Background: Heart failure (HF) affects over 64 million people globally, imposing substantial morbidity, mortality, and economic burdens. Despite advances in guideline-directed therapies, adherence remains suboptimal due to low health literacy and complex regimens. ChatGPT, an advanced large language model by OpenAI, offers conversational capabilities that could enhance HF education, management, and research. This systematic review synthesizes evidence on ChatGPT’s applications in HF, evaluating its accuracy in patient education and question-answering, enhancing readability, and clinical documentation/symptom extraction. Methods: Following PRISMA guidelines, we searched PubMed, Embase, and Cochrane up to July 2025 using the terms “ChatGPT” and “heart failure”. Inclusion: Studies on ChatGPT (3.5 or 4) in HF contexts, such as in education, readability and symptom extraction. Exclusion: Non-HF or non-ChatGPT AI. Data extraction covered design, objectives, methods, and outcomes. Thematic synthesis was applied. Results: From 59 records, 7 observational studies were included. Themes included patient education/question-answering (n = 5), readability enhancement (n = 2), and clinical documentation/symptom extraction (n = 1). Accuracy ranged 78–98%, with high reproducibility; readability improved to 6th–7th grade levels; and symptom extraction achieved up to 95% F1 score, outperforming traditional machine learning baselines. Conclusions: ChatGPT shows promise in HF care but requires further randomized validation for outcomes and bias mitigation.

Background/objectives: Acute coronary syndrome (ACS) is a major global health issue primarily caused by the rupture or erosion of vulnerable coronary plaques. Non-invasive identification of these plaques through coronary computed tomography angiography (CCTA) can be improved with radiomics, which extracts and analyses quantitative features from medical images. This systematic review aims to comprehensively evaluate the literature surrounding the role of radiomics in assessing coronary plaques via CCTA. Methods: A systematic search of Medline, EMBASE, and Web of Science was conducted up to July 2024. Nine studies met the inclusion criteria, and their methodological quality was assessed using the radiomic quality score (RQS) and the QUADAS-2 tool. Results: All studies that evaluated radiomic models for plaque vulnerability reported area under the curve (AUC) values exceeding 0.7, indicating at least modest diagnostic performance. In the four studies that made direct comparisons, radiomic models consistently outperformed conventional CCTA markers. However, RQS scores ranged from 2.7% to 41.7%, reflecting variability in study quality and underscoring the need for more robust validation. Conclusions: Radiomics has the potential to enhance CCTA-based identification of vulnerable coronary plaques, offering a promising non-invasive approach to predicting major adverse cardiovascular events. However, the current body of research is limited by the lack of external validation, reliance on small, single-centre retrospective studies, and methodological inconsistencies, which impact the generalisability and reproducibility of findings. Future research should prioritise prospective, multi-centre studies with standardised protocols and rigorous validation frameworks to effectively incorporate radiomics into clinical practice.

Background: Transcatheter aortic valve implantation (TAVI) is widely performed for severe aortic stenosis (AS), often accompanied by coronary artery disease (CAD). The optimal management of CAD in TAVI patients remains uncertain. This study reviews the evidence on percutaneous coronary intervention (PCI) timing in TAVI, and proposes an algorithmic approach for CAD management. Methods: A comprehensive search of PubMed, EMBASE and Cochrane identified studies comparing PCI timing strategies in TAVI patients. Results: Thirteen studies with 15,412 participants were included. Mortality at 30 days (OR 5.70; 95% CI [1.34–24.36]) and 2 years (OR 4.40; 95% CI [2.60–7.44]) were significantly higher in the combined pre-TAVI and concomitant PCI group than in the post-TAVI cohort. Rates of other periprocedural complications, such as stroke and bleeding, varied across studies. Conclusion: There is no clear consensus on PCI timing in TAVI due to a lack of high-quality randomised data. An individualised, algorithmic approach is proposed for managing CAD in patients undergoing TAVI.

Background: Heart failure (HF) affects over 64 million people globally, imposing substantial morbidity, mortality, and economic burdens. Despite advances in guideline-directed therapies, adherence remains suboptimal due to low health literacy and complex regimens. ChatGPT, an advanced large language model by OpenAI, offers conversational capabilities that could enhance HF education, management, and research. This systematic review synthesizes evidence on ChatGPT's applications in HF, evaluating its accuracy in patient education and question-answering, enhancing readability, and clinical documentation/symptom extraction. Methods: Following PRISMA guidelines, we searched PubMed, Embase, and Cochrane up to July 2025 using the terms "ChatGPT" and "heart failure." Inclusion: Studies on ChatGPT (3.5 or 4) in HF contexts, such as in education, readability and symptom extraction. Exclusion: Non-HF or non-ChatGPT AI. Data extraction covered design, objectives, methods, outcomes. Thematic synthesis applied. Results: From 59 records, 7 observational studies were included. Themes included: patient education/question-answering (n=5), readability enhancement (n=2), clinical documentation/symptom extraction (n=1). Accuracy ranged 78-98%, with high reproducibility; readability improved to 6th-7th grade levels; and symptom extraction achieved up to 95% F1 score, outperforming traditional machine learning baselines.Conclusions: ChatGPT shows promise in HF care but requires further randomized validation for outcomes and bias mitigation.

Background: Pericoronary adipose tissue (PCAT) radiomics on coronary computed tomography angiography (CCTA) may capture inflammatory and structural information beyond conventional attenuation measures. We assessed whether PCAT radiomics predicts future coronary events or quantifies plaque progression. Methods: PubMed/MEDLINE, Embase, and Web of Science were searched from inception to 29 August 2025. Eligible studies included adults undergoing CCTA with quantitative PCAT, or anatomically equivalent coronary perivascular adipose tissue (PVAT), radiomics, and longitudinal clinical or serial-CCTA outcomes. Findings were synthesised narratively because no common estimand supported meta-analysis. Methodological quality was assessed using QUADAS-2 and the Radiomics Quality Score (RQS). Results: Eleven full-text studies, representing at least 6307 participants were included; one study also analysed 127 plaques in 97 patients, and one reported a separate 202-person model-development sample whose overlap with its 1575-person cohort evaluation was not quantified. Eight studies evaluated coronary clinical events, and three evaluated plaque progression. Seven clinical-event studies reported better discrimination for radiomics-based or integrated models than at least one conventional comparator, although calibration and clinical-utility reporting were inconsistent. Two studies included clearly independent external validation. QUADAS-2 identified high patient-selection risk in 4/11 studies; RQSs ranged from 10/36 to 15/36 (mean: 12.5/36). Conclusions: PCAT radiomics shows preliminary prognostic potential, but confidence in the evidence is low because of retrospective designs, heterogeneous outcomes and workflows, restricted geographic sampling, and limited independent external validation. Current evidence is insufficient for routine clinical decision-making. Prospective multicentre validation with standardised acquisition, analysis, reporting, and clinically consistent endpoints is required.

Background: Myocardial scar and fibrosis predict adverse cardiac outcomes. Late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) is the reference standard for detection. However, it requires gadolinium-based contrast agents (GBCAs), which may be unsuitable for some patients. Cine balanced steady-state free precession (bSSFP) sequences are universally acquired in routine CMR. They may enable contrast-free scar detection via radiomics analysis. Aim: To systematically review the diagnostic accuracy of cine CMR radiomics for myocardial scar or fibrosis detection. The reference standard is visual or threshold-based LGE. Methods: This review followed PRISMA guidelines and was registered in PROSPERO (CRD420251121699). We searched MEDLINE, Embase, and Cochrane Library up to 8 August 2025. Eligible studies compared cine CMR radiomics with LGE-based assessment in patients with suspected or known scar/fibrosis. Quality was assessed using QUADAS-2 and Radiomics Quality Score (RQS). Results: Five retrospective studies (n = 1,484) were included. Two focused on myocardial infarction, two on hypertrophic cardiomyopathy, and one on ischaemic versus dilated cardiomyopathy. Diagnostic performance was good to excellent (AUC 0.74–0.96). Methodological heterogeneity was substantial in reference standards, segmentation, preprocessing, feature selection, and modelling. Only one study used external validation. QUADAS-2 showed high bias risk in patient selection and index test domains. RQS scores were low (30–42%), indicating limited reproducibility and validation. Conclusions: Cine CMR radiomics shows promise as a non-contrast alternative for detecting myocardial scar and fibrosis. However, methodological standardisation, multicentre validation, and prospective studies are needed before clinical adoption.