A Deep Learning Localization Method for Measuring Abdominal Muscle Dimensions in Ultrasound Images

Health professionals extensively use Two-Dimensional (2D) Ultrasound (US) videos and images to visualize and measure internal organs for various purposes including evaluation of muscle architectural changes. US images can be used to measure abdominal muscles dimensions for the diagnosis and creation of customized treatment plans for patients with Low Back Pain (LBP), however, they are difficult to interpret. Due to high variability, skilled professionals with specialized training are required to take measurements to avoid low intra-observer reliability. This variability stems from the challenging nature of accurately finding the correct spatial location of measurement endpoints in abdominal US images. In this paper, we use a Deep Learning (DL) approach to automate the measurement of the abdominal muscle thickness in 2D US images. By treating the problem as a localization task, we develop a modified Fully Convolutional Network (FCN) architecture to generate blobs of coordinate locations of measurement endpoints, similar to what a human operator does. We demonstrate that using the TrA400 US image dataset, our network achieves a Mean Absolute Error (MAE) of 0.3125 on the test set, which almost matches the performance of skilled ultrasound technicians. Our approach can facilitate next steps for automating the process of measurements in 2D US images, while reducing inter-observer as well as intra-observer variability for more effective clinical outcomes.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  C. Fede,et al.  Inter‐rater reliability and variability of ultrasound measurements of abdominal muscles and fasciae thickness , 2019, Clinical anatomy.

[3]  Jonathan Tompson,et al.  Efficient object localization using Convolutional Networks , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  Libertario Demi,et al.  Deep learning for automated detection of B-lines in lung ultrasonography , 2018, The Journal of the Acoustical Society of America.

[5]  Fei Xu,et al.  Automatic Breast Ultrasound Image Segmentation: A Survey , 2017, Pattern Recognit..

[6]  Mahmood Fathy,et al.  STFCN: Spatio-Temporal FCN for Semantic Video Segmentation , 2016, ArXiv.

[7]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[8]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Issam H. Laradji,et al.  Looc: Localize Overlapping Objects with Count Supervision , 2020, 2020 IEEE International Conference on Image Processing (ICIP).

[10]  Christopher Joseph Pal,et al.  The Importance of Skip Connections in Biomedical Image Segmentation , 2016, LABELS/DLMIA@MICCAI.

[11]  G. V. Vera,et al.  Assessment of genome damage in occupational exposure to ionising radiation and ultrasound. , 1997, Mutation research.

[12]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Mark W. Schmidt,et al.  Where are the Masks: Instance Segmentation with Image-level Supervision , 2019, BMVC.

[14]  Hai Wang,et al.  A Semantic Segmentation Algorithm Using FCN with Combination of BSLIC , 2018 .

[15]  M. Boermeester,et al.  A comparison of the Accuracy of Ultrasound and Computed Tomography in common diagnoses causing acute abdominal pain , 2011, European Radiology.

[16]  Issam H. Laradji,et al.  A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis , 2020, Scientific Reports.

[17]  Elena Celledoni,et al.  Equivariant neural networks for inverse problems , 2021, Inverse problems.

[18]  Susan J. Gordon,et al.  Measurement of transversus abdominis activation in chronic low back pain patients using a novel standardized real-time ultrasound imaging method , 2018, Ultrasound.

[19]  Il Dong Yun,et al.  Joint Weakly and Semi-Supervised Deep Learning for Localization and Classification of Masses in Breast Ultrasound Images , 2017, IEEE Transactions on Medical Imaging.

[20]  Jianwen Luo,et al.  Deep Learning for Ultrasound Localization Microscopy , 2020, IEEE Transactions on Medical Imaging.

[21]  Mark W. Schmidt,et al.  Instance Segmentation with Point Supervision , 2019, ArXiv.

[22]  V. Srikanth,et al.  Measuring ultrasound images of abdominal and lumbar multifidus muscles in older adults: A reliability study. , 2016, Manual therapy.

[23]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[24]  Line Eikvil,et al.  Automated Left Ventricle Dimension Measurement in 2D Cardiac Ultrasound via an Anatomically Meaningful CNN Approach , 2019, SUSI/PIPPI@MICCAI.

[25]  Olivier Seynnes,et al.  Fully automated analysis of muscle architecture from B-mode ultrasound images with deep learning , 2020, ArXiv.

[26]  Jason Yosinski,et al.  An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution , 2018, NeurIPS.

[27]  Qiang Chen,et al.  Network In Network , 2013, ICLR.

[28]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[29]  Junying Chen,et al.  Medical Ultrasound Imaging: To GPU or Not to GPU? , 2011, IEEE Micro.

[30]  J. Fritz,et al.  Rehabilitative ultrasound imaging is a valid measure of trunk muscle size and activation during most isometric sub-maximal contractions: a systematic review. , 2009, The Australian journal of physiotherapy.

[31]  Tareq Abed Mohammed,et al.  Understanding of a convolutional neural network , 2017, 2017 International Conference on Engineering and Technology (ICET).

[32]  P. Moran,et al.  Prenatal screening and diagnosis of neural tube defects , 2009, Prenatal diagnosis.

[33]  HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[34]  Keiron O'Shea,et al.  An Introduction to Convolutional Neural Networks , 2015, ArXiv.

[35]  Fausto Milletari,et al.  Fully Convolutional Regression Network for Accurate Detection of Measurement Points , 2017, DLMIA/ML-CDS@MICCAI.

[36]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[37]  Richard K. G. Do,et al.  Convolutional neural networks: an overview and application in radiology , 2018, Insights into Imaging.

[38]  Ian Loram,et al.  Real-Time Ultrasound Segmentation, Analysis and Visualisation of Deep Cervical Muscle Structure , 2017, IEEE Transactions on Medical Imaging.

[39]  Horst Bischof,et al.  Regressing Heatmaps for Multiple Landmark Localization Using CNNs , 2016, MICCAI.

[40]  Ruey-Feng Chang,et al.  Tumor Detection in Automated Breast Ultrasound Using 3-D CNN and Prioritized Candidate Aggregation , 2019, IEEE Transactions on Medical Imaging.

[41]  Amin Ali,et al.  LULC Segmentation of RGB Satellite Image Using FCN-8 , 2020, ArXiv.

[42]  William Parker,et al.  A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images , 2020, 2021 IEEE Winter Conference on Applications of Computer Vision (WACV).

[43]  I. Timor-Tritsch,et al.  The use of transvaginal ultrasonography in the diagnosis of ectopic pregnancy. , 1989, American journal of obstetrics and gynecology.

[44]  Chulhee Lee,et al.  Revisiting spatial dropout for regularizing convolutional neural networks , 2020, Multimedia Tools and Applications.

[45]  Mark W. Schmidt,et al.  Where are the Blobs: Counting by Localization with Point Supervision , 2018, ECCV.

[46]  Konstantinos Kamnitsas,et al.  SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound , 2016, IEEE Transactions on Medical Imaging.

[47]  K. Doma,et al.  Intra-rater reliability of transversus abdominis measurement by a novice examiner: Comparison of “freehand” to “probe force device” method of real-time ultrasound imaging , 2019, Ultrasound.