BACKGROUND:Solid organ transplantation (SOT) is the standard therapeutic approach to end-stage organ failure. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), has emerged as a promising tool for analyzing large, complex datasets, enabling both prediction of rejection risk and early detection of established graft injury. This systematic review synthesizes current evidence on AI-based approaches for predicting future rejection risk and detecting active rejection in human SOT, evaluates their performance, and identifies gaps for future research. METHODS:This review followed PRISMA 2020 guidelines. PubMed/MEDLINE, EMBASE, and Web of Science were searched up to April 30, 2025, using terms related to AI and graft rejection. Eligible studies included peer-reviewed original research using AI to predict, detect, or monitor rejection in humans. Three reviewers independently screened titles, abstracts, and full texts, resolving disagreements by consensus. Due to heterogeneity in methods and objectives, meta-analysis was not feasible. RESULTS:Of 195 studies identified, 62 met inclusion criteria. Most focused on kidney transplantation (n = 49, 79%), followed by heart (n = 6, 10%), liver (n = 4, 6%), lung (n = 1, 2%), and pancreas (n = 1, 2%). One study addressed multiple organs. Among diagnostic studies, AI, particularly ML and DL, demonstrated high diagnostic performance in non-kidney transplantation, often exceeding reported AUC of 0.90. In kidney transplantation, DL models, including convolutional neural networks and transformer-based architectures, reached accuracies up to 99.89% and AUCs up to 0.99. ML methods such as XGBoost, Bayesian classifiers, and logistic regression also performed well, with XGBoost achieving AUCs of 0.95-0.97, Bayesian classifiers reaching accuracies of 93.3% to 97%, and logistic regression models reporting AUC values up to 0.969. Among predictive studies, ML-based models similarly demonstrated strong discriminative performance. CONCLUSION:AI models using ML and DL may show strong potential, particularly in kidney transplantation, for non-invasive early detection of active graft rejection and prediction of future rejection risk, across diverse data types. The exceptionally high performance reported by some studies warrants careful interpretation. Challenges such as lack of standardization, limited validation, and interpretability must be addressed through well-designed multicentre studies to support clinical translation.
Malignancy has become a major limitation to long-term survival after lung transplantation. Its risk is shaped by recipient susceptibility, transplant anatomy, oncogenic viruses, long-term immunosuppression, and drug exposure. This narrative review integrates epidemiological, mechanistic, and clinical evidence and proposes a management framework spanning pre-transplant assessment, post-transplant surveillance, and treatment after cancer diagnosis. Lung cancer arises predominantly in the retained native lung after single-lung transplantation; early detection and curative treatment are central to improving outcomes. Keratinocyte carcinoma is common and recurrent, requiring coordinated dermatological surveillance, preventive treatment, and review of relevant drug exposures. Post-transplant lymphoproliferative disorder (PTLD) is closely linked to impaired Epstein-Barr virus (EBV) immune control. High-risk recipients warrant longitudinal viral monitoring, but diagnosis still requires histopathology, and treatment is based on reduced immunosuppression and response-adapted rituximab-based strategies. After cancer diagnosis, immunosuppression should be reconfigured according to tumour lethality, allograft reserve, and rejection risk, while coordinating anticancer therapy with allograft protection. Biomarkers may support monitoring, antibody-drug conjugates may expand treatment options, and immune checkpoint inhibitors may cause fatal allograft injury. Future work should use prospective registries and auditable pathways to evaluate cancer control and allograft outcomes together.
Cardiovascular disease is a significant cause of morbidity and mortality in kidney transplant candidates and recipients. Cardiac investigation prior to kidney transplant listing has aimed to minimise the risk of early post-operative cardiac events or premature death. Invasive coronary angiography and coronary revascularisation may be considered as part of pre-transplant work up, but there is limited evidence to support this strategy in asymptomatic patients. The joint British Transplantation Society and British Cardiovascular Society guideline on ‘Cardiovascular assessment of potential adult kidney transplant recipients’ has been written to harmonise UK practice and highlight current evidence in this area. This manuscript summarises the relevant recommendations from chapters related to coronary angiography and coronary revascularisation in adult patients being assessed for kidney transplantation.
Cardiovascular disease is a significant cause of morbidity and mortality in kidney transplant candidates and recipients. With non-invasive cardiac investigation and coronary revascularisation failing to show consistent benefits in patient outcomes, focus is being placed on medical management of coronary artery disease as a strategy to reduce cardiovascular risk. The joint British Transplantation Society and British Cardiovascular Society guideline on ‘Cardiovascular assessment of potential adult kidney transplant recipients’ has recently been developed. This manuscript summarises the relevant recommendations from chapters related to medical management of cardiovascular disease.
BACKGROUND:To address high waiting-list mortality and severe donor shortages in thoracic transplantation, clinical protocols have expanded donor selection criteria to include marginal, elderly, and controlled donation after circulatory death (cDCD) candidates. Concurrently, ex-vivo lung perfusion (EVLP), ex-vivo heart perfusion (EVHP), and normothermic regional perfusion (NRP) have reconfigured organ preservation. This study evaluates the clinical-ethical challenges triggered by these strategies to optimize graft utilization safely. METHODS:A systematic review was conducted evaluating peer-reviewed literature (2016-2026) following the PRISMA 2020 statement across PubMed/MEDLINE, Scopus, and Web of Science. Combined search strategies targeting intensive care logistics and thoracic transplantation were employed. Retrieved records were systematically screened at the title, abstract, and full-text levels according to predefined inclusion criteria. Peer-reviewed studies addressing Intensive Care Unit (ICU) donor management, withdrawal of life-sustaining treatment (WLST), and advanced thoracic procurement logistics were qualitatively analyzed. RESULTS:Nineteen studies fulfilled all inclusion criteria. Expanding thoracic donor selection targets relies heavily on precise functional warm ischemia time (fWIT) management and the chronological independence of WLST decisions. While advanced in-situ NRP and ex-vivo EVLP/EVHP preservation techniques maximize cardiothoracic donor assessment and reconditioning, they introduce critical tensions regarding the Dead Donor Rule (DDR) and the definition of circulatory irreversibility. CONCLUSIONS:Safely expanding the thoracic donor pool requires strict operational separation between end-of-life care and procurement teams. Evolving bioethical frameworks serve as analytical safeguards, reinforcing systemic integrity by ensuring advanced preservation technologies respect the donor's bodily integrity and the transition of dying without compromising clinical governance.
Urinary extracellular vesicles (uEVs) have emerged as a promising and versatile source of non-invasive biomarkers for kidney allograft monitoring. This review summarizes current evidence on uEV-derived biomarkers, including proteins, gene expression signatures, microRNAs, and extracellular vesicle-associated DNA, highlighting their potential role in detecting allograft rejection, calcineurin inhibitor nephrotoxicity, and progressive graft dysfunction. Urinary EVs reflect molecular processes occurring within the graft, including immune activation, metabolic dysregulation, and tissue remodelling. Consequently, the analysis of uEVs may offer real-time information on graft status through a readily accessible biological fluid, complementing conventional surveillance strategies and enabling earlier detection of allograft injury. For kidney transplant recipients. The effective uEV translation into clinical practice will require a collective efforts to standardized analytical workflows, validate findings in large prospective multicentric studies, and integrate uEV-based assays into precision medicine approaches.
BACKGROUND:Kidney transplantation improves survival and quality of life in patients with end-stage renal disease; however, long-term graft success depends on adherence to immunosuppressive therapy. OBJECTIVE:To quantify adherence levels among kidney transplant recipients using electronic monitoring devices. METHODS:PubMed, Web of Science, Scopus, and the Cochrane Library were systematically searched for studies published between 2000 and 2025. A random-effects meta-analysis was conducted including studies that assessed adherence using electronic monitoring devices. RESULTS:Seven studies with a total of 1025 kidney transplant recipients were included. The pooled mean adherence measured using electronic monitoring devices was 94.39%. In studies with monitoring durations of three months or longer, adherence decreased to 90.50%. Substantial heterogeneity was observed across studies. CONCLUSION:Although electronically measured adherence appears generally high, its decline over longer monitoring periods indicates that adherence is dynamic and may decrease over time. However, these estimates should be interpreted cautiously because electronic monitoring devices cannot confirm actual medication ingestion. Furthermore, intentionally non-adherent patients may be less likely to participate in or remain engaged with electronic monitoring studies and may therefore be underrepresented, potentially leading to an overestimation of true adherence. Even modest reductions in adherence may carry important clinical consequences, underscoring the need for sustained monitoring.
BACKGROUND:Acute kidney injury (AKI) after heart transplantation (HT) remains incompletely understood. This systematic review and meta-analysis aimed to estimate the pooled incidence of AKI and AKI requiring renal replacement therapy (RRT) after HT, assess associated mortality risks and identify characteristics associated with AKI requiring RRT. METHODS:A systematic literature search identified original studies reporting the incidence of AKI after HT using standardized definitions or reporting AKI requiring RRT. Meta-analyses were conducted using random-effects models. RESULTS:Forty-eight studies covering 14,389 patients were included. The pooled incidence of AKI was 57.05% (95% CI: 49.51-64.28), and AKI requiring RRT occurred in 16.99% (95% CI: 13.58-21.04). Among patients with AKI, 30-day and one-year mortality were of 12.96% (95% CI: 9.80-16.94) and 18.51% (95% CI: 14.56-23.23) respectively. AKI was associated with an approximately threefold increase in 30-day (OR: 2.92; 95% CI: 2.02-4.22) and one-year mortality (OR: 3.07; 95% CI: 1.76-5.34). In patients requiring RRT, 30-day and one-year mortality were 26.47% (95% CI: 18.46-36.41) and 35.13% (95% CI: 28.78-42.06), respectively. AKI requiring RRT was associated with and increased 30-day mortality of more than twelvefold (OR: 12.11; 95% CI: 6.05-24.24) and one-year mortality of more than fivefold (OR: 5.40; 95% CI: 3.88-7.51) Characteristics associated with AKI requiring RRT included higher body mass index, elevated baseline creatinine, chronic kidney disease, perioperative extracorporeal membrane oxygenation support, and longer cardiopulmonary bypass time. CONCLUSIONS:AKI is highly prevalent after HT and is strongly associated with increased early and late mortality, particularly in patients requiring RRT.
Artificial intelligence (AI) and machine learning (ML) are increasingly developed and evaluated for the organ transplantation pathway, yet the breadth of empirical evidence generated within European healthcare systems, and its actual translation into routine clinical use, has not been mapped. This scoping review charts the nature, clinical targets, modelling approaches and reported performance of AI applications in organ transplantation in Europe. Following Arksey and O'Malley's framework and PRISMA-ScR reporting, MEDLINE and Embase were searched using terms covering organ transplantation, AI, ML, deep learning, allocation, decision support, immunosuppression and graft survival. Empirical primary studies conducted wholly or partly in Europe were prioritised for charting, while systematic, narrative and position-paper literature informed the background and comparative discussion. Empirical European studies spanned kidney, liver, lung, heart and pancreas transplantation and addressed graft and patient survival prediction, donor-recipient matching, allocation, graft and biopsy assessment, rejection detection, and decision support. Tree-based ensembles and neural networks predominated, with discrimination commonly between 0.70 and 0.95 and frequently exceeding conventional scores such as MELD, KDRI and SOFT. Registry-based national cohorts from the United Kingdom, Spain, France, Germany, Italy and multinational European consortia were prominent. Few tools reached prospective or randomised evaluation, and benefits in shared decision-making were not consistently demonstrated. Crucially, almost all identified studies described model development or validation rather than adoption into routine care, and evidence of sustained clinical implementation within European transplant units was scarce. AI shows substantial promise across the European transplantation pathway, particularly for risk stratification, organ assessment and rejection detection, but most evidence remains retrospective and developmental rather than clinically deployed, and is concentrated in a small number of countries with mature registries. Prospective and externally validated studies, attention to equity and generalisability across diverse populations, workflow integration, and alignment with European regulation are required before routine clinical adoption.
Uterus transplantation has emerged as a promising treatment option for patients with absolute uterine factor infertility who desire gestation. Unlike solid organ transplantation performed for life-preserving indications, uterus transplantation is a temporary procedure, with graft removal planned after completion of childbearing. This distinctive feature raises a novel question: whether the same uterine graft could be recovered and reused in a subsequent recipient. In this review, we examine uterine graft reuse as a potential strategy to increase graft availability and broaden access in selected settings. Existing uterus transplantation outcome data establish that transplanted uteri can support menstruation, pregnancy, and live birth and that pregnancy after uterus transplantation carries meaningful maternal and perinatal risks. However, no clinical data establish the feasibility or safety of reusing a uterine graft in another recipient. We therefore frame graft reuse as an exploratory hypothesis requiring preclinical validation. Key considerations include the donor pathway, donor-recipient immunologic and virologic compatibility, the effects of prior transplantation and pregnancy on graft quality, vascular and operative feasibility at graft retrieval, postpartum timing, and the obstetric implications of prior cesarean delivery. Additional issues include ischemic tolerance, consent for serial graft use, and equitable allocation. Overall, uterine graft reuse represents a speculative extension of the temporary-graft model that distinguishes uterus transplantation from other transplant disciplines. Further progress will depend on animal studies, surgical modeling, and careful ethical, infectious, immunologic, and obstetric assessment. By separating established clinical observations from unanswered questions, this article provides a framework for future investigation of uterine graft reuse.
Allograft nephrectomy (AN) after graft failure remains controversial, balancing symptom control and facilitation of immunosuppression withdrawal against operative risk and uncertain immunological consequences. A systematic review of literature published from January 2000 to March 2024 in accordance with PRISMA guidelines was performed. We included studies addressing indications, timing, complications and outcomes related to subsequent transplantation. Thirty-nine studies were included. Reported prevalence varied widely but was <10% in most contemporary cohorts and declined after the cyclosporine era. Early AN was mainly performed for primary non-function, severe acute rejection, and surgical or vascular complications. Late transplantectomy was more often related to graft intolerance syndrome, refractory hypertension, or chronic infection. Overall morbidity was about 18% and mortality 4%, with higher risk in emergency settings. AN was not associated with improved recipient or graft survival after retransplantation, although it may increase sensitization and early graft-related events. In children, AN appeared more frequent than in adults, but evidence remained limited. Overall, AN appears to be an indication-driven rather than routine intervention, requiring individualized risk-benefit assessment within a retransplant strategy.
INTRODUCTION:Kidney transplantation is the intervention of choice for patients with end-stage renal disease. The advent of direct acting antiviral (DAA) medication has changed the management of hepatitis C virus (HCV), which has led to some centres expand donation criteria to include HCV-positive donors. This study aims to assess the influence of donor HCV status on post-transplantation outcomes. METHODS:A PRISMA-compliant, systematic search of PubMed, EMBASE, Cochrane and Web of Science databases was performed. Studies comparing outcomes of recipients of HCV-positive versus HCV-negative donor kidneys were included. Outcomes included graft function, survival, acute rejection and delayed graft function. Subgroup analysis was performed stratifying studies by era, recipient HCV status and the presence of viraemia. RESULTS:The systematic search identified 1461 studies which after evaluation led to the inclusion of 31 studies, with a total of 133,788 patients of whom 11,663 patients received HCV-positive donor kidneys and 122,125 received HCV-negative donor kidneys. There was a higher eGFR at 6-months for recipients of HCV-positive kidneys (MD 3.57, 95%CI: 1.68 to 5.47, p = 0.0002). No significant difference was observed between either group for all other outcomes. The subgroup analysis did not show worse outcomes for receiving a kidney from a viraemic donor or for HCV-negative recipients of HCV-positive kidneys. CONCLUSION:This meta-analysis of the best available evidence demonstrates comparable outcomes for recipients of kidneys from HCV-positive donors. Provided the availability of DAA treatment, we recommend other centres expand their donation criteria to include HCV-positive donors.
BACKGROUND:Delayed graft function (DGF) is a commonly reported outcome in kidney transplantation research, but no single definition has been uniformly adopted. Variation in how DGF is defined, measured, and analysed may limit evidence synthesis, and affect trial design. We conducted a scoping review to map contemporary DGF outcome reporting practices in adult kidney transplantation. METHODS:We conducted a scoping review using the Arksey and O'Malley framework, reported in accordance with PRISMA-ScR. MEDLINE and Embase were searched for English-language studies published from 1 January 2009 onwards. Eligible adult kidney transplant studies reported DGF as an outcome and explicitly defined it. Data were charted on study characteristics, definition type, timing, aggregation, justification, and analysis. Because some studies reported more than one definition, the unit of analysis was the individual DGF definition. RESULTS:261 included sources yielded 305 DGF definitions, with 88 definitional variants across 5 main definitional families. 60 full-text records were excluded because they used DGF as an outcome without definition. Dialysis-within-7-days predominated, which accounted for 44% of definitions. Alternative dialysis-based definitions accounted for 18%, duration- or severity-stratified definitions for 21%, creatinine-based definitions for 9%, and composite clinical definitions for 9%. The first post-transplant week was the dominant time frame. Most definitions were binary (98%). Justification for authors' chosen definition was absent in 41%. CONCLUSIONS:Contemporary DGF reporting is dominated by binary dialysis-based definitions despite the biological and clinical heterogeneity of DGF. A standardised research endpoint designed according to a novel framework may improve validity, comparability, and future trial design.
Genetic cardiomyopathies commonly cause end-stage heart failure, yet genetic testing is inconsistently applied in heart transplant recipients. Identifying pathogenic/likely pathogenic (P/LP) variants clarifies etiology and informs familial risk, but data on diagnostic yield are limited. We performed a meta-analysis to quantify genetic testing yield in heart transplant recipients and implications for cascade testing. MEDLINE and Embase were searched for studies examining cardiomyopathy-focused genetic testing in heart transplant recipients. Pooled P/LP cardiomyopathy variant yields were estimated using random-effects models, stratified by cardiomyopathy phenotype. Cascade testing outcomes were summarized. Eleven studies met inclusion criteria. Overall, 32% (95% CI: 21-46%) of heart transplant recipients carried a P/LP variant. Yield was highest in non-ischemic cardiomyopathy (34%, 95% CI: 24-47%) and non-ischemic dilated cardiomyopathy cohorts (29%, 95% CI: 12-54%), and lowest in ischemic cardiomyopathy cohorts (8%, 95% CI: 4-18%). Heterogeneity ranged from I2 35% to 92%. Among studies reporting cascade testing outcomes, 52-92% of families underwent testing; 30-40% of relatives had P/LP variants, 15-69% of which demonstrated a phenotypic CM. While exact estimates should be interpreted cautiously given heterogeneity across cohorts, the consistency of findings suggests meaningful clinical relevance. Cardiomyopathy-focused genetic testing identifies P/LP variants in many heart transplant recipients, with important implications for cascade testing, supporting integration of cardiomyopathy-related genetic evaluation into heart transplant programs.
Obesity and metabolic disease frequently shape transplant candidacy, and body mass index (BMI) remains embedded in evaluation and listing criteria. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and dual GLP-1/glucose-dependent insulinotropic polypeptide (GIP) agonists now offer non-surgical pathways to clinically meaningful weight loss, including among candidates previously limited by BMI thresholds. However, in transplantation, crossing a BMI cutoff does not necessarily establish readiness for major surgery, immunosuppression, rehabilitation, and graft-protective follow-up. This focused narrative review examines evidence on incretin-based therapy in solid organ transplant candidates and recipients and proposes a distinction between an access response (achievement of BMI-related eligibility or listing goals) and a reserve response (weight loss accompanied by preserved or improved body composition, muscle strength, function, nutrition, cardiopulmonary tolerance, treatment tolerability, and immunosuppressive stability). Available transplant data are promising but largely observational and centered on weight, glycated hemoglobin, graft function, tacrolimus troughs, listing, and transplantation. Measures of lean mass, frailty trajectory, handgrip strength, gait speed, nutritional adequacy, patient-reported outcomes, perioperative events, and tacrolimus exposure variability remain underreported. We outline reserve-sensitive monitoring domains, an operational reserve-response matrix that classifies concordant favorable, access-only, metabolic-only, depletion, intolerance, pharmacokinetic-instability, and perioperative-risk trajectories, and reporting priorities for future studies. Incretin therapy may become an important bridge to transplantation, but its success should not be defined by BMI reduction alone. Prospective studies should determine whether BMI-based access gains are accompanied by preserved or improved physiologic reserve and safer progression to transplantation.
Organ transplantation research presents promising research prospects owing to its important clinical application and translational significance. Since the advent of single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics (ST), researchers have been liberated from the constraints of traditional experimental techniques and enabled to conduct in-depth investigations at the cellular level and with high spatial resolution, ushering organ transplantation research into a new stage. By summarizing the research achievements of scRNA-seq and ST applied in conventional organ transplantation and xenotransplantation, this review illustrates the common mechanisms underlying organ transplantation as well as the clinical value of scRNA-seq and ST in the field of organ transplantation. In addition, this review also discusses the limitations and current technical challenges of scRNA-seq and ST.
Ex-situ machine perfusion (MP) technologies have revolutionised solid organ transplantation. Their core aims are to reduce ischemia-reperfusion injury (to improve recipient outcomes), prolong preservation times, and allow organ viability assessment. Different organs have unique considerations owing to their specific sensitivity to ischemia-reperfusion injury, as well as organ-specific logistical challenges and differing consequences of poor early graft function. Despite these differences, the core pathophysiologic processes underlying ischemia-reperfusion injury are shared, resulting in an organ-agnostic pathology of early graft dysfunction. The consequence of this shared biology is an opportunity for cross-organ learning. For example, the finding that hypothermic oxygenated MP ameliorates ischemia-reperfusion injury, and improves clinically relevant outcomes, has emerged as a consistent trend in randomised trials across multiple organs. In contrast, similar improvements in recipient outcomes have not been demonstrated in randomised trials of normothermic MP, whether applied to abdominal or thoracic organs. Such an integrated perspective could enable prioritisation of therapies for evaluation, particularly in settings where clinical trial activity is constrained, such as heart transplantation, by leveraging evidence generated in kidney transplantation where larger randomised trials are feasible and early graft dysfunction is more easily tolerated. In this integrated review we assess the clinical trial evidence for ex-situ MP across kidney, heart, liver, lung and pancreas. We focus on clinical trials reporting recipient outcomes, prolonged preservation and utilisation, highlighting themes for cross-organ learning.
BACKGROUND:Allograft nephrectomy (AN) is a rare but morbid procedure required after graft failure or neoplasia. We systematically reviewed current knowledge on AN surgical techniques, complications, and alternatives. PATIENTS AND METHODS:Undertaken to inform the French Association of Urology guidelines on allograft nephrectomy, this review was reported following PRISMA guidelines. PubMed/MEDLINE was searched (January 2000-March 2024) for studies of patients with non-functional first kidney transplants requiring AN or alternative strategies. Case reports, editorials, and non-systematic reviews were excluded. RESULTS:AN is predominantly performed via an open extraperitoneal approach. For non-functional grafts, consensus is lacking regarding subcapsular versus extracapsular techniques. However, the subcapsular approach reduces operative time, bleeding, and intraoperative complications, though overall morbidity data remain conflicting. The surgical approach does not impact alloimmunization. Regarding arterial embolization, protocols are not standardized. Yet, it achieves >80% success with lower mortality and morbidity than AN. Its main complication is post-embolization syndrome, potentially leading to secondary AN. Embolization prior to AN effectively reduces intraoperative bleeding and operative time. CONCLUSION:The extraperitoneal subcapsular approach appears to be the safest surgical option. Arterial embolization is a valuable alternative or adjuvant strategy to AN in managing graft intolerance syndrome after graft failure.
Rejection surveillance in heart transplantation (HT) has evolved beyond reliance on histologic grading alone to incorporate circulating biomarkers and tissue-based molecular diagnostics, including gene-expression profiling (GEP), donor-derived cell-free DNA (dd-cfDNA), and transcriptomic assessment. These tools have expanded the ability to detect biologic signals of alloimmune injury that may not be fully captured by morphology alone. At the same time, evidence that their routine use improves long-term clinical outcomes remains limited, and adoption has been uneven across healthcare systems. In several North American programs, dd-cfDNA testing and transcriptomic platforms have been incorporated into surveillance pathways, although implementation remains heterogeneous across centers and clinicians. In many European healthcare settings, availability is increasing, but integration into routine practice remains more variable, and comprehensive tissue-based profiling is still largely restricted to selected clinical scenarios. These differences in access and implementation may influence how rejection, graft stability, and longitudinal risk are interpreted in clinical practice. The next phase of the field will likely depend not on replacing histology, but on integrating molecular diagnostics into multimodal, longitudinal risk-based surveillance strategies. Coordinated validation efforts, prospective outcome-linked studies, and context-sensitive implementation frameworks will be essential to clarify the clinical role of these tools and to support equitable integration across diverse healthcare environments.
