Curr Health Sci J, vol. 47, no. 2, 2021

Assessment of Deep Learning Methods for Differentiating Autoimmune Disorders in Ultrasound Images

[Original Paper]

Corina Maria Vasile(1,2), Anca Loredana Udritoiu(3), Alice Elena Ghenea(4), Vlad Padureanu(5), Stefan Udritoiu(3), Lucian Gheorghe Gruionu(6), Gabriel Gruionu(7), Andreea Valentina Iacob(3), Mihaela Popescu(8)


(1)PhD School Department, University of Medicine and Pharmacy of Craiova, Romania,
(2)Department of Pediatric Cardiology, County Clinical Emergency Hospital of Craiova, Romania,
(3)Faculty of Automation, Computers and Electronics, University of Craiova, Romania,
(4)Department of Bacteriology-Virology-Parasitology, University of Medicine and Pharmacy of Craiova, Romania,
(5)Department of Internal Medicine, University of Medicine and Pharmacy of Craiova, Romania,
(6)Faculty of Mechanics, University of Craiova, Romania,
(7)Department of Medicine, Indiana University School of Medicine, Indianapolis, United States,
(8)Department of Endocrinology, University of Medicine and Pharmacy of Craiova, Romania


Abstract:

At present, deep learning becomes an important tool in medical image analysis, with good performance in diagnosing, pattern detection, and segmentation. Ultrasound imaging offers an easy and rapid method to detect and diagnose thyroid disorders. With the help of a computer-aided diagnosis (CAD) system based on deep learning, we have the possibility of real-time and non-invasive diagnosing of thyroidal US images. This paper proposed a study based on deep learning with transfer learning for differentiating the thyroidal ultrasound images using image pixels and diagnosis labels as inputs. We trained, assessed, and compared two pre-trained models (VGG-19 and Inception v3) using a dataset of ultrasound images consisting of 2 types of thyroid ultrasound images: autoimmune and normal. The training dataset consisted of 615 thyroid ultrasound images, from which 415 images were diagnosed as autoimmune, and 200 images as normal. The models were assessed using a dataset of 120 images, from which 80 images were diagnosed as autoimmune, and 40 images diagnosed as normal. The two deep learning models obtained very good results, as follows: the pre-trained VGG-19 model obtained 98.60% for the overall test accuracy with an overall specificity of 98.94% and overall sensitivity of 97.97%, while the Inception v3 model obtained 96.4% for the overall test accuracy with an overall specificity of 95.58% and overall sensitivity of 95.58%


Keywords:
Ultrasound imaging, autoimmune disorders; deep learning, convolutional neural networks, transfer learning



Corresponding:
Alice Elena Ghenea, Department of Bacteriology-Virology-Parasitology, University of Medicine and Pharmacy of Craiova, Romania, e-mail: gaman_alice@yahoo.com; Vlad Padureanu, Department of Internal Medicine, University of Medicine and Pharmacy of Craiova, Romania, e-mail: vladpadureanu@yahoo.com


DOI 10.12865/CHSJ.47.02.12 - Download PDF