Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network

By Jianning Chi, Ekta Walia, Paul S. Babyn, J. Wang, G. Groot, M. Eramian. Journal of digital imaging, 2017

368 citations recorded by the source corpus.

With many thyroid nodules being incidentally detected, it is important to identify as many malignant nodules as possible while excluding those that are highly likely to be benign from fine needle aspiration (FNA) biopsies or surgeries. This paper presents a computer-aided diagnosis (CAD) system for classifying thyroid nodules in ultrasound images. We use deep learning approach to extract features from thyroid ultrasound images. Ultrasound images are pre-processed to calibrate their scale and remove the artifacts. A pre-trained GoogLeNet model is then fine-tuned using the pre-processed image samples which leads to superior feature extraction. The extracted features of the thyroid ultrasound images are sent to a Cost-sensitive Random Forest classifier to classify the images into “malignant” and “benign” cases. The experimental results show the proposed fine-tuned GoogLeNet model achieves excellent classification performance, attaining 98.29% classification accuracy, 99.10% sensitivity and 93.90% specificity for the images in an open access database (Pedraza et al. 16), while 96.34% classification accuracy, 86% sensitivity and 99% specificity for the images in our local health region database.

Subfields: Artificial Intelligence, Biomedical Engineering, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/6954161b398248dce2dff204422ab555bc381d0c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Ashirbani Saha
2. Mateusz Buda
3. Maciej A. Mazurowski
4. Osamu Abe
5. Dong Ni
6. Shawn Li
7. Li Liu
8. Mustafa R. Bashir

In Biomedical Engineering

1. Maciej A. Mazurowski
2. Mustafa R. Bashir
3. Mateusz Buda
4. Ashirbani Saha
5. Xuelong Li
6. Xin Yang

In Radiology, Nuclear Medicine and Imaging

1. Maciej A. Mazurowski
2. Mustafa R. Bashir
3. Mateusz Buda
4. Ashirbani Saha
5. Xuelong Li
6. Xin Yang
7. Xin Zhu
8. Weihao Weng

Citations in the corpus

• Deep learning in radiology: an overview of the concepts and a survey of the state of the art with focus on MRI (2018, methodology): “An example of this is the study [29], which performed ultrasound imaging-based thyroid nodule classification using features extracted from a fine-tuned pre-trained GoogLeNet.”
• Deep learning with convolutional neural network in radiology (2018, methodology): “98 by using CNN models based on ultrasound (US) images [50].”
• A scoping review of transfer learning research on medical image analysis using ImageNet (2020, neutral): “, [55,57,58]), suggesting that”
• Machine Learning in Ultrasound Computer-Aided Diagnostic Systems: A Survey (2018, methodology): “Chi et al. employed the GoogLeNet to classify the thyroid nodule [27].”
• Deep Learning in Medical Ultrasound Analysis: A Review (2019, background): “for a specific task have been demonstrated to improve the diagnosis performance [48].”