GAN Jianhong,LI Wei.Automatic Calculation Method of Left Ventricular Ejection Fraction in M-mode Echocardiography[J].Journal of Chengdu University of Information Technology,2021,36(06):624-628.[doi:10.16836/j.cnki.jcuit.2021.06.007]
M型超声心动图中左室射血分数自动计算方法
- Title:
- Automatic Calculation Method of Left Ventricular Ejection Fraction in M-mode Echocardiography
- 文章编号:
- 2096-1618(2021)06-0624-05
- 关键词:
- 左心室射血分数; Teichholz法; M型超声心动图; 图像处理; 自动测量
- Keywords:
- left ventricular ejection fraction; Teichholz method; M-mode echocardiography; image processing; automatic measurement
- 分类号:
- TP311
- 文献标志码:
- A
- 摘要:
- 掌上超声仪受功率和超声频率的限制很难获取心脏B超图像,导致目前基于二维B超的左室射血分数(left ventricular ejection fraction, LVEF)测量不能在掌上超声仪上实现,而掌上超声仪能够通过M型超声获取心脏识别信息。对此,基于M型超声心动图Teichholz法提出了基于数字图像处理的LVEF自动计算方法。首先,将二尖瓣的影响考虑在内,同时为了之后的图像分割得到好的结果,将图像分为上下部分进行处理; 其次,利用图像二值化的三角法对图像进行分割,对于上部分子图像进行二尖瓣信息的消除; 最后,图像上下部分分别得到左心室室隔内膜和左心室后壁内膜的波形线,取得波峰波谷的位置数据,进行LVEF的计算。该方法所得LVEF的计算准确率达95.6%,计算效率为每张图像0.6秒。基于M型超声心动图的LVEF的自动测量方法,较手动方法测量效率更高,测量精度也在合理范围之内,对于掌上式超声诊断仪的半自动化甚至自动化有着极大的推进作用,具有较好的临床应用价值。
- Abstract:
- Due to the limitations of power and ultrasound frequency, it is difficult for handheld ultrasound systems to obtain cardiac B-ultrasound images. As a result, the current Left Ventricular Ejection Fraction(LVEF)measurement which is based on two-dimensional B-ultrasound cannot be achieved on handheld ultrasound systems. The ultrasound instrument can obtain heart identification information through M-mode ultrasound. For this, an automatic calculation method of LVEF based on digital image processing is proposed based on the Teichholz method of M-mode echocardiography. First of all, to take the influence of the mitral valve into account, and at the same time to obtain good results for the subsequent image segmentation, the image needs to be divided into upper and lower parts for processing; Secondly, the image is segmented by the triangulation method of image binarization, and the mitral valve information is eliminated for the upper part of the image; Finally, the waveform lines of the left ventricular septum and the left ventricular posterior wall are obtained in the upper and lower parts of the image, respectively, and the position data of the peaks and troughs are obtained, which are substituted into the formula for the calculation of LVEF. The calculation accuracy of LVEF obtained by this method is 95.6%, and the calculation efficiency is 0.6 seconds per image. The automatic measurement method of LVEF based on M-mode echocardiography is more efficient than manual methods, and the measurement accuracy is also within a reasonable range. It has a great promotion effect on the semi-automation or even automation of handheld ultrasound diagnostic equipment, and it also has good clinical application value.
参考文献/References:
[1] 胡孟芬,丁钧,华坚,等.负荷超声心动图对缺血性舒张性心力衰竭疗效的评估价值[J].中国医刊,2015,50(7):87-90.
[2] 王珍.强心合剂治疗慢性心力衰竭的临床研究[D].南京:南京中医药大学,2013.
[3] 彭代秋.为老年慢性充血性心力衰竭患者应用芪苈强心胶囊进行辅助治疗的效果分析[J].当代医药论丛,2015,13(20):25-26.
[4] BuckertD,KelleS,BussS,et al.Left ventricular ejection fraction and presence of myocardial necrosis assessed by cardiac magnetic resonance imaging correctly risk stratify patients with stable coronary artery diseas:a multi-center all-comerstrial[J].ClinicalResearchinCardiology,2017,106(3):219-229.
[5] 程育博,邢继岩.超声心动图Teichholz校正公式与左心室造影测量左室射血分数的对比分析[J].Zeitschrift fur Bibliothekswesen und Bibliographie,2010,8(9):1147-1148.
[6] 蒋建慧,姚静,张艳娟,等.基于深度学习的超声自动测量左室射血分数的研究[J].临床超声医学杂志,2019,21(1):70-74.
[7] BelaidA,BoukerrouiD,Maingourd Y,et al.Phase-Based Level Set Segmentation of Ultrasound Images[J].IEEE Transactionson Information Technology in Biomedicine,2011,15(1):138-147.
[8] Agarwal D,Shriram K S,Subramanian N.Automatic view classification of echocardiograms using histogram of oriented gradients[C].2013 IEEE 10th International Symposium on Biomedical Imaging.IEEE,2013:1368-1371.
[9] Marsousi M,Eftekhari A,Kocharian A,et al.Endocardial boundaryextraction in left ventricular echocardiographic images using fast adaptive B-spline snake algorithm[J].International Journal of Computer Assisted Radiology and Surgery,2010,5(5):501-513.
[10] Khamis H,Zurakhov G,Azar V,et al.Automatic apical view classification of echocardiograms using a discriminative learning dictionary[J].Medical Image Analysis,2017,36:15-21.
[11] Pedrosa J,Heyde B,Heeren L,et al.Automatic short axis orientation of the left ventricle in 3D ultrasound recordings[C].Medical Imaging 2016:Ultrasonic Imaging and Tomography.International Society for Optics and Photonics,2016:9790.
[12] Khan A,Iskandar D NF A,Ujir H,et al.Automaticsegmentation of CMRIs for LV contour detection[C].9thInternational Conferenceon Robotic,Vision,SignalProcessing and Power Applications.Springer,Singapore,2017:313-319.
[13] 徐礼胜,张书琪,牛潇,等.基于全卷积网络的左心室射血分数自动检测[J].东北大学学报:自然科学版,2018,39(11):1572-1576.
[14] 刘晓鸣,雷震,何刊,等.全卷积神经网络与全连接条件随机场中的左心室射血分数精准计算[J].计算机辅助设计与图形学学报,2019,31(3):431-438.
[15] 齐欣.超声心动图测量LVEF的方法及结果判读[J].中华心脏与心律电子杂志,2017,5(2):77-80.
备注/Memo
收稿日期:2021-04-06
基金项目:四川省科技厅应用基础研究资助项目(2019YJ0361); 四川省科技厅重点研发资助项目(2021YFG0173)