Joint Capsule Segmentation in Ultrasound Images of the Metacarpophalangeal Joint using Convolutional Neural Networks*

This work addresses the automatic segmentation of the joint capsule in ultrasound images of the metacarpophalangeal joint using an adapted version of the well known UNet model. These images are used in the diagnosis of rheumatic diseases, one of the main causes of impairment and pain in developed countries. The identification of the joint capsule gives important clues about the presence or Rheumatoid Arthritis. This structure can be used to extract metrics to help quantify the disease stage and progression. The solution proposed here has the potential to reduce the burden on the radiologists as well as the subjectivity of the diagnosis by providing quantitative measurements, such as the synovitis area. The proposed approach was compared with two other works present in the literature. Results show that our solution outperforms the two reference methods with 90% of the joint capsules identified with a DICE higher than 0.67.

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

[2]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[3]  Enrico Grisan,et al.  Improved detection of synovial boundaries in ultrasound examination by using a cascade of active-contours. , 2013, Medical engineering & physics.

[4]  Edgar Brunner,et al.  A novel ultrasonographic synovitis scoring system suitable for analyzing finger joint inflammation in rheumatoid arthritis. , 2005, Arthritis and rheumatism.

[5]  Karolina Nurzynska,et al.  Segmentation of finger joint synovitis in ultrasound images , 2016, 2016 IEEE Sixth International Conference on Communications and Electronics (ICCE).

[6]  A. Silman,et al.  Rheumatoid arthritis classifi cation criteria : an American College of Rheumatology / European League Against Rheumatism collaborative initiative , 2010 .

[7]  Karolina Nurzynska,et al.  Automatic finger joint synovitis localization in ultrasound images , 2016, Photonics Europe.

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

[9]  C. Schueller-weidekamm Quantification of synovial and erosive changes in rheumatoid arthritis with ultrasound--revisited. , 2009, European journal of radiology.

[10]  Ultrasonography in rheumatology: developing its potential in clinical practice and research. , 2007, Rheumatology.

[11]  A. Silman,et al.  UvA-DARE (Digital Academic Repository) 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative Aletaha, , 2010 .

[12]  Yoni Donner,et al.  Fully Automating Graf's Method for DDH Diagnosis Using Deep Convolutional Neural Networks , 2016, LABELS/DLMIA@MICCAI.

[13]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[14]  Miguel Tavares Coimbra,et al.  Joint capsule segmentation in ultrasound images of the metacarpophalangeal joint using a split and merge approach , 2018, 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI).