|
|
Method for Measuring Fetal Head Circumference in Ultrasound Images Based on Mask R-CNN |
Li Zonggui, Zhang Junhua*, Mei Liye |
(School of Information, Yunnan University, Kunming 650500, China) |
|
|
Abstract Fetal head circumference is one of the most important biological characteristics in prenatal ultrasound evaluation of fetal growth and development. However, manual measurement is time-consuming and labor-consuming and may have errors by the operator. According to the feature of fetal head close to ellipse shape in ultrasound image, the head circumference measurement loss function was proposed in this paper. After the segmentation branch of Mask R-CNN, ElliFit algorithm was used to fit the ellipse of the segmentation mask. Ramanujan formula was used to calculate the fitting ellipse circumference as the measurement value of the head circumference. The mean square error of the real value of the head circumference and the measurement value was added into the original loss function as head circumference measurement loss function to allow the training process of the model to be closely related to the measurement task. By this way the measurement accuracy and speed was improved. One hundred and ninety ultrasound images of fetal head were tested. Dice’s coefficient was 96.89%±1.01%, and the measurement error was (0.33±1.54) mm. The average processing time of one ultrasound image was 0.33 s. Compared with the traditional manual measurement method or the current machine learning methods, the proposed method improved the speed between 1.13 seconds and 16.87 seconds, and improved the accuracy between 0.21 mm and 1.68 mm. The results showed that the improved Mask R-CNN increased the efficiency of doctors in measuring fetal head circumference, which met the clinical needs.
|
Received: 10 December 2019
|
|
Corresponding Authors:
*E-mail: jhzhang@ynu.edu.cn
|
|
|
|
[1] Brandt I, Sticker EJ, Lentze MJ. Catch-up growth of head circumference of very low birth weight, small for gestational age preterm infants and mental development to adulthood[J]. The Journal of Pediatrics, 2003, 142(5): 463-470. [2] Peterson J, Taylor HG, Minich N, et al. Subnormal head circumference in very low birth weight children: neonatal correlates and school-age consequences[J]. Early Human Development, 2006, 82(5): 325-334. [3] Corson S. Ultrasound in Obstetrics and Gynecology[J]. The Journal of Reproductive Medicine,1978,20(2):1-22. [4] Lu Wei, Tan Jinglu, Floyd R. Automated fetal head detection and measurement in ultrasound images by iterative randomized Hough transform[J]. Ultrasound in Medicine & Miology, 2005, 31(7): 929-936. [5] Carneiro G, Georgescu B, Good S, et al. Automatic fetal measurements in ultrasound using constrained probabilistic boosting tree[J]. Med Image Comput Comput Assist Interv, 2008, 10(2):571-579. [6] Stebbing RV, McManigle JE. A boundary fragment model for head segmentation in fetal ultrasound[C]// Proceedings of Challenge US: Biometric Measurements from Fetal Ultrasound Images, ISBI. Barcelona: IEEE, 2012: 9-11. [7] Ciurte A, Bresson X, Cuisenaire O, et al. Semi-supervised segmentation of ultrasound images based on patch representation and continuous min cut[J]. PLoS ONE, 2014, 9(7): e100972. [8] Foi A, Maggioni M, Pepe A, et al. Difference of Gaussians revolved along elliptical paths for ultrasound fetal head segmentation[J]. Computerized Medical Imaging and Graphics, 2014, 38(8): 774-784. [9] van den Heuvel TLA, de Bruijn D, de Korte CL, et al. Automated measurement of fetal head circumference using 2D ultrasound images[J]. PLoS ONE, 2018, 13(8): e0200412. [10] 李璟,倪东,李胜利,等. 超声图像中胎儿头围的自动测量 [J].深圳大学学报理工版, 2014, 31(5):455-463. [11] Li Jing, Wang Yi, Lei Baiying, et al. Automatic fetal head circumference measurement in ultrasound using random forest and fast ellipse fitting[J]. IEEE Journal of Biomedical and Health Informatics, 2017, 22(1): 215-223. [12] Ni Dong, Yang Yong, Li Shengli, et al. Learning based automatic head detection and measurement from fetal ultrasound images via prior knowledge and imaging parameters[C]//2013 IEEE 10th International Symposium on Biomedical Imaging. San Francisco: IEEE, 2013: 772-775. [13] Satwika IP, Habibie I, Ma'sum MA, et al. Particle swarm optimation based 2-dimensional randomized hough transform for fetal head biometry detection and approximation in ultrasound imaging[C]// 2014 International Conference on Advanced Computer Science and Information System, Jakarta: IEEE, 2014: 468-473. [14] Prasad DK, Leung MKH, Quek C. ElliFit: An unconstrained, non-iterative, least squares based geometric Ellipse Fitting method[J]. Pattern Recognition, 2013, 46(5): 1449-1465. [15] Wu Lingyun, Xin Yang, Li Shengli, et al. Cascaded fully convolutional networks for automatic prenatal ultrasound image segmentation[C]//2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). Melbourne: VIC, 2017: 663-666. [16] Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 39(4):640-651. [17] Tu Zhuowen, Bai Xiang. Auto-context and its application to high-level vision tasks and 3d brain image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 32(10): 1744-1757. [18] He K, Gkioxari G, Dollár P, et al. Mask R-CNN [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2):386-397. [19] Barnard RW, Pearce K, Schovanec L. Inequalities for the perimeter of an ellipse [J]. Journal of Mathematical Analysis and Applications, 2001, 260(2): 295-306. [20] Rueda S, Fathima S, Knight CL, et al. Evaluation and comparison of current fetal ultrasound image segmentation methods for biometric measurements: A grand challenge [J]. IEEE Transactions on Medical Imaging, 2013, 33(4): 797-813. [21] Sinclair M, Baumgartner CF, Matthew J, et al. Human-level performance on automatic head biometrics in fetal ultrasound using fully convolutional neural networks[C]//2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Honolulu: IEEE, 2018:714-717. |
[1] |
Li Xinzheng, Jin Wei, Li Gang, Yin Caoqian. YOLO V2 Network with Asymmetric ConvolutionKernel for Lung Nodule Detection of CT Image[J]. Chinese Journal of Biomedical Engineering, 2019, 38(4): 401-408. |
[2] |
Lin Zehui, Lei Baiying, Jiang Feng, Ni Dong, Chen Siping, Li Shengli, Wang Tianfu. Quality Assessment of Fetal Head Ultrasound Images Based on Faster R-CNN[J]. Chinese Journal of Biomedical Engineering, 2019, 38(4): 392-400. |
|
|
|
|