Purpose To investigate the feasibility of a deep learning framework to automatically analyze echocardiographic color Doppler videos in detecting valvular regurgitation.Materials and Methods This study retrospectively collected echocardiographic images of 1 109 patients with valvular regurgitation in the Fourth Medical Center of PLA General Hospital,from June 2015 to September 2019 as the training and validation sets.A prospective continuous collection of 1 562 echocardiography images was used as the test set in the Fourth Medical Center of PLA General Hospital from May 13 to June 13,2023,including 378 cases of mitral regurgitation and 223 cases of aortic regurgitation.This study developed deep learning networks to establish view classification model and valvular regurgitation recognition model,including the efficiency of section classification of deep learning models.Results The deep learning view classification model in this study could automatically identify two views for diagnosing mitral regurgitation and aortic regurgitation.The recognition accuracy for the parasternal long axis color Doppler view and the apical four chamber mitral color Doppler view was 1.00 and 0.93,respectively.The sensitivity,specificity,accuracy and area under the curve of the deep learning model for diagnosing mitral regurgitation were 0.847,0.852,0.849 and 0.930,respectively.The sensitivity,specificity,accuracy and area under the curve of the deep learning model in diagnosing aortic regurgitation were 0.857,0.861,0.859 and 0.940,respectively.Conclusion Deep learning algorithms can automatically identify valvular regurgitation and have the potential to become a screening tool for valvular heart disease.

Purpose To investigate the feasibility of a deep learning framework to automatically analyze echocardiographic color Doppler videos in detecting valvular regurgitation.Materials and Methods This study retrospectively collected echocardiographic images of 1 109 patients with valvular regurgitation in the Fourth Medical Center of PLA General Hospital,from June 2015 to September 2019 as the training and validation sets.A prospective continuous collection of 1 562 echocardiography images was used as the test set in the Fourth Medical Center of PLA General Hospital from May 13 to June 13,2023,including 378 cases of mitral regurgitation and 223 cases of aortic regurgitation.This study developed deep learning networks to establish view classification model and valvular regurgitation recognition model,including the efficiency of section classification of deep learning models.Results The deep learning view classification model in this study could automatically identify two views for diagnosing mitral regurgitation and aortic regurgitation.The recognition accuracy for the parasternal long axis color Doppler view and the apical four chamber mitral color Doppler view was 1.00 and 0.93,respectively.The sensitivity,specificity,accuracy and area under the curve of the deep learning model for diagnosing mitral regurgitation were 0.847,0.852,0.849 and 0.930,respectively.The sensitivity,specificity,accuracy and area under the curve of the deep learning model in diagnosing aortic regurgitation were 0.857,0.861,0.859 and 0.940,respectively.Conclusion Deep learning algorithms can automatically identify valvular regurgitation and have the potential to become a screening tool for valvular heart disease.

Objective To investigate the progress of two-dimensional echocardiographic parameters in patients with different severity of aortic stenosis.Methods A retrospective analysis was per-formed on 96 patients diagnosed with aortic stenosis with at least 2 times of transthoracic echo-cardiography(interval ≥1 year)in Department of Cardiology,Fourth Medical Center of Chinese PLA General Hospital from March 2017 to December 2023.According to aortic stenosis severity,they were divided into a mild group(72 cases),a moderate group(14 cases)and a severe group(10 cases).Peak pressure gradient(PPG)across aortic valve,Vmax,mean aortic valve pressure gradient(ΔPm),pulmonary artery systolic pressure(PASP)were collected,and the changes and annual progress of these echocardiographic parameters at baseline and before and after follow-up were analyzed.Results The values of IVST,LVPWT,Vmax,aortic valve PPG and ΔPm were sig-nificantly increased in the mild,moderate and severe stenosis groups in turn(P＜0.05,P＜0.01).The values of Vmax,PPG and ΔPm were significantly lower in the mild stenosis group than the moderate and severe stenosis groups,and the LVPWT value was obviously lower in the mild ste-nosis group than the severe stenosis group(P＜0.05).The aortic valve PPG and ΔPm values at follow-up were significantly higher than those before the follow-up in the three stenosis groups(P＜0.05,P＜0.01).After follow-up,the Vmax values in mild and moderate stenosis groups were notably higher than before(P＜0.01).The PASP value at follow-up was significantly higher than before in the severe stenosis group(P＜0.05).The annual progression rate of Vmax,PASP,LVEF were gradually increased in the mild,moderate,and severe stenosis groups(P＞0.05).The annual progression rate of ΔPm was gradually increased in the three groups in turn(2.30±1.77 mm Hg/year vs 2.40±1.18 mm Hg/year vs 6.08±1.70 mm Hg/year,P＜0.05).Conclusion As the severity of baseline aortic stenosis increases,obvious changes are observed in cardiac structure and function.Before and after follow-up,the serious the aortic stenosis severity is,the faster the annual progression rates of Vmax,PPG,LVEF and PASP are.