.","authors":"Ricardo Carvalheiro, Vera Vaz Ferreira, Ana Raquel Santos, Isabel Cardoso, António Valentim Gonçalves, Rita Ilhão Moreira, Tiago Pereira da Silva, Sílvia Aguiar Rosa, Rui Cruz Ferreira","doi":"10.3390/jpm16040201","DOIUrl":"https://doi.org/10.3390/jpm16040201","url":null,"abstract":"<p><p><b>Background</b><b>:</b> Heart transplantation remains the definitive therapy for selected patients with end-stage heart failure, but outcomes are limited by acute rejection, chronic allograft injury, and cardiac allograft vasculopathy. Endomyocardial biopsy (EMB) remains the reference standard for rejection surveillance but is invasive and imperfectly captures diffuse myocardial injury. Cardiovascular magnetic resonance (CMR) offers noninvasive, multiparametric assessment of graft structure, function, tissue composition, and perfusion. We aimed to review current evidence supporting CMR in post-heart transplant surveillance and to evaluate the performance of serial CMR for acute cellular rejection in a prospective cohort. <b>Methods:</b> We performed a focused narrative review of the literature on CMR for detection of acute rejection, assessment of chronic allograft injury and prognosis, and evaluation of cardiac allograft vasculopathy and microvascular disease. In parallel, we conducted a prospective observational study of adult heart transplant recipients undergoing early post-transplant CMR (CMR1) and follow-up CMR (CMR2) with temporally matched EMB. Multiparametric CMR included cine imaging, native T1 and T2 mapping, extracellular volume fraction (ECV), and late gadolinium enhancement (LGE). Clinically significant acute cellular rejection was defined as ISHLT grade ≥ 2R. <b>Results:</b> Eighteen recipients were included (median 53 days to CMR1 and 192 days to CMR2). Baseline CMR parameters correlated with invasive hemodynamic and biomarkers. Two patients had biopsy-proven ≥2R rejection at follow-up. T2 values at CMR2 were significantly higher in rejection versus non-rejection patients (59.0 ± 1.4 ms vs. 51.1 ± 1.9 ms; <i>p</i> = 0.015), with greater LGE burden in rejection (<i>p</i> = 0.029). In longitudinal analyses, rejection was associated with divergent patterns of cardiac remodelling and tissue characterization, including increases in indexed ventricular volumes and T2 over time, whereas non-rejection patients demonstrated stable ventricular volumes and a decline in T2. <b>Conclusions:</b> Multiparametric CMR, anchored by T2 mapping, provides clinically meaningful, non-invasive information for acute rejection surveillance after heart transplantation and complements EMB within a personalized, risk-adapted follow-up framework. Establishing individualized baseline CMR phenotypes and monitoring longitudinal changes may support more personalized, less invasive graft surveillance strategies. Larger multicentre prospective studies are needed to define standardized implementation pathways.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117184/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating AI Segmentation, Simulated Digital Twins, and Extended Reality into Medical Education: A Narrative Technical Review and Proof-of-Concept Case Study.","authors":"Parhesh Kumar, Ingharan Siddarthan, Catharine Kelsh Keim, Daniel K Cho, John E Rubin, Robert S White, Rohan Jotwani","doi":"10.3390/jpm16040202","DOIUrl":"https://doi.org/10.3390/jpm16040202","url":null,"abstract":"<p><p><b>Background/Objectives</b>: Simulation digital twins (DT) models that integrate patient-specific imaging with artificial intelligence (AI)-based segmentation and extended reality (XR) technologies are rapidly increasing in relevance in personalized medicine. While their clinical applications are expanding, their role as reusable educational tools and the technical pipeline utilized for their development remain incompletely characterized. This narrative review examines current approaches to digital twin creation and XR integration, illustrated by a scoliosis-specific proof-of-concept educational case study. <b>Methods:</b> A narrative technical review was conducted by identifying relevant search keywords within the fields of AI-based image segmentation, extended reality in medicine, and medical education based on the authors' expertise and familiarity with the subject. PubMed, Google Scholar, and Scopus were searched for English-language studies published primarily between 2015 and 2025 addressing patient-specific three-dimensional modeling, AI-driven segmentation, and XR applications in spine, orthopedic, anesthesiology, and interventional care. A de-identified case of scoliosis is used to present a proof-of-concept example of this process of creating a simulated digital twin for the purpose of medical education in a recorded XR format. <b>Results:</b> Prior studies demonstrated benefits of patient-specific 3D models for anatomical understanding and procedural planning, while highlighting limitations in segmentation accuracy and workflow integration. Nevertheless, while DTs have traditionally served clinical roles in surgical planning or pre-procedural rehearsal, their pedagogical potential remains under-explored. In the proof-of-concept case study, AI-assisted segmentation enabled rapid creation of an anatomically detailed scoliosis digital twin that was incorporated into XR and used to produce a reusable, spatially anchored instructional experience focused on neuraxial access. <b>Conclusions:</b> AI-enabled digital twin models integrated with XR represent a promising approach for personalized, anatomy-driven medical education. Further evaluation is needed to assess educational outcomes, scalability, and integration into clinical training workflows.</p>","PeriodicalId":16722,"journal":{"name":"Journal of Personalized Medicine","volume":"16 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13117035/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147774620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}