Objective To explore the value of deep learning(DL)models semi-automatic training system for automatic optimization of clinical image quality control of transthoracic echocardiography(TTE).Methods Totally 1250 TTE vid...Objective To explore the value of deep learning(DL)models semi-automatic training system for automatic optimization of clinical image quality control of transthoracic echocardiography(TTE).Methods Totally 1250 TTE videos from 402 patients were retrospectively collected,including 490 apical four chamber(A4C),310 parasternal long axis view of left ventricle(PLAX)and 450 parasternal short axis view of great vessel(PSAX GV).The videos were divided into development set(245 A4C,155 PLAX,225 PSAX GV),semi-automated training set(98 A4C,62 PLAX,90 PSAX GV)and test set(147 A4C,93 PLAX,135 PSAX GV)at the ratio of 5∶2∶3.Based on development set and semi-automatic training set,DL model of quality control was semi-automatically iteratively optimized,and a semi-automatic training system was constructed,then the efficacy of DL models for recognizing TTE views and assessing imaging quality of TTE were verified in test set.Results After optimization,the overall accuracy,precision,recall,and F1 score of DL models for recognizing TTE views in test set improved from 97.33%,97.26%,97.26%and 97.26%to 99.73%,99.65%,99.77%and 99.71%,respectively,while the overall accuracy for assessing A4C,PLAX and PSAX GV TTE as standard views in test set improved from 89.12%,83.87%and 90.37%to 93.20%,90.32%and 93.33%,respectively.Conclusion The developed DL models semi-automatic training system could improve the efficiency of clinical imaging quality control of TTE and increase iteration speed.展开更多
Background To report quality control methods and baseline reproducibility data of the ultrasound measurements of carotid artery intima-media thickness in the project of Establishment of an Integrated System for Corona...Background To report quality control methods and baseline reproducibility data of the ultrasound measurements of carotid artery intima-media thickness in the project of Establishment of an Integrated System for Coronary Heart Disease Prevention and Treatment. Methods Standard ultrasound scanning and measuring protocols were established by the study group. All sonographers and readers were trained by the carotid ultrasound core lab and all digital ultrasound images were centrally read. Ten subjects were scanned twice （with 1 week interval） by 2 sonographers independently and images were read by a single reader to evaluate the sonographer variability. Twenty subjects＇ images were read twice （with 1 week interval） by a single reader to assess the reader variability and the reproducibility of IMT measured at different carotid segments. Results The intraclass correlation （ICC） of intra- and inter-sonographer and intra- reader for mean IMT measurements was 0.99, 0.98 and 0.97 respectively; while for max IMT, it was 0.97, 0.99 and 0.95 respectively. Among different carotid segments and sites, ICC for mean 1MT measurements of common carotid （CCA）, carotid artery bulb （Bulb）, internal carotid artery （ICA）, overall near wall and overall far wall was 0.97, 0.99, 0.89, 0.93 and 0.98 respectively. Conclusion The reproducibility of IMT measurements according to our protocol is acceptable, although better reproducibility is found when measuring the mean IMT than max IMT, CCA and Bulb IMT than ICA IMT, and far wall IMT than near wall IMT.展开更多
文摘Objective To explore the value of deep learning(DL)models semi-automatic training system for automatic optimization of clinical image quality control of transthoracic echocardiography(TTE).Methods Totally 1250 TTE videos from 402 patients were retrospectively collected,including 490 apical four chamber(A4C),310 parasternal long axis view of left ventricle(PLAX)and 450 parasternal short axis view of great vessel(PSAX GV).The videos were divided into development set(245 A4C,155 PLAX,225 PSAX GV),semi-automated training set(98 A4C,62 PLAX,90 PSAX GV)and test set(147 A4C,93 PLAX,135 PSAX GV)at the ratio of 5∶2∶3.Based on development set and semi-automatic training set,DL model of quality control was semi-automatically iteratively optimized,and a semi-automatic training system was constructed,then the efficacy of DL models for recognizing TTE views and assessing imaging quality of TTE were verified in test set.Results After optimization,the overall accuracy,precision,recall,and F1 score of DL models for recognizing TTE views in test set improved from 97.33%,97.26%,97.26%and 97.26%to 99.73%,99.65%,99.77%and 99.71%,respectively,while the overall accuracy for assessing A4C,PLAX and PSAX GV TTE as standard views in test set improved from 89.12%,83.87%and 90.37%to 93.20%,90.32%and 93.33%,respectively.Conclusion The developed DL models semi-automatic training system could improve the efficiency of clinical imaging quality control of TTE and increase iteration speed.
文摘Background To report quality control methods and baseline reproducibility data of the ultrasound measurements of carotid artery intima-media thickness in the project of Establishment of an Integrated System for Coronary Heart Disease Prevention and Treatment. Methods Standard ultrasound scanning and measuring protocols were established by the study group. All sonographers and readers were trained by the carotid ultrasound core lab and all digital ultrasound images were centrally read. Ten subjects were scanned twice （with 1 week interval） by 2 sonographers independently and images were read by a single reader to evaluate the sonographer variability. Twenty subjects＇ images were read twice （with 1 week interval） by a single reader to assess the reader variability and the reproducibility of IMT measured at different carotid segments. Results The intraclass correlation （ICC） of intra- and inter-sonographer and intra- reader for mean IMT measurements was 0.99, 0.98 and 0.97 respectively; while for max IMT, it was 0.97, 0.99 and 0.95 respectively. Among different carotid segments and sites, ICC for mean 1MT measurements of common carotid （CCA）, carotid artery bulb （Bulb）, internal carotid artery （ICA）, overall near wall and overall far wall was 0.97, 0.99, 0.89, 0.93 and 0.98 respectively. Conclusion The reproducibility of IMT measurements according to our protocol is acceptable, although better reproducibility is found when measuring the mean IMT than max IMT, CCA and Bulb IMT than ICA IMT, and far wall IMT than near wall IMT.
