EasyPISA: Automatic Integrated PISA Measurements of Mitral Regurgitation From 2-D Color-Doppler Using Deep Learning.

OBJECTIVE: The proximal isovelocity surface area (PISA) method is a well-established approach for mitral regurgitation (MR) quantification. However, it exhibits high inter-observer variability and inaccuracies in cases of non-hemispherical flow convergence and non-holosystolic MR. To address this, we present EasyPISA, a framework for automated integrated PISA measurements taken directly from 2-D color-Doppler sequences. METHODS: We trained convolutional neural networks (UNet/Attention UNet) on 1171 images from 196 recordings (54 patients) to detect and segment flow convergence zones in 2-D color-Doppler images. Different preprocessing schemes and model architectures were compared. Flow convergence surface areas were estimated, accounting for non-hemispherical convergence, and regurgitant volume (RVol) was computed by integrating the flow rate over time. EasyPISA was retrospectively applied to 26 MR patient examinations, comparing results with reference PISA RVol measurements, severity grades, and cMRI RVol measurements for 13 patients. RESULTS: The UNet trained on duplex images achieved the best results (precision: 0.63, recall: 0.95, dice: 0.58, flow rate error: 10.4 ml/s). Mitigation of false-positive segmentation on the atrial side of the mitral valve was achieved through integration with a mitral valve segmentation network. The intraclass correlation coefficient was 0.83 between EasyPISA and PISA, and 0.66 between EasyPISA and cMRI. Relative standard deviations were 46% and 53%, respectively. Receiver operator characteristics demonstrated a mean area under the curve between 0.90 and 0.97 for EasyPISA RVol estimates and reference severity grades. CONCLUSION: EasyPISA demonstrates promising results for fully automated integrated PISA measurements in MR, offering potential benefits in workload reduction and mitigating inter-observer variability in MR assessment.

Five-year outcomes of mitral valve repair for leaflet prolapse at a medium-sized Norwegian university hospital.

Objective. To evaluate patient characteristics and 5-year outcomes after surgical mitral valve (MV) repair for leaflet prolapse at a medium-sized cardiothoracic center. Background. Contemporary reports on the outcome of MV repair at medium-sized cardiothoracic centers are sparse. Methods. Patients receiving open-heart surgery with MV repair due to primary mitral regurgitation caused by leaflet prolapse between 2015 and 2021, without active endocarditis, were included. Clinical data, complications, re-interventions, mortality, and echocardiographic data were retrospectively registered from electronical patient charts, both pre-operatively and from post-operative follow-ups. Results. One hundred and three patients were included, 83% male, with a mean age of 62 years. All-cause mortality was 9% during a median follow-up time of 4.9 years. Re-intervention rate on the MV was 4%. Post-operative complications before last available follow-up visit at median 3.0 years were infrequent, with new-onset atrial fibrillation/flutter in 16%, post-operative MV regurgitation grade II or above in 17% and post-operative tricuspid regurgitation grade II or above in 14%. Conclusions. These data demonstrate that surgical MV repair for leaflet prolapse at a medium-sized cardiothoracic center was associated with low re-intervention rate and few severe complications. The presented results are comparable to data from surgical high-volume centers, indicating that surgical MV repair can be safely performed at selected medium-sized cardiothoracic centers.

Systemic embolization in infective endocarditis.

Embolism is a common complication in infective endocarditis which may lead to serious complications, such as stroke, intestinal ischemia, and peripheral embolization. A comprehensive literature search was performed and the registry at our centre, including 390 cases of infective endocarditis, diagnosed between 2010 and 2020, was investigated. Large registries show that 20-40% of patients with infective endocarditis (IE) are affected by embolism. In many instances, embolism is present already at the time of diagnosis. The rate of embolism during the hospital stay in our data was 11%. However, only 2% developed clinical embolism during or following surgery. According to recent guidelines, previous embolism, and the presence of vegetations > 10 mm present an indication for surgical treatment. Routine imaging revealed non-symptomatic cerebral embolism in 8.5% of surgical patients. However, it is not clear whether detection of non-symptomatic embolism and consecutive surgical treatment improves the prognosis of infective endocarditis.

Mitral Valve Segmentation and Tracking from Transthoracic Echocardiography Using Deep Learning.

OBJECTIVE: Valvular heart diseases (VHDs) pose a significant public health burden, and deciding the best treatment strategy necessitates accurate assessment of heart valve function. Transthoracic echocardiography (TTE) is the key modality to evaluate VHDs, but the lack of standardized quantitative measurements leads to subjective and time-consuming assessments. We aimed to use deep learning to automate the extraction of mitral valve (MV) leaflets and annular hinge points from echocardiograms of the MV, improving standardization and reducing workload in quantitative assessment of MV disease. METHODS: We annotated the MV leaflets and annulus points in 2931 images from 127 patients. We propose an approach for segmenting the annotated features using Attention UNet with deep supervision and weight scheduling of the attention coefficients to enforce saliency surrounding the MV. The derived segmentation masks were used to extract quantitative biomarkers for specific MV leaflet scallops throughout the heart cycle. RESULTS: Evaluation performance was summarized using a Dice score of 0.63 ± 0.14, annulus error of 3.64 ± 2.53 and leaflet angle error of 8.7 ± 8.3°. Leveraging Attention UNet with deep supervision robustness of clinically relevant metrics was improved compared with UNet, reducing standard deviations by 2.7° (angle error) and 0.73 mm (annulus error). We correctly identified cases of MV prolapse, cases of stenosis and healthy references from a clinical material using the derived biomarkers. CONCLUSION: Robust deep learning segmentation and tracking of MV morphology and motion is possible by leveraging attention gates and deep supervision, and holds promise for enhancing VHD diagnosis and treatment monitoring.