Yang Zhang1, Jeon-Hor Chen2, Yezhi Lin3, Siwa Chan4, Jiejie Zhou3, Daniel Chow1, Peter Chang1, Tiffany Kwong1, Dah-Cherng Yeh4, Xinxin Wang1, Ritesh Parajuli5, Rita S Mehta5, Meihao Wang6, Min-Ying Su7,8. 1. Department of Radiological Sciences, University of California, Irvine, CA, USA. 2. Department of Radiology, E-Da Hospital and I-Shou University, No. 1, Yida Road, Jiaosu Village, Yanchao District, 8244, Kaohsiung, Taiwan. 3. Department of Radiology, First Affiliate Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, People's Republic of China. 4. Department of Medical Imaging, Taichung Tzu-Chi Hospital, Taichung, Taiwan. 5. Department of Medicine, University of California, Irvine, CA, USA. 6. Department of Radiology, First Affiliate Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, People's Republic of China. wzwmh@wmu.edu.cn. 7. Department of Radiological Sciences, University of California, Irvine, CA, USA. msu@uci.edu. 8. John Tu and Thomas Yuen Center for Functional Onco-Imaging, 164 Irvine Hall, University of California, Irvine, CA, 92697-5020, USA. msu@uci.edu.
Abstract
OBJECTIVES: To apply deep learning algorithms using a conventional convolutional neural network (CNN) and a recurrent CNN to differentiate three breast cancer molecular subtypes on MRI. METHODS: A total of 244 patients were analyzed, 99 in training dataset scanned at 1.5 T and 83 in testing-1 and 62 in testing-2 scanned at 3 T. Patients were classified into 3 subtypes based on hormonal receptor (HR) and HER2 receptor: (HR+/HER2-), HER2+, and triple negative (TN). Only images acquired in the DCE sequence were used in the analysis. The smallest bounding box covering tumor ROI was used as the input for deep learning to develop the model in the training dataset, by using a conventional CNN and the convolutional long short-term memory (CLSTM). Then, transfer learning was applied to re-tune the model using testing-1(2) and evaluated in testing-2(1). RESULTS: In the training dataset, the mean accuracy evaluated using tenfold cross-validation was higher by using CLSTM (0.91) than by using CNN (0.79). When the developed model was applied to the independent testing datasets, the accuracy was 0.4-0.5. With transfer learning by re-tuning parameters in testing-1, the mean accuracy reached 0.91 by CNN and 0.83 by CLSTM, and improved accuracy in testing-2 from 0.47 to 0.78 by CNN and from 0.39 to 0.74 by CLSTM. Overall, transfer learning could improve the classification accuracy by greater than 30%. CONCLUSIONS: The recurrent network using CLSTM could track changes in signal intensity during DCE acquisition, and achieved a higher accuracy compared with conventional CNN during training. For datasets acquired using different settings, transfer learning can be applied to re-tune the model and improve accuracy. KEY POINTS: • Deep learning can be applied to differentiate breast cancer molecular subtypes. • The recurrent neural network using CLSTM could track the change of signal intensity in DCE images, and achieved a higher accuracy compared with conventional CNN during training. • For datasets acquired using different scanners with different imaging protocols, transfer learning provided an efficient method to re-tune the classification model and improve accuracy.
OBJECTIVES: To apply deep learning algorithms using a conventional convolutional neural network (CNN) and a recurrent CNN to differentiate three breast cancer molecular subtypes on MRI. METHODS: A total of 244 patients were analyzed, 99 in training dataset scanned at 1.5 T and 83 in testing-1 and 62 in testing-2 scanned at 3 T. Patients were classified into 3 subtypes based on hormonal receptor (HR) and HER2 receptor: (HR+/HER2-), HER2+, and triple negative (TN). Only images acquired in the DCE sequence were used in the analysis. The smallest bounding box covering tumor ROI was used as the input for deep learning to develop the model in the training dataset, by using a conventional CNN and the convolutional long short-term memory (CLSTM). Then, transfer learning was applied to re-tune the model using testing-1(2) and evaluated in testing-2(1). RESULTS: In the training dataset, the mean accuracy evaluated using tenfold cross-validation was higher by using CLSTM (0.91) than by using CNN (0.79). When the developed model was applied to the independent testing datasets, the accuracy was 0.4-0.5. With transfer learning by re-tuning parameters in testing-1, the mean accuracy reached 0.91 by CNN and 0.83 by CLSTM, and improved accuracy in testing-2 from 0.47 to 0.78 by CNN and from 0.39 to 0.74 by CLSTM. Overall, transfer learning could improve the classification accuracy by greater than 30%. CONCLUSIONS: The recurrent network using CLSTM could track changes in signal intensity during DCE acquisition, and achieved a higher accuracy compared with conventional CNN during training. For datasets acquired using different settings, transfer learning can be applied to re-tune the model and improve accuracy. KEY POINTS: • Deep learning can be applied to differentiate breast cancer molecular subtypes. • The recurrent neural network using CLSTM could track the change of signal intensity in DCE images, and achieved a higher accuracy compared with conventional CNN during training. • For datasets acquired using different scanners with different imaging protocols, transfer learning provided an efficient method to re-tune the classification model and improve accuracy.
Entities:
Keywords:
Breast neoplasms; Machine learning; Magnetic resonance imaging; Receptor, ErbB-2; Receptors, estrogen