Artificial intelligence-enhanced automation of left ventricular diastolic assessment: a pilot study for feasibility, diagnostic validation, and outcome prediction.

Background: Evaluating left ventricular diastolic function (LVDF) is crucial in echocardiography; however, the complexity and time demands of current guidelines challenge clinical use. This study aimed to develop an artificial intelligence (AI)-based framework for automatic LVDF assessment to reduce subjectivity and improve accuracy and outcome prediction. Methods: We developed an AI-based LVDF assessment framework using a nationwide echocardiographic dataset from five tertiary hospitals. This framework automatically identifies views, calculates diastolic parameters, including mitral inflow and annular velocities (E/A ratio, e' velocity, and E/e' ratio), maximal tricuspid regurgitation velocity, left atrial (LA) volume index, and left atrial reservoir strain (LARS). Subsequently, it grades LVDF according to guidelines. The AI-framework was validated on an external dataset composed of randomly screened 173 outpatients who underwent transthoracic echocardiography with suspicion for diastolic dysfunction and 33 individuals from medical check-ups with normal echocardiograms at Seoul National University Bundang Hospital, tertiary medical center in Korea, between May 2012 and June 2022. Additionally, we assessed the predictive value of AI-derived diastolic parameters and LVDF grades for a clinical endpoint, defined as a composite of all-cause death and hospitalization for heart failure, using Cox-regression risk modelling. Results: In an evaluation with 200 echocardiographic examinations (167 suspected diastolic dysfunction patients, 33 controls), it achieves an overall accuracy of 99.1% in identifying necessary views. Strong correlations (Pearson coefficient 0.901-0.959) were observed between AI-derived and manually-derived measurements of diastolic parameters, including LARS as well as conventional parameters. When following the guidelines, whether utilizing AI-derived or manually-derived parameters, the evaluation of LVDF consistently showed high concordance rates (94%). However, both methods exhibited lower concordance rates with the clinician's prior assessments (77.5% and 78.5%, respectively). Importantly, both AI-derived and manually-derived LVDF grades independently demonstrated significant prognostic value [adjusted hazard ratio (HR) =3.03; P=0.03 and adjusted HR =2.75; P=0.04, respectively] for predicting clinical outcome. In contrast, the clinician's prior grading lost its significance as a prognostic indicator after adjusting for clinical risk factors (adjusted HR =1.63; P=0.36). AI-derived LARS values significantly decreased with worsening LVDF (P for trend <0.001), and low LARS (<17%) was associated with increased risk for the clinical outcome (Log-rank P=0.04) relative to that for preserved LARS (≥17%). Conclusions: Our AI-based approach for automatic LVDF assessment on echocardiography is feasible, potentially enhancing clinical diagnosis and outcome prediction.

Artificial intelligence-enhanced automation for M-mode echocardiographic analysis: ensuring fully automated, reliable, and reproducible measurements.

To enhance M-mode echocardiography's utility for measuring cardiac structures, we developed and evaluated an artificial intelligence (AI)-based automated analysis system for M-mode images through the aorta and left atrium [M-mode (Ao-LA)], and through the left ventricle [M-mode (LV)]. Our system, integrating two deep neural networks (DNN) for view classification and image segmentation, alongside an auto-measurement algorithm, was developed using 5,958 M-mode images [3,258 M-mode (LA-Ao), and 2,700 M-mode (LV)] drawn from a nationwide echocardiographic dataset collated from five tertiary hospitals. The performance of view classification and segmentation DNNs were evaluated on 594 M-mode images, while automatic measurement accuracy was tested on separate internal test set with 100 M-mode images as well as external test set with 280 images (140 sinus rhythm and 140 atrial fibrillation). Performance evaluation showed the view classification DNN's overall accuracy of 99.8% and segmentation DNN's Dice similarity coefficient of 94.3%. Within the internal test set, all automated measurements, including LA, Ao, and LV wall and cavity, resonated strongly with expert evaluations, exhibiting Pearson's correlation coefficients (PCCs) of 0.81-0.99. This performance persisted in the external test set for both sinus rhythm (PCC, 0.84-0.98) and atrial fibrillation (PCC, 0.70-0.97). Notably, automatic measurements, consistently offering multi-cardiac cycle readings, showcased a stronger correlation with the averaged multi-cycle manual measurements than with those of a single representative cycle. Our AI-based system for automatic M-mode echocardiographic analysis demonstrated excellent accuracy, reproducibility, and speed. This automated approach has the potential to improve efficiency and reduce variability in clinical practice.

Generative adversarial network with radiomic feature reproducibility analysis for computed tomography denoising.

BACKGROUND: Most computed tomography (CT) denoising algorithms have been evaluated using image quality analysis (IQA) methods developed for natural image, which do not adequately capture the texture details in medical imaging. Radiomics is an emerging image analysis technique that extracts texture information to provide a more objective basis for medical imaging diagnostics, overcoming the subjective nature of traditional methods. By utilizing the difficulty of reproducing radiomics features under different imaging protocols, we can more accurately evaluate the performance of CT denoising algorithms. METHOD: We introduced radiomic feature reproducibility analysis as an evaluation metric for a denoising algorithm. Also, we proposed a low-dose CT denoising method based on a generative adversarial network (GAN), which outperformed well-known CT denoising methods. RESULTS: Although the proposed model produced excellent results visually, the traditional image assessment metrics such as peak signal-to-noise ratio and structural similarity failed to show distinctive performance differences between the proposed method and the conventional ones. However, radiomic feature reproducibility analysis provided a distinctive assessment of the CT denoising performance. Furthermore, radiomic feature reproducibility analysis allowed fine-tuning of the hyper-parameters of the GAN. CONCLUSION: We demonstrated that the well-tuned GAN architecture outperforms the well-known CT denoising methods. Our study is the first to introduce radiomics reproducibility analysis as an evaluation metric for CT denoising. We look forward that the study may bridge the gap between traditional objective and subjective evaluations in the clinical medical imaging field.

Fully automated quantification of cardiac chamber and function assessment in 2-D echocardiography: clinical feasibility of deep learning-based algorithms.

We aimed to compare the segmentation performance of the current prominent deep learning (DL) algorithms with ground-truth segmentations and to validate the reproducibility of the manually created 2D echocardiographic four cardiac chamber ground-truth annotation. Recently emerged DL based fully-automated chamber segmentation and function assessment methods have shown great potential for future application in aiding image acquisition, quantification, and suggestion for diagnosis. However, the performance of current DL algorithms have not previously been compared with each other. In addition, the reproducibility of ground-truth annotations which are the basis of these algorithms have not yet been fully validated. We retrospectively enrolled 500 consecutive patients who underwent transthoracic echocardiogram (TTE) from December 2019 to December 2020. Simple U-net, Res-U-net, and Dense-U-net algorithms were compared for the segmentation performances and clinical indices such as left atrial volume (LAV), left ventricular end diastolic volume (LVEDV), left ventricular end systolic volume (LVESV), LV mass, and ejection fraction (EF) were evaluated. The inter- and intra-observer variability analysis was performed by two expert sonographers for a randomly selected echocardiographic view in 100 patients (apical 2-chamber, apical 4-chamber, and parasternal short axis views). The overall performance of all DL methods was excellent [average dice similarity coefficient (DSC) 0.91 to 0.95 and average Intersection over union (IOU) 0.83 to 0.90], with the exception of LV wall area on PSAX view (average DSC of 0.83, IOU 0.72). In addition, there were no significant difference in clinical indices between ground truth and automated DL measurements. For inter- and intra-observer variability analysis, the overall intra observer reproducibility was excellent: LAV (ICC = 0.995), LVEDV (ICC = 0.996), LVESV (ICC = 0.997), LV mass (ICC = 0.991) and EF (ICC = 0.984). The inter-observer reproducibility was slightly lower as compared to intraobserver agreement: LAV (ICC = 0.976), LVEDV (ICC = 0.982), LVESV (ICC = 0.970), LV mass (ICC = 0.971), and EF (ICC = 0.899). The three current prominent DL-based fully automated methods are able to reliably perform four-chamber segmentation and quantification of clinical indices. Furthermore, we were able to validate the four cardiac chamber ground-truth annotation and demonstrate an overall excellent reproducibility, but still with some degree of inter-observer variability.

Reconnection of fragmented parts of coronary arteries using local geometric features in X-ray angiography images.

The segmentation of coronary arteries in X-ray images is essential for image-based guiding procedures and the diagnosis of cardiovascular disease. However, owing to the complex and thin structures of the coronary arteries, it is challenging to accurately segment arteries in X-ray images using only a single neural network model. Consequently, coronary artery images obtained by segmentation with a single model are often fragmented, with parts of the arteries missing. Sophisticated post-processing is then required to identify and reconnect the fragmented regions. In this paper, we propose a method to reconstruct the missing regions of coronary arteries using X-ray angiography images. METHOD: We apply an independent convolutional neural network model considering local details, as well as a local geometric prior, for reconnecting the disconnected fragments. We implemented and compared the proposed method with several convolutional neural networks with customized encoding backbones as baseline models. RESULTS: When integrated with our method, existing models improved considerably in terms of similarity with ground truth, with a mean increase of 0.330 of the Dice similarity coefficient in local regions of disconnected arteries. The method is efficient and is able to recover missing fragments in a short number of iterations. CONCLUSION AND SIGNIFICANCE: Owing to the restoration of missing fragments of coronary arteries, the proposed method enables a significant enhancement of clinical impact. The method is general and can simply be integrated into other existing methods for coronary artery segmentation.