RESUMEN
AIMS: The randomized, double-blind, placebo-controlled HOPE-HF trial assessed the benefit of atrio-ventricular (AV) delay optimization delivered using His bundle pacing. It recruited patients with left ventricular ejection fraction ≤40%, PR interval ≥200 ms, and baseline QRS ≤140 ms or right bundle branch block. Overall, there was no significant increase in peak oxygen uptake (VO2max) but there was significant improvement in heart failure specific quality of life. In this pre-specified secondary analysis, we evaluated the impact of baseline PR interval, echocardiographic E-A fusion, and the magnitude of acute high-precision haemodynamic response to pacing, on outcomes. METHODS AND RESULTS: All 167 randomized participants underwent measurement of PR interval, acute haemodynamic response at optimized AV delay, and assessment of presence of E-A fusion. We tested the impact of these baseline parameters using a Bayesian ordinal model on VO2max, quality of life and activity measures. There was strong evidence of a beneficial interaction between the baseline acute haemodynamic response and the blinded benefit of pacing for VO2 (Pr 99.9%), Minnesota Living With Heart Failure (MLWHF) (Pr 99.8%), MLWHF physical limitation score (Pr 98.9%), EQ-5D visual analogue scale (Pr 99.6%), and exercise time (Pr 99.4%). The baseline PR interval and the presence of baseline E-A fusion did not have this reliable ability to predict the clinical benefit of pacing over placebo across multiple endpoints. CONCLUSIONS: In the HOPE-HF trial, the acute haemodynamic response to pacing reliably identified patients who obtained clinical benefit. Patients with a long PR interval (≥200 ms) and left ventricular impairment who obtained acute haemodynamic improvement with AV-optimized His bundle pacing were likely to obtain clinical benefit, consistent across multiple endpoints. Importantly, this gradation can be reliably tested for before randomization, but does require high-precision AV-optimized haemodynamic assessment to be performed.
RESUMEN
BACKGROUND: Global longitudinal strain (GLS) is reported to be more reproducible and prognostic than ejection fraction. Automated, transparent methods may increase trust and uptake. OBJECTIVES: The authors developed open machine-learning-based GLS methodology and validate it using multiexpert consensus from the Unity UK Echocardiography AI Collaborative. METHODS: We trained a multi-image neural network (Unity-GLS) to identify annulus, apex, and endocardial curve on 6,819 apical 4-, 2-, and 3-chamber images. The external validation dataset comprised those 3 views from 100 echocardiograms. End-systolic and -diastolic frames were each labelled by 11 experts to form consensus tracings and points. They also ordered the echocardiograms by visual grading of longitudinal function. One expert calculated global strain using 2 proprietary packages. RESULTS: The median GLS, averaged across the 11 individual experts, was -16.1 (IQR: -19.3 to -12.5). Using each case's expert consensus measurement as the reference standard, individual expert measurements had a median absolute error of 2.00 GLS units. In comparison, the errors of the machine methods were: Unity-GLS 1.3, proprietary A 2.5, proprietary B 2.2. The correlations with the expert consensus values were for individual experts 0.85, Unity-GLS 0.91, proprietary A 0.73, proprietary B 0.79. Using the multiexpert visual ranking as the reference, individual expert strain measurements found a median rank correlation of 0.72, Unity-GLS 0.77, proprietary A 0.70, and proprietary B 0.74. CONCLUSIONS: Our open-source approach to calculating GLS agrees with experts' consensus as strongly as the individual expert measurements and proprietary machine solutions. The training data, code, and trained networks are freely available online.
Asunto(s)
Consenso , Ecocardiografía , Interpretación de Imagen Asistida por Computador , Aprendizaje Automático , Redes Neurales de la Computación , Valor Predictivo de las Pruebas , Humanos , Fenómenos Biomecánicos , Conjuntos de Datos como Asunto , Tensión Longitudinal Global , Contracción Miocárdica , Variaciones Dependientes del Observador , Reproducibilidad de los Resultados , Reino Unido , Disfunción Ventricular Izquierda/diagnóstico por imagen , Disfunción Ventricular Izquierda/fisiopatología , Función Ventricular IzquierdaRESUMEN
BACKGROUND: Accurate measurements from cardiac magnetic resonance (CMR) images require precise positioning of scan planes and elimination of movement artefacts from arrhythmia or breathing. Unidentified or incorrectly managed artefacts degrade image quality, invalidate clinical measurements and decrease diagnostic confidence. Currently, radiographers must manually inspect each acquired image to confirm diagnostic quality and decide whether reacquisition or a change in sequences is warranted. We aimed to develop an artificial intelligence (AI) to provide continuous quality scores across different quality domains, and from these, determine whether cines are clinically adequate, require replanning, or warrant a change in protocol. METHODS: A three-dimensional convolutional neural network was trained to predict cine quality graded on a continuous scale by a level 3 CMR expert, focusing separately on planning and movement artefacts. It incorporated four distinct output heads for the assessment of image quality in terms of: (a, b, c)2-, 3- and 4-chamber misplanning, and (d)long- and short-axis arrhythmia/breathing artefact. Back-propagation was selectively performed across these heads based on the labels present for each cine. Each image in the testing set was reported by four level 3 CMR experts, providing a consensus on clinical adequacy. The AI's assessment of image quality and ability to identify images requiring replanning or sequence changes were evaluated with Spearman's rho and the area under receiver operating characteristic (AUROC) respectively. RESULTS: A total of 1940 cines across 1387 studies were included. On the test set of 383 cines, AI-judged image quality correlated strongly with expert judgement, with Spearman's rho of 0.84, 0.84, 0.81 and 0.81 for 2-, 3- and 4-chamber planning quality and the extent of arrhythmia or breathing artefacts, respectively. The AI also showed high efficacy in flagging clinically inadequate cines (AUROC 0.88, 0.93, 0.93 and for identifying replanning of 2-, 3- and 4-chamber cines, and 0.90 for identifying movement artefacts). CONCLUSION: AI can assess distinct domains of CMR cine quality and provide continuous quality scores that correlate closely with a consensus of experts. These ratings could be used to identify cases where reacquisition is warranted, and guide corrective actions to optimise image quality, including replanning, prospective gating, or real-time imaging.
RESUMEN
BACKGROUND: Neurocardiogenic syncope is a common condition with significant associated psychological and physical morbidity. The effectiveness of therapeutic options for neurocardiogenic syncope beyond placebo remains uncertain. METHODS: The primary endpoint was the risk ratio (RR) of spontaneously recurring syncope following any therapeutic intervention. We also examined the effect of blinding on treatment efficacy. We identified all randomised trials which evaluated the effect of any pharmacological, device-based or supportive intervention on patients with a history of syncope. A systematic search was conducted on Medline, Embase, PubMed databases and Cochrane Central Register for Controlled Trials from 1950 to 25 April 2023. Event rates, their RRs and 95% CIs were calculated, and a random-effects meta-analysis was conducted for each intervention. Data analysis was performed in R using RStudio. RESULTS: We identified 47 eligible trials randomising 3518 patients. Blinded trials assessing syncope recurrence were neutral for beta blockers, fludrocortisone and conventional dual-chamber pacing but were favourable for selective serotonin reuptake inhibitors (SSRIs) (RR 0.40, 95% CI 0.26 to 0.63, p<0.001), midodrine (RR 0.70, 95% CI 0.53 to 0.94, p=0.016) and closed-loop stimulation (CLS) pacing (RR 0.15, 95% CI 0.07 to 0.35, p<0.001). Unblinded trials reported significant benefits for all therapy categories other than beta blockers and consistently showed larger benefits than blinded trials. CONCLUSIONS: Under blinded conditions, SSRIs, midodrine and CLS pacing significantly reduced syncope recurrence. Future trials for syncope should be blinded to avoid overestimating treatment effects. PROSPERO REGISTRATION NUMBER: CRD42022330148.
Asunto(s)
Ensayos Clínicos Controlados Aleatorios como Asunto , Síncope Vasovagal , Humanos , Síncope Vasovagal/terapia , Síncope Vasovagal/diagnóstico , Ensayos Clínicos Controlados Aleatorios como Asunto/métodos , Resultado del Tratamiento , RecurrenciaRESUMEN
BACKGROUND: In stable coronary artery disease, 30% to 60% of patients remain symptomatic despite successful revascularization. Perhaps not all symptoms reported by a patient with myocardial ischemia are, in fact, angina. OBJECTIVES: This study sought to determine whether independent symptom verification using a placebo-controlled ischemic stimulus could distinguish which patients achieve greatest symptom relief from percutaneous coronary intervention (PCI). METHODS: ORBITA-STAR was a multicenter, n-of-1, placebo-controlled study in patients undergoing single-vessel PCI for stable symptoms. Participants underwent 4 episodes (60 seconds each) of low-pressure balloon occlusion across their coronary stenosis, randomly paired with 4 episodes of placebo inflation. Following each episode, patients reported the similarity of the induced symptom in comparison with their usual symptom. The similarity score ranged from -10 (placebo replicated the symptom more than balloon occlusion) to +10 (balloon occlusion exactly replicated the symptom). The primary endpoint was the ability of the similarity score to predict symptom relief with PCI. RESULTS: Fifty-one patients were recruited, aged 62.9 ± 8.6 years. The median fractional flow reserve was 0.68 (Q1-Q3: 0.57-0.79), and the instantaneous wave-free ratio was 0.80 (Q1-Q3: 0.48-0.89). The median similarity score was 3 (Q1-Q3: 0.875-5.25). The similarity score was a strong predictor of symptom improvement following PCI: a patient with an upper quartile similarity score of 5.25 was significantly more likely to have lower angina frequency at follow-up (OR: 8.01; 95% credible interval: 2.39-15.86) than a patient with a lower quartile similarity score of 0.875 (OR: 1.31; 95% credible interval: 0.71-1.99), Pr(difference) >99.9%. CONCLUSIONS: Similarity score powerfully predicted symptom improvement from PCI. These data lay the foundation for independent symptom mapping to target PCI to those patients most likely to benefit. (Systematic Trial of Angina Assessment Before Revascularization [ORBITA-STAR]; NCT04280575).
Asunto(s)
Angina de Pecho , Intervención Coronaria Percutánea , Humanos , Masculino , Persona de Mediana Edad , Femenino , Intervención Coronaria Percutánea/métodos , Angina de Pecho/terapia , Anciano , Resultado del TratamientoRESUMEN
BACKGROUND: Placebo-controlled evidence from ORBITA-2 (Objective Randomised Blinded Investigation with Optimal Medical Therapy of Angioplasty in Stable Angina-2) found that percutaneous coronary intervention (PCI) in stable coronary artery disease with little or no antianginal medication relieved angina, but residual symptoms persisted in many patients. The reason for this was unclear. OBJECTIVES: This ORBITA-2 secondary analysis investigates the relationship between presenting symptoms and disease severity (anatomic, noninvasive, and invasive ischemia) and the ability of symptoms to predict the placebo-controlled efficacy of PCI. METHODS: Prerandomization symptom severity and nature were assessed using the ORBITA smartphone application and symptom and quality of life questionnaires including the World Health Organization Rose angina questionnaire (Rose). Disease severity was assessed using quantitative coronary angiography, stress echocardiography, fractional flow reserve, and instantaneous wave-free ratio. Bayesian ordinal regression was used. RESULTS: At prerandomization, the median number of daily angina episodes was 0.8 (Q1-Q3: 0.4-1.6), 64% had Rose angina, quantitative coronary angiography diameter stenosis was 61% (Q1-Q3: 49%-74%), stress echocardiography score was 1.0 (Q1-Q3: 0.0-2.7), fractional flow reserve was 0.63 (Q1-Q3: 0.49-0.75), and instantaneous wave-free ratio was 0.78 (Q1-Q3: 0.55-0.87). There was little relationship between symptom severity and nature and disease severity: angina symptom score with quantitative coronary angiography ordinal correlation coefficient: 0.06 (95% credible interval [CrI]: 0.00-0.08); stress echocardiography: 0.09 (95% CrI: 0.02-0.10); fractional flow reserve: 0.04 (95% CrI: -0.03 to 0.07); and instantaneous wave-free ratio: 0.04 (95% CrI: -0.01 to 0.07). However, Rose angina and guideline-based typical angina were strong predictors of placebo-controlled PCI efficacy (angina symptom score: OR: 1.9; 95% CrI: 1.6-2.1; probability of interaction [PrInteraction] = 99.9%; and OR: 1.8; 95% CrI: 1.6-2.1; PrInteraction = 99.9%, respectively). CONCLUSIONS: Although symptom severity and nature were poorly associated with disease severity, the nature of symptoms powerfully predicted the placebo-controlled efficacy of PCI.
Asunto(s)
Enfermedad de la Arteria Coronaria , Intervención Coronaria Percutánea , Humanos , Masculino , Femenino , Intervención Coronaria Percutánea/métodos , Persona de Mediana Edad , Enfermedad de la Arteria Coronaria/terapia , Enfermedad de la Arteria Coronaria/diagnóstico , Anciano , Resultado del Tratamiento , Angiografía Coronaria , Índice de Severidad de la Enfermedad , Angina Estable/terapia , Angina Estable/diagnóstico , Angina Estable/fisiopatología , Calidad de VidaRESUMEN
BACKGROUND: Long-term clinical outcomes of catheter ablation (CA) compared to thoracoscopic surgical ablation (SA) to treat patients with long-standing persistent atrial fibrillation (LSPAF) are not known. OBJECTIVE: The purpose of this study was to compare the long-term (36-month) clinical efficacy, quality of life, and cost-effectiveness of SA and CA in LSPAF. METHODS: Participants were followed up for 3 years using implantable loop recorders and questionnaires to assess the change in quality of life. Intention-to-treat analyses were used to report the findings. RESULTS: Of the 115 patients with LSPAF treated, 104 (90.4%) completed 36-month follow-up [CA: n = 57 (95%); SA: n = 47 (85%)]. After a single procedure without antiarrhythmic drugs, 7 patients (12%) in the CA arm and 5 (11%) in the SA arm [hazard ratio 1.22; 95% confidence interval (CI) 0.81-1.83; P = .41] were free from atrial fibrillation/tachycardia (AF/AT) ≥30 seconds at 36 months. Thirty-three patients (58%) in the CA arm and 26 (55%) in the SA arm (hazard ratio 1.04; 95% CI 0.57-1.88; P = .91) had their AF/AT burden reduced by ≥75%. The overall impact on health-related quality of life was similar, with mean quality-adjusted life year estimates of 2.45 (95% CI 2.31-2.59) for CA and 2.32 (95% CI 2.13-2.52) for SA. Estimated costs were higher for SA (mean £24,682; 95% CI £21,746-£27,618) than for CA (mean £18,002; 95% CI £15,422-£20,581). CONCLUSION: In symptomatic LSPAF, CA and SA were equally effective at achieving arrhythmia outcomes (freedom from AF/AT ≥30 seconds and ≥75% burden reduction) after a single procedure without antiarrhythmic drugs. However, SA is significantly more costly than CA. TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT04280042.
RESUMEN
BACKGROUND: Ventricular tachycardia (VT) reduces cardiac output through high heart rates, loss of atrioventricular synchrony, and loss of ventricular synchrony. We studied the contribution of each mechanism and explored the potential therapeutic utility of His bundle pacing to improve cardiac output during VT. METHODS: Study 1 aimed to improve the understanding of mechanisms of harm during VT (using pacing simulated VT). In 23 patients with left ventricular impairment, we recorded continuous ECG and beat-by-beat blood pressure measurements. We assessed the hemodynamic impact of heart rate and restoration of atrial and biventricular synchrony. Study 2 investigated novel pacing interventions during clinical VT by evaluating the hemodynamic effects of His bundle pacing at 5 bpm above the VT rate in 10 patients. RESULTS: In Study 1, at progressively higher rates of simulated VT, systolic blood pressure declined: at rates of 125, 160, and 190 bpm, -22.2%, -42.0%, and -58.7%, respectively (ANOVA p < 0.0001). Restoring atrial synchrony alone had only a modest beneficial effect on systolic blood pressure (+ 3.6% at 160 bpm, p = 0.2117), restoring biventricular synchrony alone had a greater effect (+ 9.1% at 160 bpm, p = 0.242), and simultaneously restoring both significantly increased systolic blood pressure (+ 31.6% at 160 bpm, p = 0.0003). In Study 2, the mean rate of clinical VT was 143 ± 21 bpm. His bundle pacing increased systolic blood pressure by + 14.2% (p = 0.0023). In 6 of 10 patients, VT terminated with His bundle pacing. CONCLUSIONS: Restoring atrial and biventricular synchrony improved hemodynamic function in simulated and clinical VT. Conduction system pacing could improve VT tolerability and treatment.
RESUMEN
BACKGROUND: The coronary sinus reducer (CSR) is proposed to reduce angina in patients with stable coronary artery disease by improving myocardial perfusion. We aimed to measure its efficacy, compared with placebo, on myocardial ischaemia reduction and symptom improvement. METHODS: ORBITA-COSMIC was a double-blind, randomised, placebo-controlled trial conducted at six UK hospitals. Patients aged 18 years or older with angina, stable coronary artery disease, ischaemia, and no further options for treatment were eligible. All patients completed a quantitative adenosine-stress perfusion cardiac magnetic resonance scan, symptom and quality-of-life questionnaires, and a treadmill exercise test before entering a 2-week symptom assessment phase, in which patients reported their angina symptoms using a smartphone application (ORBITA-app). Patients were randomly assigned (1:1) to receive either CSR or placebo. Both participants and investigators were masked to study assignment. After the CSR implantation or placebo procedure, patients entered a 6-month blinded follow-up phase in which they reported their daily symptoms in the ORBITA-app. At 6 months, all assessments were repeated. The primary outcome was myocardial blood flow in segments designated ischaemic at enrolment during the adenosine-stress perfusion cardiac magnetic resonance scan. The primary symptom outcome was the number of daily angina episodes. Analysis was done by intention-to-treat and followed Bayesian methodology. The study is registered with ClinicalTrials.gov, NCT04892537, and completed. FINDINGS: Between May 26, 2021, and June 28, 2023, 61 patients were enrolled, of whom 51 (44 [86%] male; seven [14%] female) were randomly assigned to either the CSR groupâ(n=25) or the placebo group (n=26). Of these, 50 patients were included in the intention-to-treat analysis (24 in the CSR group and 26 in the placebo group). 454 (57%) of 800 imaged cardiac segments were ischaemic at enrolment, with a median stress myocardial blood flow of 1·08 mL/min per g (IQR 0·77-1·41). Myocardial blood flow in ischaemic segments did not improve with CSR compared with placebo (difference 0·06 mL/min per g [95% CrI -0·09 to 0·20]; Pr(Benefit)=78·8%). The number of daily angina episodes was reduced with CSR compared with placebo (OR 1·40 [95% CrI 1·08 to 1·83]; Pr(Benefit)=99·4%). There were two CSR embolisation events in the CSR group, and no acute coronary syndrome events or deaths in either group. INTERPRETATION: ORBITA-COSMIC found no evidence that the CSR improved transmural myocardial perfusion, but the CSR did improve angina compared with placebo. These findings provide evidence for the use of CSR as a further antianginal option for patients with stable coronary artery disease. FUNDING: Medical Research Council, Imperial College Healthcare Charity, National Institute for Health and Care Research Imperial Biomedical Research Centre, St Mary's Coronary Flow Trust, British Heart Foundation.
Asunto(s)
Angina Estable , Enfermedad de la Arteria Coronaria , Seno Coronario , Intervención Coronaria Percutánea , Humanos , Masculino , Femenino , Enfermedad de la Arteria Coronaria/terapia , Angina Estable/tratamiento farmacológico , Seno Coronario/diagnóstico por imagen , Teorema de Bayes , Resultado del Tratamiento , Intervención Coronaria Percutánea/efectos adversos , Método Doble Ciego , Isquemia , AdenosinaRESUMEN
Importance: Unlike medications, procedural interventions are rarely trialed against placebo prior to becoming accepted in clinical practice. When placebo-controlled trials are eventually conducted, procedural interventions may be less effective than previously believed. Objective: To investigate the importance of including a placebo arm in trials of surgical and interventional procedures by comparing effect sizes from trials of the same procedure that do and do not include a placebo arm. Data Sources: Searches of MEDLINE and Embase identified all placebo-controlled trials for procedural interventions in any specialty of medicine and surgery from inception to March 31, 2019. A secondary search identified randomized clinical trials assessing the same intervention, condition, and end point but without a placebo arm for paired comparison. Study Selection: Placebo-controlled trials of anatomically site-specific procedures requiring skin incision or endoscopic techniques were eligible for inclusion; these were then matched to trials without placebo control that fell within prespecified limits of heterogeneity. Data Extraction and Synthesis: Random-effects meta-regression, with placebo and blinding as a fixed effect and intervention and end point grouping as random effects, was used to calculate the impact of placebo control for each end point. Data were analyzed from March 2019 to March 2020. Main Outcomes and Measures: End points were examined in prespecified subgroups: patient-reported or health care professional-assessed outcomes, quality of life, pain, blood pressure, exercise-related outcomes, recurrent bleeding, and all-cause mortality. Results: Ninety-seven end points were matched from 72 blinded, placebo-controlled trials (hereafter, blinded) and 55 unblinded trials without placebo control (hereafter, unblinded), including 111â¯500 individual patient end points. Unblinded trials had larger standardized effect sizes than blinded trials for exercise-related outcomes (standardized mean difference [SMD], 0.59; 95% CI, 0.29 to 0.89; P < .001) and quality-of-life (SMD, 0.32; 95% CI, 0.11 to 0.53; P = .003) and health care professional-assessed end points (SMD, 0.40; 95% CI, 0.18 to 0.61; P < .001). The placebo effect accounted for 88.1%, 55.2%, and 61.3% of the observed unblinded effect size for these end points, respectively. There was no significant difference between unblinded and blinded trials for patient-reported end points (SMD, 0.31; 95% CI, -0.02 to 0.64; P = .07), blood pressure (SMD, 0.26; 95% CI, -0.10 to 0.62; P = .15), all-cause mortality (odds ratio [OR], 0.23; 95% CI, -0.26 to 0.72; P = .36), pain (SMD, 0.03; 95% CI, -0.52 to 0.57; P = .91), or recurrent bleeding events (OR, -0.12; 95% CI, -1.11 to 0.88; P = .88). Conclusions and Relevance: The magnitude of the placebo effect found in this systematic review and meta-regression was dependent on the end point. Placebo control in trials of procedural interventions had the greatest impact on exercise-related, quality-of-life, and health care professional-assessed end points. Randomized clinical trials of procedural interventions may consider placebo control accordingly.
Asunto(s)
Ensayos Clínicos Controlados Aleatorios como Asunto , Humanos , PlacebosRESUMEN
BACKGROUND AND OBJECTIVE: Training deep learning models for medical image segmentation require large annotated datasets, which can be expensive and time-consuming to create. Active learning is a promising approach to reduce this burden by strategically selecting the most informative samples for segmentation. This study investigates the use of active learning for efficient left ventricle segmentation in echocardiography with sparse expert annotations. METHODS: We adapt and evaluate various sampling techniques, demonstrating their effectiveness in judiciously selecting samples for segmentation. Additionally, we introduce a novel strategy, Optimised Representativeness Sampling, which combines feature-based outliers with the most representative samples to enhance annotation efficiency. RESULTS: Our findings demonstrate a substantial reduction in annotation costs, achieving a remarkable 99% upper bound performance while utilising only 20% of the labelled data. This equates to a reduction of 1680 images needing annotation within our dataset. When applied to a publicly available dataset, our approach yielded a remarkable 70% reduction in required annotation efforts, representing a significant advancement compared to baseline active learning strategies, which achieved only a 50% reduction. Our experiments highlight the nuanced performance of diverse sampling strategies across datasets within the same domain. CONCLUSIONS: The study provides a cost-effective approach to tackle the challenges of limited expert annotations in echocardiography. By introducing a distinct dataset, made publicly available for research purposes, our work contributes to the field's understanding of efficient annotation strategies in medical image segmentation.
Asunto(s)
Ecocardiografía , Ventrículos Cardíacos , Ventrículos Cardíacos/diagnóstico por imagen , Procesamiento de Imagen Asistido por ComputadorRESUMEN
BACKGROUND: Late gadolinium enhancement (LGE) of the myocardium has significant diagnostic and prognostic implications, with even small areas of enhancement being important. Distinguishing between definitely normal and definitely abnormal LGE images is usually straightforward, but diagnostic uncertainty arises when reporters are not sure whether the observed LGE is genuine or not. This uncertainty might be resolved by repetition (to remove artifact) or further acquisition of intersecting images, but this must take place before the scan finishes. Real-time quality assurance by humans is a complex task requiring training and experience, so being able to identify which images have an intermediate likelihood of LGE while the scan is ongoing, without the presence of an expert is of high value. This decision-support could prompt immediate image optimization or acquisition of supplementary images to confirm or refute the presence of genuine LGE. This could reduce ambiguity in reports. METHODS: Short-axis, phase-sensitive inversion recovery late gadolinium images were extracted from our clinical cardiac magnetic resonance (CMR) database and shuffled. Two, independent, blinded experts scored each individual slice for "LGE likelihood" on a visual analog scale, from 0 (absolute certainty of no LGE) to 100 (absolute certainty of LGE), with 50 representing clinical equipoise. The scored images were split into two classes-either "high certainty" of whether LGE was present or not, or "low certainty." The dataset was split into training, validation, and test sets (70:15:15). A deep learning binary classifier based on the EfficientNetV2 convolutional neural network architecture was trained to distinguish between these categories. Classifier performance on the test set was evaluated by calculating the accuracy, precision, recall, F1-score, and area under the receiver operating characteristics curve (ROC AUC). Performance was also evaluated on an external test set of images from a different center. RESULTS: One thousand six hundred and forty-five images (from 272 patients) were labeled and split at the patient level into training (1151 images), validation (247 images), and test (247 images) sets for the deep learning binary classifier. Of these, 1208 images were "high certainty" (255 for LGE, 953 for no LGE), and 437 were "low certainty". An external test comprising 247 images from 41 patients from another center was also employed. After 100 epochs, the performance on the internal test set was accuracy = 0.94, recall = 0.80, precision = 0.97, F1-score = 0.87, and ROC AUC = 0.94. The classifier also performed robustly on the external test set (accuracy = 0.91, recall = 0.73, precision = 0.93, F1-score = 0.82, and ROC AUC = 0.91). These results were benchmarked against a reference inter-expert accuracy of 0.86. CONCLUSION: Deep learning shows potential to automate quality control of late gadolinium imaging in CMR. The ability to identify short-axis images with intermediate LGE likelihood in real-time may serve as a useful decision-support tool. This approach has the potential to guide immediate further imaging while the patient is still in the scanner, thereby reducing the frequency of recalls and inconclusive reports due to diagnostic indecision.
Asunto(s)
Medios de Contraste , Aprendizaje Profundo , Interpretación de Imagen Asistida por Computador , Valor Predictivo de las Pruebas , Humanos , Medios de Contraste/administración & dosificación , Reproducibilidad de los Resultados , Interpretación de Imagen Asistida por Computador/normas , Bases de Datos Factuales , Miocardio/patología , Masculino , Femenino , Imagen por Resonancia Cinemagnética/normas , Persona de Mediana Edad , Cardiopatías/diagnóstico por imagen , Garantía de la Calidad de Atención de Salud/normas , Variaciones Dependientes del Observador , Anciano , Imagen por Resonancia Magnética/normasRESUMEN
BACKGROUND: Artificial intelligence (AI) could improve accuracy and reproducibility of echocardiographic measurements in dogs. HYPOTHESIS: A neural network can be trained to measure echocardiographic left ventricular (LV) linear dimensions in dogs. ANIMALS: Training dataset: 1398 frames from 461 canine echocardiograms from a single specialist center. VALIDATION: 50 additional echocardiograms from the same center. METHODS: Training dataset: a right parasternal 4-chamber long axis frame from each study, labeled by 1 of 18 echocardiographers, marking anterior and posterior points of the septum and free wall. VALIDATION DATASET: End-diastolic and end-systolic frames from 50 studies, annotated twice (blindly) by 13 experts, producing 26 measurements of each site from each frame. The neural network also made these measurements. We quantified its accuracy as the deviation from the expert consensus, using the individual-expert deviation from consensus as context for acceptable variation. The deviation of the AI measurement away from the expert consensus was assessed on each individual frame and compared with the root-mean-square-variation of the individual expert opinions away from that consensus. RESULTS: For the septum in end-diastole, individual expert opinions deviated by 0.12 cm from the consensus, while the AI deviated by 0.11 cm (P = .61). For LVD, the corresponding values were 0.20 cm for experts and 0.13 cm for AI (P = .65); for the free wall, experts 0.20 cm, AI 0.13 cm (P < .01). In end-systole, there were no differences between individual expert and AI performances. CONCLUSIONS AND CLINICAL IMPORTANCE: An artificial intelligence network can be trained to adequately measure linear LV dimensions, with performance indistinguishable from that of experts.
Asunto(s)
Inteligencia Artificial , Ecocardiografía , Perros , Animales , Reproducibilidad de los Resultados , Ecocardiografía/veterinaria , Ecocardiografía/métodos , Ventrículos Cardíacos/diagnóstico por imagen , DiástoleRESUMEN
Doppler echocardiography is a widely utilised non-invasive imaging modality for assessing the functionality of heart valves, including the mitral valve. Manual assessments of Doppler traces by clinicians introduce variability, prompting the need for automated solutions. This study introduces an innovative deep learning model for automated detection of peak velocity measurements from mitral inflow Doppler images, independent from Electrocardiogram information. A dataset of Doppler images annotated by multiple expert cardiologists was established, serving as a robust benchmark. The model leverages heatmap regression networks, achieving 96% detection accuracy. The model discrepancy with the expert consensus falls comfortably within the range of inter- and intra-observer variability in measuring Doppler peak velocities. The dataset and models are open-source, fostering further research and clinical application.
Asunto(s)
Aprendizaje Profundo , Velocidad del Flujo Sanguíneo , Ecocardiografía Doppler/métodos , Válvula Mitral/diagnóstico por imagen , Ultrasonografía DopplerRESUMEN
Background: The prognostic impact of ventricular tachycardia (VT) catheter ablation is an important outstanding research question. We undertook a reconstructed individual patient data meta-analysis of randomised controlled trials comparing ablation to medical therapy in patients developing VT after MI. Methods: We systematically identified all trials comparing catheter ablation to medical therapy in patients with VT and prior MI. The prespecified primary endpoint was reconstructed individual patient assessment of all-cause mortality. Prespecified secondary endpoints included trial-level assessment of all-cause mortality, VT recurrence or defibrillator shocks and all-cause hospitalisations. Prespecified subgroup analysis was performed for ablation approaches involving only substrate modification without VT activation mapping. Sensitivity analyses were performed depending on the proportion of patients with prior MI included. Results: Eight trials, recruiting a total of 874 patients, were included. Of these 874 patients, 430 were randomised to catheter ablation and 444 were randomised to medical therapy. Catheter ablation reduced all-cause mortality compared with medical therapy when synthesising individual patient data (HR 0.63; 95% CI [0.41-0.96]; p=0.03), but not in trial-level analysis (RR 0.91; 95% CI [0.67-1.23]; p=0.53; I2=0%). Catheter ablation significantly reduced VT recurrence, defibrillator shocks and hospitalisations compared with medical therapy. Sensitivity analyses were consistent with the primary analyses. Conclusion: In patients with postinfarct VT, catheter ablation reduces mortality.
RESUMEN
The ability of statins to reduce the morbidity and mortality of cardiovascular disease has ensured that they are among the most prescribed drugs in modern medicine. Unfortunately, most patients who start taking statins will end up stopping them, most commonly due to side effects. Confusingly, however, in blinded placebo-controlled trials, side effects appear no more common in those taking statins than those taking placebo. One possible explanation is that ever-present background symptoms are being falsely attributed to statins. However, another explanation is the nocebo effect, where the act of just taking a tablet (even a placebo) causes genuine side effects in patients. Two recent randomized placebo-controlled personalized (N-of-1) trials have been reported: StatinWise and SAMSON. In these trials, each participant was randomized to multiple periods of statin and placebo, with regular symptom burden assessments. Together, these trials support the existence of a significant nocebo effect from taking statins. Possibly even more importantly, they demonstrate the ability of personalized trials to inform and empower patients: up to half of the patients in these trials were able to successfully restart statins after taking part, despite previously having been statin intolerant. StatinWise and SAMSON have raised public awareness of the nocebo effect in statin intolerance. However, they also demonstrate a potential role for the personalized design outside of clinical trials. If taking part in these personalized experiments allows half of our patients to successfully restart life-saving medications, maybe we should be able to prescribe personalized experiments to our patients in the clinical setting?