Effect of domain knowledge encoding in CNN model architecture—a prostate cancer study using mpMRI images

PeerJ

article

Piotr Sobecki¹  Rafał Jóźwiak¹  Katarzyna Sklinda³  Artur Przelaskowski² 

[1] Applied Artificial Intelligence Laboratory, National Information Processing Institute;Faculty of Mathematics and Information Science, Warsaw University of Technology;Department of Radiology, Centre of Postgraduate Medical Education

关键词: Artificial intelligence;    Machine learning;    Prostate cancer;    PI-RADS;    mpMRI;    Prostate cancer diagnostics;    Knowledge-based modeling;    Neural network architectures;    Deep learning;    Multimodal convolutional neural networks;   

DOI  :  10.7717/peerj.11006

学科分类：社会科学、人文和艺术（综合）

来源: Inra

【 摘 要 】

BackgroundProstate cancer is one of the most common cancers worldwide. Currently, convolution neural networks (CNNs) are achieving remarkable success in various computer vision tasks, and in medical imaging research. Various CNN architectures and methodologies have been applied in the field of prostate cancer diagnosis. In this work, we evaluate the impact of the adaptation of a state-of-the-art CNN architecture on domain knowledge related to problems in the diagnosis of prostate cancer. The architecture of the final CNN model was optimised on the basis of the Prostate Imaging Reporting and Data System (PI-RADS) standard, which is currently the best available indicator in the acquisition, interpretation, and reporting of prostate multi-parametric magnetic resonance imaging (mpMRI) examinations.MethodsA dataset containing 330 suspicious findings identified using mpMRI was used. Two CNN models were subjected to comparative analysis. Both implement the concept of decision-level fusion for mpMRI data, providing a separate network for each multi-parametric series. The first model implements a simple fusion of multi-parametric features to formulate the final decision. The architecture of the second model reflects the diagnostic pathway of PI-RADS methodology, using information about a lesion’s primary anatomic location within the prostate gland. Both networks were experimentally tuned to successfully classify prostate cancer changes.ResultsThe optimised knowledge-encoded model achieved slightly better classification results compared with the traditional model architecture (AUC = 0.84 vs. AUC = 0.82). We found the proposed model to achieve convergence significantly faster.ConclusionsThe final knowledge-encoded CNN model provided more stable learning performance and faster convergence to optimal diagnostic accuracy. The results fail to demonstrate that PI-RADS-based modelling of CNN architecture can significantly improve performance of prostate cancer recognition using mpMRI.

【 授权许可】

CC BY