Search Results (761)

If we do not urgently educate current and future medical professionals to critically evaluate and distinguish credible AI-assisted diagnostic tools from those whose performance is artificially inflated by data leakage or improper validation, we risk undermining clinician trust in all AI diagnostics and jeopardizing future advances in patient care. For instance, machine learning models have shown high accuracy in diagnosing Parkinson’s Disease when trained on clinical features that are themselves diagnostic, such as tremor and rigidity. This study systematically investigates the impact of data leakage and feature selection on the true clinical utility of machine learning models for early Parkinson’s Disease detection. We constructed two experimental pipelines: one excluding all overt motor symptoms to simulate a subclinical scenario and a control including these features. Nine machine learning algorithms were evaluated using a robust three-way data split and comprehensive metric analysis. Results reveal that, without overt features, all models exhibited superficially acceptable F1 scores but failed catastrophically in specificity, misclassifying most healthy controls as Parkinson’s Disease. The inclusion of overt features dramatically improved performance, confirming that high accuracy was due to data leakage rather than genuine predictive power. These findings underscore the necessity of rigorous experimental design, transparent reporting, and critical evaluation of machine learning models in clinically realistic settings. Our work highlights the risks of overestimating model utility due to data leakage and provides guidance for developing robust, clinically meaningful machine learning tools for early disease detection. Full article

Background/Objectives: Acute kidney injury (AKI) is a common and severe complication in patients with acute myocardial infarction (AMI). Very elderly patients are at a heightened risk of developing AKI. Fibrinogen and albumin are well-known biomarkers of inflammation and nutrition, which are highly related to AKI. We aim to explore the predictive value of the fibrinogen-to-albumin ratio (FAR) for AKI in very elderly patients with AMI. Methods: A retrospective cohort of AMI patients ≥ 75 years old hospitalized at the First Affiliated Hospital of Xi’an Jiaotong University between January 2018 and December 2022 was established. Clinical data and medication information were collected through the biospecimen information resource center at the hospital. Univariate and multivariable logistic regression models were used to analyze the association between FAR and the risk of AKI in patients with AMI. FAR was calculated as the ratio of fibrinogen (FIB) to serum albumin (ALB) level (FAR = FIB/ALB). The primary outcome is acute kidney injury, which was diagnosed based on KDIGO 2012 criteria. Results: Among 1236 patients enrolled, 66.8% of them were male, the median age was 80.00 years (77.00–83.00), and acute kidney injury occurred in 18.8% (n = 232) of the cohort. Comparative analysis revealed significant disparities in clinical characteristics between patients with or without AKI. Patients with AKI exhibited a markedly higher prevalence of arrhythmia (51.9% vs. 28.1%, p < 0.001) and lower average systolic blood pressure (115.77 ± 25.96 vs. 122.64 ± 22.65 mmHg, p = 0.013). In addition, after adjusting for age, sex, history of hypertension, left ventricular ejection fraction (LVEF), and other factors, FAR remained an independent risk factor for acute kidney injury (OR = 1.47, 95%CI: 1.36–1.58). ROC analysis shows that FAR predicted stage 2–3 AKI with superior accuracy (AUC 0.94, NPV 98.6%) versus any AKI (AUC 0.79, NPV 93.0%), enabling risk-stratified management. Conclusions: FAR serves as both a high-sensitivity screening tool for any AKI and a high-specificity sentinel for severe AKI, with NPV-driven thresholds guiding resource allocation in the fragile elderly. Full article

Background: Neonatal jaundice is common and can cause severe hyperbilirubinemia if untreated. The early identification of at-risk newborns is challenging despite the existing guidelines. Objective: This study aimed to identify the key maternal and neonatal risk factors for jaundice requiring phototherapy using machine learning. Methods: In this study hospital, phototherapy was administered following the American Academy of Pediatrics (AAP) guidelines when a neonate’s transcutaneous bilirubin level was in the high-risk zone. To identify the risk factors for phototherapy, we retrospectively analyzed the electronic medical records of 8242 neonates admitted between 2017 and 2022. Predictive models were trained using maternal and neonatal data. XGBoost showed the best performance (AUROC = 0.911). SHAP values interpreted the model. Results: Mode of delivery, neonatal feeding indicators (including daily formula intake and breastfeeding frequency), maternal BMI, and maternal white blood cell count were strong predictors. Cesarean delivery and lower birth weight were linked to treatment need. Conclusions: Machine learning models using perinatal data accurately predict the risk of neonatal jaundice requiring phototherapy, potentially aiding early clinical decisions and improving outcomes. Full article

Background: Artificial intelligence (AI) and machine learning (ML) have been reshaping maternal, fetal, neonatal, and reproductive healthcare by enhancing risk prediction, diagnostic accuracy, and operational efficiency across the perinatal continuum. However, no comprehensive synthesis has yet been published. Objective: To conduct a scoping review of reviews of AI/ML applications spanning reproductive, prenatal, postpartum, neonatal, and early child-development care. Methods: We searched PubMed, Embase, the Cochrane Library, Web of Science, and Scopus through April 2025. Two reviewers independently screened records, extracted data, and assessed methodological quality using AMSTAR 2 for systematic reviews, ROBIS for bias assessment, SANRA for narrative reviews, and JBI guidance for scoping reviews. Results: Thirty-nine reviews met our inclusion criteria. In preconception and fertility treatment, convolutional neural network-based platforms can identify viable embryos and key sperm parameters with over 90 percent accuracy, and machine-learning models can personalize follicle-stimulating hormone regimens to boost mature oocyte yield while reducing overall medication use. Digital sexual-health chatbots have enhanced patient education, pre-exposure prophylaxis adherence, and safer sexual behaviors, although data-privacy safeguards and bias mitigation remain priorities. During pregnancy, advanced deep-learning models can segment fetal anatomy on ultrasound images with more than 90 percent overlap compared to expert annotations and can detect anomalies with sensitivity exceeding 93 percent. Predictive biometric tools can estimate gestational age within one week with accuracy and fetal weight within approximately 190 g. In the postpartum period, AI-driven decision-support systems and conversational agents can facilitate early screening for depression and can guide follow-up care. Wearable sensors enable remote monitoring of maternal blood pressure and heart rate to support timely clinical intervention. Within neonatal care, the Heart Rate Observation (HeRO) system has reduced mortality among very low-birth-weight infants by roughly 20 percent, and additional AI models can predict neonatal sepsis, retinopathy of prematurity, and necrotizing enterocolitis with area-under-the-curve values above 0.80. From an operational standpoint, automated ultrasound workflows deliver biometric measurements at about 14 milliseconds per frame, and dynamic scheduling in IVF laboratories lowers staff workload and per-cycle costs. Home-monitoring platforms for pregnant women are associated with 7–11 percent reductions in maternal mortality and preeclampsia incidence. Despite these advances, most evidence derives from retrospective, single-center studies with limited external validation. Low-resource settings, especially in Sub-Saharan Africa, remain under-represented, and few AI solutions are fully embedded in electronic health records. Conclusions: AI holds transformative promise for perinatal care but will require prospective multicenter validation, equity-centered design, robust governance, transparent fairness audits, and seamless electronic health record integration to translate these innovations into routine practice and improve maternal and neonatal outcomes. Full article

In recent years, data augmentation techniques have become the predominant approach for addressing highly imbalanced classification problems in machine learning. Algorithms such as the Synthetic Minority Over-sampling Technique (SMOTE) and Conditional Tabular Generative Adversarial Network (CTGAN) have proven effective in synthesizing minority class samples. However, these methods often introduce distributional bias and noise, potentially leading to model overfitting, reduced predictive performance, increased computational costs, and elevated cybersecurity risks. To overcome these limitations, we propose a novel architecture, FADEL, which integrates feature-type awareness with a supervised discretization strategy. FADEL introduces a unique feature augmentation ensemble framework that preserves the original data distribution by concurrently processing continuous and discretized features. It dynamically routes these feature sets to their most compatible base models, thereby improving minority class recognition without the need for data-level balancing or augmentation techniques. Experimental results demonstrate that FADEL, solely leveraging feature augmentation without any data augmentation, achieves a recall of 90.8% and a G-mean of 94.5% on the internal test set from Kaohsiung Chang Gung Memorial Hospital in Taiwan. On the external validation set from Kaohsiung Medical University Chung-Ho Memorial Hospital, it maintains a recall of 91.9% and a G-mean of 86.7%. These results outperform conventional ensemble methods trained on CTGAN-balanced datasets, confirming the superior stability, computational efficiency, and cross-institutional generalizability of the FADEL architecture. Altogether, FADEL uses feature augmentation to offer a robust and practical solution to extreme class imbalance, outperforming mainstream data augmentation-based approaches. Full article

Epilepsy is a chronic neurological disease with a relatively high incidence in the pediatric population. Anti-seizure medication (ASM) may cause adverse drug reactions (ADRs), which may occur repeatedly. Objective: This study aimed to analyze the recurrence of ADRs caused by ASMs over a period of 122 months in hospitalized Mexican pediatric epilepsy patients. The patients were under monotherapy or polytherapy treatment, with valproic acid (VPA), phenytoin (PHT), and levetiracetam (LEV), among others. A total of 313 patients met the inclusion criteria: 211 experienced ADRs, whereas 102 did not. Patient sex, age, seizure type, nutritional status and related drugs were considered explanatory variables. Methods: Four statistical models were used to analyze recurrent events that were defined as “one or more ADRs occurred on a single day”, considering both the classification of ADR seriousness and the ASM causing the ADR. Results: A total of 499 recurrence events were identified. The recurrence risk was significantly greater among younger patients for both nonsevere and severe ADRs and among those with focal seizures for nonsevere ADRs. Interestingly, malnutrition was negatively associated with the risk of nonsevere ADRs, and obesity was positively associated with the risk of severe ADRs. Finally, LEV was associated with a significantly greater risk of causing nonsevere ADRs than VPA. However, LEV significantly reduced the risk of severe ADRs compared with VPA, and PHT increased the risk in comparison with VPA. In conclusion, this study offers a robust clinical tool to predict risk factors for the presence and recurrence of ASM-ADRs in pediatric patients with epilepsy. Full article

Background/Objectives: Artificial intelligence (AI)-enabled electrocardiogram (ECG) analysis may assist in objective and reproducible risk stratification. However, the prognostic utility of serial ECGs, particularly the follow-up ECG prior to discharge, has not been extensively studied. This study aimed to evaluate whether dynamic changes in AI-predicted ECG risk scores could enhance prediction of post-discharge outcomes. Methods: This retrospective cohort study included 11,508 ED visits from a single medical center where patients underwent two ECGs and were directly discharged. We stratified the mortality risk of patients as low risk, medium risk, and high risk based on the first and follow-up ECG prior to discharge using AI-enabled ECG models. The Area Under the Curve (AUC) was calculated for the predictive performance of the two ECGs. Kaplan–Meier (KM) curves were used for 90-day mortality analysis, and the Cox proportional hazards model was utilized to compare the risk of death across categories. Results: The AI-enabled ECG risk prediction model, based on the initial and follow-up ECGs prior to discharge, indicated risk transitions among different groups. The AUC for mortality risk was 78.6% for the first ECG and 83.3% for the follow-up ECG. KM curves revealed a significant increase in 90-day mortality for patients transitioning from low to medium/high risk upon discharge (Hazard Ratio: 6.01; Confidence Interval: 1.70–21.27). Conclusions: AI-enabled ECGs obtained prior to discharge provide superior mortality risk stratification for ED patients compared to initial ECGs. Patients classified as medium- or high-risk at discharge require careful consideration, whereas those at low risk can generally be discharged safely. Although AI-ECG alone does not replace comprehensive risk assessment, it offers a practical tool to support clinical judgment, particularly in the dynamic ED environment, by aiding safer discharge decisions. Full article

Background: Intravascular albumin mass represents the total quantity of albumin circulating within the bloodstream and may serve as a physiologically relevant marker of vascular integrity and fluid distribution in sepsis. While low serum albumin levels are acknowledged as prognostic indicators, dynamic assessments based on albumin mass remain insufficiently explored in patients outside the intensive care unit. Objectives: To describe the temporal changes in intravascular albumin mass in patients with community-acquired sepsis and to examine its relationship with fluid balance and thirty-day mortality. Methods: This prospective observational study encompassed 247 adults diagnosed with community-acquired sepsis who were admitted to a high-dependency hospital ward specializing in acute medical care. The intravascular albumin mass was calculated daily for a duration of up to five days, utilizing plasma albumin concentration and estimated plasma volume derived from anthropometric and hematologic data. Net albumin leakage was defined as the variation in intravascular albumin mass between consecutive days. Fluid administration and urine output were documented to ascertain cumulative fluid balance. Repeated-measures statistical models were employed to evaluate the associations between intravascular albumin mass, fluid balance, and mortality, with adjustments made for age, comorbidity, and clinical severity scores. Results: The intravascular albumin mass exhibited a significant decrease during the initial five days of hospitalization and demonstrated an inverse correlation with the cumulative fluid balance. A greater net leakage of albumin was associated with a positive fluid balance and elevated mortality rates. Furthermore, a reduced intravascular albumin mass independently predicted an increased risk of mortality at thirty days. Conclusions: A reduction in intravascular albumin mass may suggest ineffective fluid retention and the onset of capillary leak syndrome. This parameter holds promise as a clinically valuable, non-invasive indicator for guiding fluid resuscitation in cases of sepsis. Full article

Background: Osteodysplastic syndromes comprise a very diverse group of clinically and genetically heterogeneous disorders characterized by defects in bone and connective tissue development, as well as in bone density. Here, we report the case of a 48-year-old female with a complex medical history characterized by bone dysplasia, hyperostosis, and partial tooth agenesis. Methods: Genetic testing was performed using WES analysis and Sanger sequencing. Molecular modeling analysis and dynamics simulation explored the impact of detected pathogenic variants. Results: The genetic analysis detected multiple pathogenic variants in genes CREB3L1, SLCO2A1, SFRP4, LRP5, and LRP6, each of which has been associated with rare osteodysplastic syndromes. The patient was homozygous for the same rare alleles associated with three of the identified autosomal recessive disorders osteogenesis imperfecta type XVI, primary hypertrophic osteoarthropathy, and metaphyseal dysplasia Pyle type. She also had a variant linked to autosomal dominant endosteal hyperostosis and a variant previously associated with increased risk of osteoporosis and bone fractures. Two of the detected variants are predicted to cause abnormal splicing, while molecular modeling and dynamics simulations analysis suggest that the other three variants probably confer altered local secondary structure and flexibility that may have functionally devastating consequences. Conclusions: Our case highlights the rare coexistence of multiple osteodysplastic syndromes in a single patient that may complicate differential diagnosis. Furthermore, this case emphasizes the necessity for early genetic investigation of such complex cases with overlying phenotypic traits, followed by genetic counseling, facilitating orchestration of clinical interventions and allowing prevention and/or prompt management of manifestations. Full article

Food addiction (FA) is an emerging psychiatric condition that presents behavioral and neurobiological similarities with other addictions, and its early identification is essential to prevent the development of more severe disorders. The aim of the present study was to determine the ability of anthropometric measures, eating habits, symptoms related to eating disorders (ED), and lifestyle features to predict the symptoms of food addiction. Methodology: A cross-sectional study was conducted in a sample of 702 university students (77.3% women; age: 22 ± 6 years). The Food Frequency Questionnaire (FFQ), the Yale Food Addiction Scale 2.0 (YFAS 2.0), the Eating Attitudes Test (EAT-26), anthropometric measurements, and a set of self-report questions on substance use, physical activity level, and other questions were administered. A total of 6.4% of participants presented symptoms compatible with food addiction, and 8.1% were at risk for ED. Additionally, 26.5% reported daily smoking, 70.6% consumed alcohol, 2.9% used illicit drugs, and 29.4% took medication; 35.3% did not engage in physical activity. Individuals with food addiction had higher BMI (p = 0.010), waist circumference (p = 0.001), and body fat (p < 0.001) values, and a higher risk of eating disorders (p = 0.010) compared to those without this condition. In the multivariate logistic model, non-dairy beverage consumption (such as coffee or alcohol), vitamin D deficiency, and waist circumference predicted food addiction symptoms (R²_Nagelkerke = 0.349). Indeed, the machine learning approaches confirmed the influence of these variables. Conclusions: The prediction models allowed an accurate prediction of FA in the university students; moreover, the individualized approach improved the identification of people with FA, involving complex dimensions of eating behavior, body composition, and potential nutritional deficits not previously studied. Full article

Background/Objectives: Peripheral artery disease (PAD) impacts more than 200 million individuals globally and leads to mortality and morbidity secondary to progressive limb dysfunction and amputation. However, clinical management of PAD remains suboptimal, in part because of the lack of standardized biomarkers to predict patient outcomes. Growth differentiation factor 15 (GDF15) is a stress-responsive cytokine that has been studied extensively in cardiovascular disease, but its investigation in PAD remains limited. This study aimed to use explainable statistical and machine learning methods to assess the prognostic value of GDF15 for limb outcomes in patients with PAD. Methods: This prognostic investigation was carried out using a prospectively enrolled cohort comprising 454 patients diagnosed with PAD. At baseline, plasma GDF15 levels were measured using a validated multiplex immunoassay. Participants were monitored over a two-year period to assess the occurrence of major adverse limb events (MALE), a composite outcome encompassing major lower extremity amputation, need for open/endovascular revascularization, or acute limb ischemia. An Extreme Gradient Boosting (XGBoost) model was trained to predict 2-year MALE using 10-fold cross-validation, incorporating GDF15 levels along with baseline variables. Model performance was primarily evaluated using the area under the receiver operating characteristic curve (AUROC). Secondary model evaluation metrics were accuracy, sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). Prediction histogram plots were generated to assess the ability of the model to discriminate between patients who develop vs. do not develop 2-year MALE. For model interpretability, SHapley Additive exPlanations (SHAP) analysis was performed to evaluate the relative contribution of each predictor to model outputs. Results: The mean age of the cohort was 71 (SD 10) years, with 31% (n = 139) being female. Over the two-year follow-up period, 157 patients (34.6%) experienced MALE. The XGBoost model incorporating plasma GDF15 levels and demographic/clinical features achieved excellent performance for predicting 2-year MALE in PAD patients: AUROC 0.84, accuracy 83.5%, sensitivity 83.6%, specificity 83.7%, PPV 87.3%, and NPV 86.2%. The prediction probability histogram for the XGBoost model demonstrated clear separation for patients who developed vs. did not develop 2-year MALE, indicating strong discrimination ability. SHAP analysis showed that GDF15 was the strongest predictive feature for 2-year MALE, followed by age, smoking status, and other cardiovascular comorbidities, highlighting its clinical relevance. Conclusions: Using explainable statistical and machine learning methods, we demonstrated that plasma GDF15 levels have important prognostic value for 2-year MALE in patients with PAD. By integrating clinical variables with GDF15 levels, our machine learning model can support early identification of PAD patients at elevated risk for adverse limb events, facilitating timely referral to vascular specialists and aiding in decisions regarding the aggressiveness of medical/surgical treatment. This precision medicine approach based on a biomarker-guided prognostication algorithm offers a promising strategy for improving limb outcomes in individuals with PAD. Full article

Background: Preparing patients for surgery considers assessing the patient’s somatic health, for example by the American Society of Anesthesiology (ASA) scale or the Revised Cardiac Risk Index (RCRI), known as the Lee index. This process usually ignores mental functioning (personality and anxiety), which is known to influence health. The purpose of this study is to analyze the existence of a relationship between personality traits (the Big Five model and trait-anxiety) and anesthesia scales (ASA scale, Lee index) used for the preoperative evaluation of patients. Methods: The study group comprised 102 patients (59 women, 43 men) scheduled for hip replacement surgery. Patients completed two psychological questionnaires: the NEO-FFI (NEO Five Factors Inventory) and the X-2 STAI (State-Trait Anxiety Inventory) sheet. Next, the presence and possible strength of the relationship between personality traits and demographic and medical variables were analyzed using Spearman’s rho rank correlation coefficient. Results: Patients with a high severity of trait anxiety are classified higher on the ASA scale (rs = 0.359; p < 0.001). Neuroticism, defined according to the Big Five model, significantly correlates with scales of preoperative patient assessment: the ASA classification (rs = 0.264; p < 0.001) and the Lee index (rs = 0.202; p = 0.044). A hierarchical regression model was created to test the possibility of predicting ASA scores based on personality. It explained more than 34% of the variance and was a good fit to the data (p < 0.05). The controlled variables of age and gender accounted for more than 23% of the variance. Personality indicators (trait anxiety, neuroticism) additionally accounted for slightly more than 11% of the variance. Trait anxiety (Beta = 0.293) proved to be a better predictor than neuroticism (Beta = 0.054). Conclusions: These results indicate that inclusion of personality screening in the preoperative patient evaluation might help to introduce a more individualized approach to patients, which could result in better surgical outcomes. Full article

On a worldwide scale, neurodegenerative diseases, including multiple sclerosis, Parkinson’s, and Alzheimer’s, face considerable healthcare challenges demanding the development of novel approaches to early detection and efficient treatment. With its ability to provide real-time patient monitoring, customized medical care, and advanced predictive analytics, artificial intelligence (AI) is fundamentally transforming the way healthcare is provided. Through the integration of wearable physiological sensors, motion sensors, and neurological assessment tools, the NeuroPredict platform harnesses AI and smart sensor technologies to enhance the management of specific neurodegenerative diseases. Machine learning algorithms process these data flows to find patterns that point out disease evolution. This paper covers the design and architecture of the NeuroPredict platform, stressing the ethical and regulatory requirements that guide its development. Initial development of AI algorithms for disease monitoring, technical achievements, and constant enhancements driven by early user feedback are addressed in the discussion section. To ascertain the platform’s trustworthiness and data security, it also points towards risk analysis and mitigation approaches. The NeuroPredict platform’s capability for achieving AI-driven smart healthcare solutions is highlighted, even though it is currently in the development stage. Subsequent research is expected to focus on boosting data integration, expanding AI models, and providing regulatory compliance for clinical application. The current results are based on incremental laboratory tests using simulated user roles, with no clinical patient data involved so far. This study reports an experimental technology evaluation of modular components of the NeuroPredict platform, integrating multimodal sensors and machine learning pipelines in a laboratory-based setting, with future co-design and clinical validation foreseen for a later project phase. Full article

Background and Objectives: Investigating the adherence to antihypertensive medication and identifying patients with low adherence allows targeted interventions to improve therapeutic outcomes. Artificial intelligence (AI) offers advanced tools for analyzing medication adherence data. This study aimed to develop and validate several predictive models for non-adherence, using patient-reported data collected via a structured questionnaire. Materials and Methods: A cross-sectional, multi-center study was conducted on 3095 hypertensive patients from community pharmacies. A structured questionnaire was administered, collecting data on sociodemographic factors, medical history, self-monitoring behaviors, and informational exposure, alongside medication adherence measured using the Romanian-translated and validated ARMS (Adherence to Refills and Medications Scale). Five machine learning models were developed to predict non-adherence, defined by ARMS quartile-based thresholds. The models included Logistic Regression, Random Forest, and boosting algorithms (CatBoost, LightGBM, and XGBoost). Models were evaluated based on their ability to stratify patients according to adherence risk. Results: A total of 79.13% of respondents had an ARMS Score ≥ 15, indicating a high prevalence of suboptimal adherence. Better adherence was statistically associated (adjusted for age and sex) with more frequent blood pressure self-monitoring, a reduced salt intake, fewer daily supplements, more frequent reading of medication leaflets, and the receipt of specific information from pharmacists. Among the ML models, CatBoost achieved the highest ROC AUC Scores across the non-adherence classifications, although none exceeded 0.75. Conclusions: Several machine learning models were developed and validated to estimate levels of medication non-adherence. While the performance was moderate, the results demonstrate the potential of AI in identifying and stratifying patients by adherence profiles. Notably, to our knowledge, this study represents the first application of permutation and SHapley Additive exPlanations feature importance in combination with probability-based adherence stratification, offering a novel framework for predictive adherence modelling. Full article

Digital transformation is reshaping the healthcare field by streamlining diagnostic workflows and improving disease management. Within this transformation, Digital Twins (DTs), which are virtual representations of physical systems continuously updated by real-world data, stand out for their ability to capture the complexity of human physiology and behavior. When coupled with Artificial Intelligence (AI), DTs enable data-driven experimentation, precise diagnostic support, and predictive modeling without posing direct risks to patients. However, their integration into healthcare requires careful consideration of ethical, regulatory, and safety constraints in light of the sensitivity and nonlinear nature of human data. In this review, we examine recent progress in DTs over the past seven years and explore broader trends in AI-augmented DTs, focusing particularly on movement rehabilitation. Our goal is to provide a comprehensive understanding of how DTs bolstered by AI can transform healthcare delivery, medical research, and personalized care. We discuss implementation challenges such as data privacy, clinical validation, and scalability along with opportunities for more efficient, safe, and patient-centered healthcare systems. By addressing these issues, this review highlights key insights and directions for future research to guide the proactive and ethical adoption of DTs in healthcare. Full article