ABSTRACT
Problem: Chest radiography is a crucial tool for diagnosing thoracic disorders, but interpretation errors and a lack of qualified practitioners can cause delays in treatment. Aim: This study aimed to develop a reliable multi-classification artificial intelligence (AI) tool to improve the accuracy and efficiency of chest radiograph diagnosis. Methods: We developed a convolutional neural network (CNN) capable of distinguishing among 26 thoracic diagnoses. The model was trained and externally validated using 795,055 chest radiographs from 13 datasets across 4 countries. Results: The CNN model achieved an average area under the curve (AUC) of 0.961 across all 26 diagnoses in the testing set. COVID-19 detection achieved perfect accuracy (AUC 1.000, [95% confidence interval {CI}, 1.000 to 1.000]), while effusion or pleural effusion detection showed the lowest accuracy (AUC 0.8453, [95% CI, 0.8417 to 0.8489]). In external validation, the model demonstrated strong reproducibility and generalizability within the local dataset, achieving an AUC of 0.9634 for lung opacity detection (95% CI, 0.9423 to 0.9702). The CNN outperformed both radiologists and nonradiological physicians, particularly in trans-device image recognition. Even for diseases not specifically trained on, such as aortic dissection, the AI model showed considerable scalability and enhanced diagnostic accuracy for physicians of varying experience levels (all P < 0.05). Additionally, our model exhibited no gender bias (P > 0.05). Conclusion: The developed AI algorithm, now available as professional web-based software, substantively improves chest radiograph interpretation. This research advances medical imaging and offers substantial diagnostic support in clinical settings.
ABSTRACT
Background and objectives: Comprehensive data analyses in heart failure research can provide academics with information and help policymakers formulate relevant policies. We collected data from reports published between 1945 and 2021 to identify research topics, trends, and cross-domains in the heart failure disease literature. Methods: Text fragments were extracted and clustered from the titles and abstracts in 270617 publications using artificial intelligence techniques. Two algorithms were used to corroborate the results and ensure that they were reliable. Experts named themes and document clusters based on the results of these semiautomated methods. Using consistent methods, we identified and flagged 107 heart failure topics and 16 large document clusters (divided into two groups by time). The annual vocabularies of research hotspots were calculated to draw attention to niche research fields. Results: Clinical research is an expanding field, followed by basic research and population research. The most frequently raised issues were intensive care treatment for heart failure, applications of artificial intelligence technologies, cardiac assist devices, stem cells, genetics, and regional distribution and use of heart failure-related health care. Risk scoring and classification, care for patients, readmission, health economics of treatment and care, and cell regeneration and signaling pathways were among the fastest-growing themes. Drugs, signaling pathways, and biomarkers were all crucial issues for clinical and basic research in the entire population. Studies on intelligent medicine and telemedicine, interventional therapy for valvular disease, and novel coronavirus have emerged recently. Conclusion: Clinical and population research is increasingly focusing on the customization of intelligent treatments, improving the quality of patients' life, and developing novel treatments. Basic research is increasingly focusing on regenerative medicine, translational medicine, and signaling pathways. Additionally, each research field exhibits mutual fusion characteristics. Medical demands, new technologies, and social support are all potential drivers for these changes.
ABSTRACT
INTRODUCTION AND OBJECTIVES: This study aimed to retrospectively analyze the anatomical characteristics and classification of multiple coronary artery fistulas (MCAFs), and to compare the outcomes of transcatheter closure between MCAFs and single fistulas. METHODS: All patients who underwent attempts at transcatheter closure of coronary artery fistulas (CAFs) at Fuwai Hospital from 2010 to 2023 were retrospectively reviewed. Patients were categorized into single fistula and MCAFs groups, and anatomical characteristics and transcatheter closure outcomes were compared between the 2 groups. RESULTS: This retrospective study included 146 patients who underwent attempted transcatheter closure of CAFs, with a 14.38% failure rate. Among the 146 patients with CAFs, 32.19% were identified as having MCAFs, with types I, II, and III constituting 40.43%, 42.55%, and 17.02%, respectively. Unlike single fistulas, which predominantly originated from the right coronary artery and terminated in the left ventricle, MCAFs mainly had simultaneous origins from the right coronary artery and left anterior descending artery (29.79%), and predominantly drained into the pulmonary artery (70.21%), with a notable prevalence of plexus-like morphology (38.3% vs 2.02%, P<.001). The success rate of transcatheter closure was significantly lower for multiple fistulas compared with single fistula (64.29% vs 84.34%, P=.011). Multivariate regression analysis indicated that the risk of closure failure for MCAFs was 2.64 times that of single fistulas. CONCLUSIONS: MCAFs are common among CAFs and can be classified into 3 types based on the number and location of their origins and terminations. The risk of failure of transcatheter closure is significantly higher in MCAFs than in single fistulas.
Subject(s)
Cardiac Catheterization , Heart Valve Prosthesis Implantation , Mitral Valve Insufficiency , Mitral Valve , Recovery of Function , Severity of Illness Index , Humans , Mitral Valve Insufficiency/physiopathology , Mitral Valve Insufficiency/surgery , Mitral Valve Insufficiency/diagnostic imaging , Heart Valve Prosthesis Implantation/instrumentation , Heart Valve Prosthesis Implantation/adverse effects , Treatment Outcome , Mitral Valve/physiopathology , Mitral Valve/surgery , Mitral Valve/diagnostic imaging , Cardiac Catheterization/instrumentation , Cardiac Catheterization/adverse effects , Time Factors , Risk Factors , Female , Male , Heart Valve Prosthesis , Aged , Hemodynamics , Aged, 80 and overABSTRACT
AIMS: This study aimed to analyse the global prevalence and disability trends of heart failure (HF) from 1990 to 2019, considering both sexes and country-specific economic strata. METHODS: This study conducted a secondary analysis employing data from the Global Burden of Disease (GBD) study. The analysis is stratified by sex and Socio-demographic Index (SDI) levels. Through age-period-cohort and Joinpoint regression analyses, we investigated the temporal trends in HF prevalence and years lived with disability (YLDs) during this period. RESULTS: Between 1990 and 2019, the global prevalence of HF surged by 106.3% (95% uncertainty interval: 99.3% to 114.3%), reaching 56.2 million cases in 2019. While all-age prevalence and YLDs increased over the 30 year span, age-standardized rates decreased by 2019. Countries with higher SDI experienced a more pronounced percentage decrease compared with those with lower SDI. Longitudinal analysis revealed an overall improvement in both prevalence and YLDs for HF, albeit with notable disparities between SDI quintiles and sexes. Ischaemic heart disease and hypertensive heart disease emerged as the most rapidly increasing and primarily contributing causes of HF, albeit with variations observed across different countries. The average annual percentage change for prevalence and YLDs over the period was -0.26% and -0.25%, respectively. CONCLUSIONS: This study offers valuable insights into the global burden of HF, considering factors such as population aging, regional disparities, sex differences and aetiological variations. The findings hold significant implications for healthcare planning and resource allocation. Continued assessment of these trends and innovative strategies for HF prevention and management are crucial for addressing this pressing global health concern.
Subject(s)
Cardiac Catheterization , Robotic Surgical Procedures , Humans , Cardiac Catheterization/instrumentation , Treatment Outcome , Robotic Surgical Procedures/instrumentation , Mitral Valve/surgery , Mitral Valve/diagnostic imaging , Mitral Valve/physiopathology , Mitral Valve Insufficiency/surgery , Mitral Valve Insufficiency/diagnostic imaging , Mitral Valve Insufficiency/physiopathology , Ultrasonography, Interventional , Echocardiography, Transesophageal , Male , Heart Valve Prosthesis Implantation/instrumentation , Heart Valve Prosthesis Implantation/adverse effects , Aged , FemaleABSTRACT
The transcatheter edge-to-edge mitral valve repair (TEER) has been recommended as a reliable treatment option for selected patients with severe degenerative and functional mitral regurgitation (MR). Although MR patients with rheumatic etiology were excluded from two significant trials (EVEREST II and COAPT) that established a role for the TEER in degenerative and functional MR. However, it has been reported that the TEER procedure could be safely and effectively performed in carefully selected rheumatic MR patients. Therefore, we share a case report of successfully treating severe rheumatic MR using a novel-designed TEER system (JensClipTM).
Subject(s)
Cardiac Catheterization , Heart Valve Prosthesis Implantation , Mitral Valve Insufficiency , Mitral Valve , Rheumatic Heart Disease , Severity of Illness Index , Humans , Mitral Valve Insufficiency/diagnostic imaging , Mitral Valve Insufficiency/surgery , Mitral Valve Insufficiency/physiopathology , Mitral Valve Insufficiency/etiology , Cardiac Catheterization/instrumentation , Treatment Outcome , Rheumatic Heart Disease/diagnostic imaging , Rheumatic Heart Disease/surgery , Rheumatic Heart Disease/physiopathology , Rheumatic Heart Disease/therapy , Mitral Valve/diagnostic imaging , Mitral Valve/surgery , Mitral Valve/physiopathology , Heart Valve Prosthesis Implantation/instrumentation , Heart Valve Prosthesis Implantation/adverse effects , Female , Echocardiography, Transesophageal , Echocardiography, Doppler, Color , Prosthesis Design , Heart Valve Prosthesis , Middle Aged , MaleABSTRACT
Transcatheter edge-to-edge repair (TEER) has been established as a safe and effective option for treating patients with severe symptomatic degenerative mitral regurgitation (MR) who are at prohibitive surgical risk. However, the significant cost presents a considerable disease burden in low-income countries. This case details the treatment of a high-surgical-risk patient with severe degenerative MR by using the GeminiOne (Peijia Medical) system-a novel Chinese TEER device.
ABSTRACT
Transcatheter intervention has been the preferred treatment for congenital structural heart diseases by implanting occluders into the heart defect site through minimally invasive access. Biodegradable polymers provide a promising alternative for cardiovascular implants by conferring therapeutic function and eliminating long-term complications, but inducing in situ cardiac tissue regeneration remains a substantial clinical challenge. PGAG (polydioxanone/poly (l-lactic acid)-gelatin-A5G81) occluders are prepared by covalently conjugating biomolecules composed of gelatin and layer adhesive protein-derived peptides (A5G81) to the surface of polydioxanone and poly (l-lactic acid) fibers. The polymer microfiber-biomacromolecule-peptide frame with biophysical and biochemical cues could orchestrate the biomaterial-host cell interactions, by recruiting endogenous endothelial cells, promoting their adhesion and proliferation, and polarizing immune cells into anti-inflammatory phenotypes and augmenting the release of reparative cytokines. In a porcine atrial septal defect (ASD) model, PGAG occluders promote in situ tissue regeneration by accelerating surface endothelialization and regulating immune response, which mitigate inflammation and fibrosis formation, and facilitate the fusion of occluder with surrounding heart tissue. Collectively, this work highlights the modulation of cell-biomaterial interactions for tissue regeneration in cardiac defect models, ensuring endothelialization and extracellular matrix remodeling on polymeric scaffolds. Bioinspired cell-material interface offers a highly efficient and generalized approach for constructing bioactive coatings on medical devices.
Subject(s)
Gelatin , Septal Occluder Device , Animals , Swine , Gelatin/chemistry , Polydioxanone , Endothelial Cells , Polymers , Biocompatible Materials , Lactic Acid , PeptidesABSTRACT
BACKGROUND: Previous studies have examined the impact of antithrombotic agents on Patent Foramen Ovale (PFO) in relation to migraine. However, differences in effectiveness of different antithrombotic agents and traditional migraine medications are not known. METHODS/DESIGN: This study is an investigator-initiated, randomized, multicenter, single-masked (outcomes assessor), and active-controlled parallel-group trial (ClinicalTrials.gov Identifier: NCT05546320), with the objective of evaluating the prevention efficacy of antithrombotic agents compared to first-line migraine medication in PFO patients. The trial involves 1,000 migraine patients with a right-to-left shunt at the atrial level, randomized in a 1:1:1:1 fashion to receive either aspirin 300 mg QD, clopidogrel 75 mg QD, rivaroxaban 20 mg QD, or the active-control metoprolol 25 mg BID. The primary efficacy end point is the response rate, defined as a 50% or greater reduction in the average migraine attack days per month or in the average number of migraine attacks per month at 12-week visit compared to baseline. CONCLUSIONS: The COMPETE trial aims to provide valuable insights into the comparative effectiveness of antithrombotic agents and standard migraine therapies in patients with PFO. This study holds the promise of advancing treatment approaches for individuals having migraines associated with PFO, thus addressing an important gap in current migraine management strategies.
Subject(s)
Foramen Ovale, Patent , Migraine Disorders , Humans , Foramen Ovale, Patent/complications , Foramen Ovale, Patent/drug therapy , Fibrinolytic Agents , Migraine Disorders/complications , Migraine Disorders/drug therapy , Research Design , Treatment OutcomeABSTRACT
BACKGROUND: The global and national burden of rheumatic mitral valve disease (MVD) has been well studied and estimated before. However, little is known about non-rheumatic degenerative MVD. Therefore, this study aimed to assess the trends in non-rheumatic degenerative MVD (NRDMVD) epidemiology, with an emphasis on NRDMVD mortality, leading risk factors, and their associations with age, period, and birth cohort. METHODS: Using the data derived from the Global Burden of Disease Study 2019, including prevalence, mortality, and disability-adjusted life years, we analyzed the burden of NRDMVD and the detailed trends of NRDMVD mortality over the past 30 years in 204 countries and territories by implementing the age-period-cohort framework. RESULTS: Globally, the number of deaths due to NRDMVD increased from 5695.89 (95% uncertainty interval [UI]: 5405.19 to 5895.4) × 1000 in 1990 to 9137.79 (95% UI: 8395.68 to 9743.55) × 1000 in 2019. The all-age mortality rate increased from 106.47 (95% UI: 101.03 to 110.2) per 100,000 to 118.1 (95% UI: 108.51 to 125.93) per 100,000, whereas the age-standardized mortality rate decreased from 170.45 (95% UI: 159.61 to 176.94) per 100,000 to 117.95 (95% UI: 107.83 to 125.92) per 100,000. The estimated net drift of mortality per year was -1.1% (95% confidence interval: -1.17 to -1.04). The risk of death due to NRDMVD increased with age, reaching its peak after 85 years old globally. Despite female patients being associated with lower local drift than male patients, no significant gender differences were observed in the age effect across countries and regions for all sociodemographic index (SDI) levels, except low-SDI regions. CONCLUSIONS: We estimated the global disease prevalence of and mortality due to NRDMVD over approximately a 30-year period. The health-related burden of NRDMVD has declined worldwide; however, the condition persisted in low-SDI regions. Moreover, higher attention should be paid to female patients.
Subject(s)
Global Burden of Disease , Mitral Valve , Humans , Male , Female , Aged, 80 and over , Risk Factors , Sex Distribution , Global Health , Prevalence , Cohort Studies , Quality-Adjusted Life Years , IncidenceABSTRACT
Background: As the global burden of hypertension continues to increase, early diagnosis and treatment play an increasingly important role in improving the prognosis of patients. In this study, we developed and evaluated a method for predicting abnormally high blood pressure (HBP) from infrared (upper body) remote thermograms using a deep learning (DL) model. Methods: The data used in this cross-sectional study were drawn from a coronavirus disease 2019 (COVID-19) pilot cohort study comprising data from 252 volunteers recruited from 22 July to 4 September 2020. Original video files were cropped at 5 frame intervals to 3,800 frames per slice. Blood pressure (BP) information was measured using a Welch Allyn 71WT monitor prior to infrared imaging, and an abnormal increase in BP was defined as a systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg. The PanycNet DL model was developed using a deep neural network to predict abnormal BP based on infrared thermograms. Results: A total of 252 participants were included, of which 62.70% were male and 37.30% were female. The rate of abnormally high HBP was 29.20% of the total number. In the validation group (upper body), precision, recall, and area under the receiver operating characteristic curve (AUC) values were 0.930, 0.930, and 0.983 [95% confidence interval (CI): 0.904-1.000], respectively, and the head showed the strongest predictive ability with an AUC of 0.868 (95% CI: 0.603-0.994). Conclusions: This is the first technique that can perform screening for hypertension without contact using existing equipment and data. It is anticipated that this technique will be suitable for mass screening of the population for abnormal BP in public places and home BP monitoring.
ABSTRACT
Recent studies suggest that pleiotropic effects may explain the genetic architecture of cardiovascular diseases (CVDs). We conducted a comprehensive gene-centric pleiotropic association analysis for ten CVDs using genome-wide association study (GWAS) summary statistics to identify pleiotropic genes and pathways that may underlie multiple CVDs. We found shared genetic mechanisms underlying the pathophysiology of CVDs, with over two-thirds of the diseases exhibiting common genes and single-nucleotide polymorphisms (SNPs). Significant positive genetic correlations were observed in more than half of paired CVDs. Additionally, we investigated the pleiotropic genes shared between different CVDs, as well as their functional pathways and distribution in different tissues. Moreover, six hub genes, including ALDH2, XPO1, HSPA1L, ESR2, WDR12, and RAB1A, as well as 26 targeted potential drugs, were identified. Our study provides further evidence for the pleiotropic effects of genetic variants on CVDs and highlights the importance of considering pleiotropy in genetic association studies.
ABSTRACT
Objective: Echocardiography (ECG) is the most common method used to diagnose heart failure (HF). However, its accuracy relies on the experience of the operator. Additionally, the video format of the data makes it challenging for patients to bring them to referrals and reexaminations. Therefore, this study used a deep learning approach to assist physicians in assessing cardiac function to promote the standardization of echocardiographic findings and compatibility of dynamic and static ultrasound data. Methods: A deep spatio-temporal convolutional model r2plus1d-Pan (trained on dynamic data and applied to static data) was improved and trained using the idea of "regression training combined with classification application," which can be generalized to dynamic ECG and static cardiac ultrasound views to identify HF with a reduced ejection fraction (EF < 40%). Additionally, three independent datasets containing 8976 cardiac ultrasound views and 10085 cardiac ultrasound videos were established. Subsequently, a multinational, multi-center dataset of EF was labeled. Furthermore, model training and independent validation were performed. Finally, 15 registered ultrasonographers and cardiologists with different working years in three regional hospitals specialized in cardiovascular disease were recruited to compare the results. Results: The proposed deep spatio-temporal convolutional model achieved an area under the receiveroperating characteristic curve (AUC) value of 0.95 (95% confidence interval [CI]: 0.947 to 0.953) on the training set of dynamic ultrasound data and an AUC of 1 (95% CI, 1 to 1) on the independent validation set. Subsequently, the model was applied to the static cardiac ultrasound view (validation set) with simultaneous input of 1, 2, 4, and 8 images of the same heart, with classification accuracies of 85%, 81%, 93%, and 92%, respectively. On the static data, the classification accuracy of the artificial intelligence (AI) model was comparable with the best performance of ultrasonographers and cardiologists with more than 3 working years (P = 0.344), but significantly better than the median level (P = 0.0000008). Conclusion: A new deep spatio-temporal convolution model was constructed to identify patients with HF with reduced EF accurately (< 40%) using dynamic and static cardiac ultrasound images. The model outperformed the diagnostic performance of most senior specialists. This may be the first HF-related AI diagnostic model compatible with multi-dimensional cardiac ultrasound data, and may thereby contribute to the improvement of HF diagnosis. Additionally, the model enables patients to carry "on-the-go" static ultrasound reports for referral and reexamination, thus saving healthcare resources.
ABSTRACT
AIM: To increase the comprehensive understanding of trends in the burden of cardiovascular disease (CVD) attributable to low physical activity in the Western Pacific Region. METHODS: Based on data from the Global Burden of Disease (GBD) study for the years 1990-2019, an age-period-cohort (APC) analysis was conducted to investigate trends in CVD-related mortality attributable to low physical activity in the Western Pacific Region and associations with age, period, and birth cohort. We also used joinpoint regression analysis to identify the periods with the most substantial changes. RESULTS: The Western Pacific Region witnessed a substantial increase in CVD deaths attributable to low physical activity, accompanied by a rise in all-age CVD-related mortality. However, the age-standardized death rate was lower in the region than the global level, highlighting the importance of considering the age composition of CVD burden in the region. Countries with higher SDI levels exhibited lower mortality than those with lower SDI levels. The longitudinal analysis using the APC model indicated an overall improvement in CVD-related mortality attributable to low physical activity in the region, but with differences between sexes and CVD subtypes. Specific period in which CVD-related mortality decreased significantly were 2011-2016, for the average annual percentage change for the period was -0.69%. CONCLUSION: The study highlights the significance of addressing low physical activity as a modifiable risk factor for CVD burden in the Western Pacific Region. Further research is essential to understand the factors contributing to inter-country variations, sex disparities, and CVD subtypes distinctions.
ABSTRACT
BACKGROUND: TP53 mutation heterogeneity should be considered when using TP53 as a predictive biomarker for anti-programmed death (ligand) 1 (PD-(L)1) monotherapy in lung adenocarcinoma (LUAD). However, whether TP53 variant allele frequency (VAF) should also be considered remains unknown. METHODS: Patients with LUAD from both published research and the local cohort were included to discover and validate the relationship between TP53 VAF and the efficacy of PD-(L)1 inhibitors. The Cancer Genome Atlas (TCGA) LUAD data were included for genomic, transcriptomic, and tumor microenvironment analysis. RESULTS: Among 159 patients in the discovery cohort, low TP53 VAF patients (VAF ≤ 25%) experienced significantly longer progression-free survival (PFS) than both high TP53 VAF (5.4 vs. 3.3 months; p = .021) and TP53-wild-type patients (5.4 vs. 2.5 months; p = .011). Multivariate Cox regression revealed low TP53 VAF as an independent biomarker of better efficacy. Among 50 patients in the combined validation cohort, median PFS of low TP53 VAF patients was also significantly longer than that of high TP53 VAF patients (12.0 vs. 2.1 months; p = .037). Analyzed with 469 TCGA LUAD samples, low TP53 VAF is associated with significantly higher PD-L1 expression, enrichment of gene sets related to T-cell activation, T cell-mediated immunity, and interferon-γ signaling pathways, and independently associated with more tumor-infiltrating CD8+ T cells compared with both high TP53 VAF and TP53-wild type. CONCLUSIONS: TP53 VAF should also be considered when using TP53 as a predictive biomarker. Only low TP53 VAF is independently associated with better efficacy of anti-PD-(L)1 monotherapy, which may result from higher PD-L1 expression and more tumor-infiltrating CD8+ T cells.
Subject(s)
Adenocarcinoma of Lung , Lung Neoplasms , Humans , B7-H1 Antigen/genetics , CD8-Positive T-Lymphocytes , Ligands , Adenocarcinoma of Lung/drug therapy , Adenocarcinoma of Lung/genetics , Biomarkers, Tumor/genetics , Lung Neoplasms/drug therapy , Lung Neoplasms/genetics , Gene Frequency , Mutation , Tumor Microenvironment/genetics , Tumor Suppressor Protein p53/geneticsABSTRACT
PURPOSE: Studies relating to the right ventricle (RV) are inadequate, and specific diagnostic algorithms still need to be improved. This essay is designed to make exploration and verification on an algorithm of deep learning based on imaging and clinical data to detect RV abnormalities. METHODS: The Automated Cardiac Diagnosis Challenge dataset includes 20 subjects with RV abnormalities (an RV cavity volume which is higher than 110 mL/m2 or RV ejection fraction which is lower than 40%) and 20 normal subjects who suffered from both cardiac MRI. The subjects were separated into training and validation sets in a ratio of 7:3 and were modeled by utilizing a nerve net of deep-learning and six machine-learning algorithms. Eight MRI specialists from multiple centers independently determined whether each subject in the validation group had RV abnormalities. Model performance was evaluated based on the AUC, accuracy, recall, sensitivity and specificity. Furthermore, a preliminary assessment of patient disease risk was performed based on clinical information using a nomogram. RESULTS: The deep-learning neural network outperformed the other six machine-learning algorithms, with an AUC value of 1 (95% confidence interval: 1-1) on both training group and validation group. This algorithm surpassed most human experts (87.5%). In addition, the nomogram model could evaluate a population with a disease risk of 0.2-0.8. CONCLUSIONS: A deep-learning algorithm could effectively identify patients with RV abnormalities. This AI algorithm developed specifically for right ventricular abnormalities will improve the detection of right ventricular abnormalities at all levels of care units and facilitate the timely diagnosis and treatment of related diseases. In addition, this study is the first to validate the algorithm's ability to classify RV abnormalities by comparing it with human experts.