Background/Objectives: Artificial intelligence (AI), machine-learning (ML), and deep-learning (DL) models are increasingly used for prognostic prediction in intensive care unit (ICU) sepsis, but their clinical readiness remains uncertain. This systematic review aimed to evaluate AI-, ML-, and DL-based models for mortality and clinically relevant outcome prediction in adult ICU patients with sepsis or septic shock. Methods: PubMed/MEDLINE, Scopus, and the Cochrane Library were searched up to April 2026. Eligible studies included adult ICU sepsis or septic shock cohorts evaluating AI/ML/DL-based prognostic models. Screening, full-text assessment, and data extraction were performed independently by two reviewers. Outcomes, model families, validation strategies, discrimination, calibration, clinical utility, explainability, comparative performance versus conventional severity scores, risk of bias, and reporting completeness were synthesized. Risk of bias was assessed using PROBAST domains supplemented by PROBAST + AI considerations, and reporting completeness was evaluated according to TRIPOD/TRIPOD + AI domains. Results: Seventy-five studies were included, comprising 50 PubMed-derived and 25 additional Scopus-derived studies. AUROC or C-statistic was extractable in 64 studies, external validation was reported in 27, prospective evaluation in three, calibration in 38, decision-curve analysis or clinical utility assessment in 37, and explainability in 64. Across 17 directly extractable within-study comparisons from nine studies, AI/ML models usually, but not uniformly, achieved higher discrimination than conventional severity scores, with a median paired ΔAUROC of +0.108 (IQR, +0.082 to +0.148; range, −0.013 to +0.203). Externally validated fixed-horizon models showed clinically relevant but heterogeneous discrimination across sepsis phenotypes, with stronger evidence in selected sepsis-induced coagulopathy cohorts and more variable transportability in respiratory and liver-injury subgroups. However, 45 studies were judged at high risk of bias, mainly because of limitations in the analysis domain. Conclusions: AI/ML models for adult ICU sepsis show a recurrent signal of prognostic discrimination and often perform comparably to or better than conventional severity scores in directly extractable within-study comparisons; however, this signal should be interpreted cautiously given clinical and methodological heterogeneity, limited prospective validation, incomplete calibration, and frequent high risk of bias. The strongest evidence comes from externally validated, phenotype-specific models, although routine clinical implementation remains limited by heterogeneous endpoints, incomplete calibration, insufficient prospective validation, and scarce workflow-level evaluation. Future studies should shift from retrospective AUROC optimization toward calibrated, externally validated, clinically actionable, and workflow-integrated decision-support tools tested in prospective ICU settings.

The complete transposition of the great arteries (C-TGA) is a congenital cardiac anomaly characterized by the reversal of the main arteries. Early detection and precise management are crucial for optimal outcomes. This review emphasizes the integral role of multimodal imaging, including fetal echocardiography, transthoracic echocardiography (TTE), cardiovascular magnetic resonance (CMR), and cardiac computed tomography (CCT) in the diagnosis, treatment planning, and long-term follow-up of C-TGA. Fetal echocardiography plays a pivotal role in prenatal detection, enabling early intervention strategies. Despite technological advances, the detection rate varies, highlighting the need for improved screening protocols. TTE remains the cornerstone for initial diagnosis, surgical preparation, and postoperative evaluation, providing essential information on cardiac anatomy, ventricular function, and the presence of associated defects. CMR and CCT offer additional value in C-TGA assessment. CMR, free from ionizing radiation, provides detailed anatomical and functional insights from fetal life into adulthood, becoming increasingly important in evaluating complex cardiac structures and post-surgical outcomes. CCT, with its high-resolution imaging, is indispensable in delineating coronary anatomy and vascular structures, particularly when CMR is contraindicated or inconclusive. This review advocates for a comprehensive imaging approach, integrating TTE, CMR, and CCT to enhance diagnostic accuracy, guide therapeutic interventions, and monitor postoperative conditions in C-TGA patients. Such a multimodal strategy is vital for advancing patient care and improving long-term prognoses in this complex congenital heart disease.

Congenital heart diseases (CHDs) represent a heterogeneous group of congenital defects, with high prevalence worldwide. Non-invasive imaging is essential to guide medical and surgical planning, to follow the patient over time in the evolution of the disease, and to reveal potential complications of the chosen treatment. The application of cardiac magnetic resonance imaging (CMRI) in this population allows for obtaining detailed information on the defects without the necessity of ionizing radiations. This review emphasizes the central role of CMR in the overall assessment of CHDs, considering also the limitations and challenges of this imaging technique. CMR, with the application of two-dimensional (2D) and tri-dimensional (3D) steady-state free precession (SSFP), permits the obtaining of very detailed and accurate images about the cardiac anatomy, global function, and volumes’ chambers, giving essential information in the intervention planning and optimal awareness of the postoperative anatomy. Nevertheless, CMR supplies tissue characterization, identifying the presence of fat, fibrosis, or oedema in the myocardial tissue. Using a contrast agent for angiography sequences or 2D/four-dimensional (4D) flows offers information about the vascular, valvular blood flow, and, in general, the cardiovascular system hemodynamics. Furthermore, 3D SSFP CMR acquisitions allow the identification of coronary artery abnormalities as an alternative to invasive angiography and cardiovascular computed tomography (CCT). However, CMR requires expertise in CHDs, and it can be contraindicated in patients with non-conditional devices. Furthermore, its relatively longer acquisition time and the necessity of breath-holding may limit its use, particularly in children under eight years old, sometimes requiring anesthesia. The purpose of this review is to elucidate the application of CMR during the pediatric age.

Cardiac arrhythmias in the pediatric population represent both diagnostic and therapeutic challenges due to their heterogeneous etiologies, age-dependent clinical presentations, and the variety of management strategies required. Although many electrophysiological mechanisms are shared with adults, several arrhythmias are unique to infancy and adolescence or are associated with congenital heart disease and inherited channelopathies. Clinical presentation varies considerably with age and may be unclear or nonspecific, especially in neonates and infants. In some cases, arrhythmias are benign and self-limiting; in others, they may represent the earliest manifestation of an underlying cardiac disorder, potentially progressing to hemodynamic instability, serious arrhythmic events, or sudden cardiac death. Recent advancements in electrophysiology, catheter ablation, and device technology have broadened the therapeutic landscape, improving outcomes even in neonatal and infant patients. This review aims to summarize current knowledge on pediatric arrhythmias, with a focus on clinical recognition, diagnostic evaluation, and personalized treatment strategies.

Although advances in medical and surgical management have significantly improved clinical outcomes, infective endocarditis (IE) remains a significant threat to patients with congenital heart disease (CHD). The complexity of cardiac anatomy, the presence of prosthetic materials, and the emergence of novel pathogens pose unique diagnostic challenges in this specific population. However, the use of personalized imaging, integrating the strengths of each modality, has the potential to refine the diagnostic process, thereby optimizing diagnostic accuracy, guiding therapeutic decisions, and, ultimately, improving patient clinical outcomes. This review delves into the critical role of the multimodality imaging approach in the care of patients with IE and CHD, underscoring the importance of tailored and patient-centered management strategies in this vulnerable cohort.