1. Image-driven classification of functioning and nonfunctioning pituitary adenoma by deep convolutional neural networks
- Author
-
Xiaobing Jiang, Hongyu Li, Lunshan Xu, Yubin Xie, Yihua Zhang, Jian Ren, Yonggao Mou, Qi Zhao, and Ke Sai
- Subjects
Computer science ,Biophysics ,Biochemistry ,Convolutional neural network ,Image (mathematics) ,03 medical and health sciences ,0302 clinical medicine ,Structural Biology ,Pituitary adenoma ,Genetics ,medicine ,Segmentation ,Pituitary adenomas ,030304 developmental biology ,ComputingMethodologies_COMPUTERGRAPHICS ,0303 health sciences ,medicine.diagnostic_test ,business.industry ,Deep learning ,Pattern recognition ,Magnetic resonance imaging ,medicine.disease ,Computer Science Applications ,030220 oncology & carcinogenesis ,Artificial intelligence ,Transfer of learning ,Model interpretation ,business ,TP248.13-248.65 ,Research Article ,MRI ,Biotechnology - Abstract
Graphical abstract, The secreting function of pituitary adenomas (PAs) plays a critical role in making the treatment strategies. However, Magnetic Resonance Imaging (MRI) analysis for pituitary adenomas is labor intensive and highly variable among radiologists. In this work, by applying convolutional neural network (CNN), we built a segmentation and classification model to help distinguish functioning pituitary adenomas from non-functioning subtypes with 3D MRI images from 185 patients with PAs (two centers). Specifically, the classification model adopts the concept of transfer learning and uses the pre-trained segmentation model to extract deep features from conventional MRI images. As a result, both segmentation and classification models obtained high performance in two internal validation datasets and an external testing dataset (for segmentation model: Dice score = 0.8188, 0.8091 and 0.8093 respectively; for classification model: AUROC = 0.8063, 0.7881 and 0.8478, respectively). In addition, the classification model considers the attention mechanism for better model interpretation. Taken together, this work provides the first deep learning-based tumor region segmentation and classification models of PAs, which enables early diagnosis and subtyping PAs from MRI images.
- Published
- 2021