Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers
Purpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US). Materials and Methods We included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Korean Society of Radiology
2020-09-01
|
| Series: | 대한영상의학회지 |
| Subjects: | |
| Online Access: | https://doi.org/10.3348/jksr.2019.0147 |
| _version_ | 1818523172567252992 |
|---|---|
| author | Jieun Kil Kwang Gi Kim Young Jae Kim Hye Ryoung Koo Jeong Seon Park |
| author_facet | Jieun Kil Kwang Gi Kim Young Jae Kim Hye Ryoung Koo Jeong Seon Park |
| author_sort | Jieun Kil |
| collection | DOAJ |
| description | Purpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative
ultrasonography (US).
Materials and Methods We included representative images from 229 US-based patients
(male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on
preoperative US and subsequently underwent thyroid surgery. After selecting each representative
transverse or longitudinal US image, we created a data set from the resulting database of
898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks
were used for deep learning with a convolutional neural network. We compared the clinical
and histological features between patients with and without recurrence. The predictive performance
of the deep learning model between groups was evaluated using receiver operating
characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the
prognostic performance of the deep learning model to predict recurrent thyroid cancer.
Results Tumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and
multifocality varied significantly between the groups with and without recurrence (p < 0.05).
The overall mean area under the curve (AUC) value of the deep learning model for prediction of
recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma
and 0.79 ± 0.16 in microcarcinoma.
Conclusion A deep learning model for analysis of US images of thyroid cancer showed the possibility
of predicting recurrence of thyroid cancer. |
| first_indexed | 2024-12-11T05:41:54Z |
| format | Article |
| id | doaj.art-c73580f221bc466a8c92d7fd3ef681a5 |
| institution | Directory Open Access Journal |
| issn | 1738-2637 2288-2928 |
| language | English |
| last_indexed | 2024-12-11T05:41:54Z |
| publishDate | 2020-09-01 |
| publisher | The Korean Society of Radiology |
| record_format | Article |
| series | 대한영상의학회지 |
| spelling | doaj.art-c73580f221bc466a8c92d7fd3ef681a52022-12-22T01:19:06ZengThe Korean Society of Radiology대한영상의학회지1738-26372288-29282020-09-0181511641174https://doi.org/10.3348/jksr.2019.0147Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid CancersJieun Kil0Kwang Gi Kim1Young Jae Kim2Hye Ryoung Koo3Jeong Seon Park4Department of Radiology, Hanyang University College of Medicine, Seoul, KoreaDepartment of Biomedical Engineering, College of Medicine, Gachon University, Incheon, KoreaDepartment of Biomedical Engineering, College of Medicine, Gachon University, Incheon, KoreaDepartment of Radiology, Hanyang University College of Medicine, Seoul, KoreaDepartment of Radiology, Hanyang University College of Medicine, Seoul, KoreaPurpose To evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US). Materials and Methods We included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on preoperative US and subsequently underwent thyroid surgery. After selecting each representative transverse or longitudinal US image, we created a data set from the resulting database of 898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks were used for deep learning with a convolutional neural network. We compared the clinical and histological features between patients with and without recurrence. The predictive performance of the deep learning model between groups was evaluated using receiver operating characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the prognostic performance of the deep learning model to predict recurrent thyroid cancer. Results Tumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and multifocality varied significantly between the groups with and without recurrence (p < 0.05). The overall mean area under the curve (AUC) value of the deep learning model for prediction of recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma and 0.79 ± 0.16 in microcarcinoma. Conclusion A deep learning model for analysis of US images of thyroid cancer showed the possibility of predicting recurrence of thyroid cancer.https://doi.org/10.3348/jksr.2019.0147deep learningthyroid cancerpapillaryultrasonographyrecurrence |
| spellingShingle | Jieun Kil Kwang Gi Kim Young Jae Kim Hye Ryoung Koo Jeong Seon Park Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers 대한영상의학회지 deep learning thyroid cancer papillary ultrasonography recurrence |
| title | Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers |
| title_full | Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers |
| title_fullStr | Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers |
| title_full_unstemmed | Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers |
| title_short | Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers |
| title_sort | deep learning in thyroid ultrasonography to predict tumor recurrence in thyroid cancers |
| topic | deep learning thyroid cancer papillary ultrasonography recurrence |
| url | https://doi.org/10.3348/jksr.2019.0147 |
| work_keys_str_mv | AT jieunkil deeplearninginthyroidultrasonographytopredicttumorrecurrenceinthyroidcancers AT kwanggikim deeplearninginthyroidultrasonographytopredicttumorrecurrenceinthyroidcancers AT youngjaekim deeplearninginthyroidultrasonographytopredicttumorrecurrenceinthyroidcancers AT hyeryoungkoo deeplearninginthyroidultrasonographytopredicttumorrecurrenceinthyroidcancers AT jeongseonpark deeplearninginthyroidultrasonographytopredicttumorrecurrenceinthyroidcancers |