High-accuracy detection of supraspinatus fatty infiltration in shoulder MRI using convolutional neural network algorithms

被引:5
|
作者
Saavedra, Juan Pablo [1 ]
Droppelmann, Guillermo [2 ,3 ,4 ]
Garcia, Nicolas [2 ]
Jorquera, Carlos [5 ]
Feijoo, Felipe [1 ]
机构
[1] Pontificia Univ Catolica Valparaiso, Sch Ind Engn, Valparaiso, Chile
[2] MEDS Clin, Res Ctr Med Exercise Sport & Hlth, Santiago, Chile
[3] Univ Catol Murcia UCAM, Hlth Sci PhD Program, Murcia, Spain
[4] Harvard TH Chan Sch Publ Hlth, Principles & Practice Clin Res PPCR, Boston, MA 02115 USA
[5] Univ Mayor, Escuela Nutr & Dietet, Fac Ciencias, Santiago, Chile
关键词
classification; deep learning; fatty infiltration; MRI; supraspinatus; ROTATOR CUFF TEARS; DEGENERATION; MUSCLES; ATROPHY;
D O I
10.3389/fmed.2023.1070499
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
BackgroundThe supraspinatus muscle fatty infiltration (SMFI) is a crucial MRI shoulder finding to determine the patient's prognosis. Clinicians have used the Goutallier classification to diagnose it. Deep learning algorithms have been demonstrated to have higher accuracy than traditional methods. AimTo train convolutional neural network models to categorize the SMFI as a binary diagnosis based on Goutallier's classification using shoulder MRIs. MethodsA retrospective study was performed. MRI and medical records from patients with SMFI diagnosis from January 1st, 2019, to September 20th, 2020, were selected. 900 T2-weighted, Y-view shoulder MRIs were evaluated. The supraspinatus fossa was automatically cropped using segmentation masks. A balancing technique was implemented. Five binary classification classes were developed into two as follows, A: 0, 1 v/s 3, 4; B: 0, 1 v/s 2, 3, 4; C: 0, 1 v/s 2; D: 0, 1, 2, v/s 3, 4; E: 2 v/s 3, 4. The VGG-19, ResNet-50, and Inception-v3 architectures were trained as backbone classifiers. An average of three 10-fold cross-validation processes were developed to evaluate model performance. AU-ROC, sensitivity, and specificity with 95% confidence intervals were used. ResultsOverall, 606 shoulders MRIs were analyzed. The Goutallier distribution was presented as follows: 0 = 403; 1 = 114; 2 = 51; 3 = 24; 4 = 14. Case A, VGG-19 model demonstrated an AU-ROC of 0.991 +/- 0.003 (accuracy, 0.973 +/- 0.006; sensitivity, 0.947 +/- 0.039; specificity, 0.975 +/- 0.006). B, VGG-19, 0.961 +/- 0.013 (0.925 +/- 0.010; 0.847 +/- 0.041; 0.939 +/- 0.011). C, VGG-19, 0.935 +/- 0.022 (0.900 +/- 0.015; 0.750 +/- 0.078; 0.914 +/- 0.014). D, VGG-19, 0.977 +/- 0.007 (0.942 +/- 0.012; 0.925 +/- 0.056; 0.942 +/- 0.013). E, VGG-19, 0.861 +/- 0.050 (0.779 +/- 0.054; 0.706 +/- 0.088; 0.831 +/- 0.061). ConclusionConvolutional neural network models demonstrated high accuracy in MRIs SMFI diagnosis.
引用
收藏
页数:12
相关论文
共 50 条
  • [41] Transparent Object Detection Using Convolutional Neural Network
    Khaing, May Phyo
    Masayuki, Mukunoki
    BIG DATA ANALYSIS AND DEEP LEARNING APPLICATIONS, 2019, 744 : 86 - 93
  • [42] Early Wildfire Detection Using Convolutional Neural Network
    Oh, Seon Ho
    Ghyme, Sang Won
    Jung, Soon Ki
    Kim, Geon-Woo
    FRONTIERS OF COMPUTER VISION, 2020, 1212 : 18 - 30
  • [43] Brain Cancer Tumor Classification from Motion-Corrected MRI Images Using Convolutional Neural Network
    Mengash, Hanan Abdullah
    Mahmoud, Hanan A. Hosni
    CMC-COMPUTERS MATERIALS & CONTINUA, 2021, 68 (02): : 1551 - 1563
  • [44] Automatic Detection and Classification of Rib Fractures on Thoracic CT Using Convolutional Neural Network: Accuracy and Feasibility
    Zhou, Qing-Qing
    Wang, Jiashuo
    Tang, Wen
    Hu, Zhang-Chun
    Xia, Zi-Yi
    Li, Xue-Song
    Zhang, Rongguo
    Yin, Xindao
    Zhang, Bing
    Zhang, Hong
    KOREAN JOURNAL OF RADIOLOGY, 2020, 21 (07) : 869 - 879
  • [45] Detection and Classification of Brain Tumors From MRI Images Using a Deep Convolutional Neural Network Approach
    Menaouer, Brahami
    El-Houda, Kebir Nour
    Zoulikha, Dermane
    Mohammed, Sabri
    Matta, Nada
    INTERNATIONAL JOURNAL OF SOFTWARE INNOVATION, 2022, 10 (01)
  • [46] A hyperspectral anomaly detection framework based on segmentation and convolutional neural network algorithms
    Hosseiny, Benyamin
    Shah-Hosseini, Reza
    INTERNATIONAL JOURNAL OF REMOTE SENSING, 2020, 41 (18) : 6946 - 6975
  • [47] Arabic Text Classification Using Convolutional Neural Network and Genetic Algorithms
    Alsaleh, Deem
    Larabi-Marie-Sainte, Souad
    IEEE ACCESS, 2021, 9 (09): : 91670 - 91685
  • [48] Nodule Detection with Convolutional Neural Network Using Apache Spark and GPU Frameworks
    Venkatesan, Nikitha Johnsirani
    Shin, Dong Ryeol
    Nam, Choon Sung
    APPLIED SCIENCES-BASEL, 2021, 11 (06):
  • [49] Sound-Based Cough Detection System using Convolutional Neural Network
    Tomaszewska, Julia Zofia
    Chousidis, Christos
    Donati, Eugenio
    2022 IEEE INTERNATIONAL SYMPOSIUM ON MEDICAL MEASUREMENTS AND APPLICATIONS (MEMEA 2022), 2022,
  • [50] A novel approach for detection of dyslexia using convolutional neural network with EOG signals
    Ileri, Ramis
    Latifoglu, Fatma
    Demirci, Esra
    MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING, 2022, 60 (11) : 3041 - 3055