Contrast-enhanced CT image assessment of cervical lymph node metastasis in oral cancer patients using a deep learning system of artificial intelligence

Publication date: Available online 15 October 2018

Source: Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology

Author(s): Yoshiko Ariji, Motoki Fukuda, Yoshitaka Kise, Michihito Nozawa, Yudai Yanashita, Hiroshi Fujita, Akitoshi Katsumata, Eiichiro Ariji

Abstract

Objective

Although deep learning has been applied to medical images, its application to the diagnosis of cervical lymph nodes in patients with oral cancer has not yet been reported. The purpose of this study was to evaluate the performance of deep-learning image classification for diagnosis of lymph node metastasis.

Study Design

The imaging data used for evaluation consisted of CT images of 127 histologically proven positive cervical lymph nodes and 314 histologically proven negative lymph nodes from 45 patients with oral squamous cell carcinoma. The performance of a deep-learning image classification system for the diagnosis of lymph node metastasis on CT images was compared with the diagnostic outcomes of two experienced radiologists, using the Mann-Whitney U Test and chi-squared analysis.

Results

The performance of the deep-learning image classification system resulted in accuracy of 78.2%, sensitivity of 75.4%, specificity of 81.0%, positive predictive value of 79.9, negative predictive value of 77.1%, and area under the receiver operating characteristic curve of 0.80. These values were not significantly different from the radiologists' performance.

Conclusion

The deep learning system produced diagnostic outcomes similar to those of the radiologists, which suggests that it may be valuable for diagnostic support.

https://ift.tt/2CM8oyR