A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI
Prostate Cancer (PCa) is the most common oncological disease in Western men. Even though a growing effort has been carried out by the scientific community in recent years, accurate and reliable automated PCa detection methods on multiparametric Magnetic Resonance Imaging (mpMRI) are still a compelli...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/1/338 |
| _version_ | 1811270329753927680 |
|---|---|
| author | Paulo Lapa Mauro Castelli Ivo Gonçalves Evis Sala Leonardo Rundo |
| author_facet | Paulo Lapa Mauro Castelli Ivo Gonçalves Evis Sala Leonardo Rundo |
| author_sort | Paulo Lapa |
| collection | DOAJ |
| description | Prostate Cancer (PCa) is the most common oncological disease in Western men. Even though a growing effort has been carried out by the scientific community in recent years, accurate and reliable automated PCa detection methods on multiparametric Magnetic Resonance Imaging (mpMRI) are still a compelling issue. In this work, a Deep Neural Network architecture is developed for the task of classifying clinically significant PCa on non-contrast-enhanced MR images. In particular, we propose the use of Conditional Random Fields as a Recurrent Neural Network (CRF-RNN) to enhance the classification performance of XmasNet, a Convolutional Neural Network (CNN) architecture specifically tailored to the PROSTATEx17 Challenge. The devised approach builds a hybrid end-to-end trainable network, CRF-XmasNet, composed of an initial CNN component performing feature extraction and a CRF-based probabilistic graphical model component for structured prediction, without the need for two separate training procedures. Experimental results show the suitability of this method in terms of classification accuracy and training time, even though the high-variability of the observed results must be reduced before transferring the resulting architecture to a clinical environment. Interestingly, the use of CRFs as a separate postprocessing method achieves significantly lower performance with respect to the proposed hybrid end-to-end approach. The proposed hybrid end-to-end CRF-RNN approach yields excellent peak performance for all the CNN architectures taken into account, but it shows a high-variability, thus requiring future investigation on the integration of CRFs into a CNN. |
| first_indexed | 2024-04-12T21:58:54Z |
| format | Article |
| id | doaj.art-e0a0df3e9c694043aa6b6ebfa2fc50b3 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-04-12T21:58:54Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-e0a0df3e9c694043aa6b6ebfa2fc50b32022-12-22T03:15:13ZengMDPI AGApplied Sciences2076-34172020-01-0110133810.3390/app10010338app10010338A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRIPaulo Lapa0Mauro Castelli1Ivo Gonçalves2Evis Sala3Leonardo Rundo4Nova Information Management School (NOVA IMS), Campus de Campolide, Universidade Nova de Lisboa, 1070-332 Lisboa, PortugalNova Information Management School (NOVA IMS), Campus de Campolide, Universidade Nova de Lisboa, 1070-332 Lisboa, PortugalINESC Coimbra, DEEC, University of Coimbra, Pólo 2, 3030-290 Coimbra, PortugalDepartment of Radiology, University of Cambridge, Cambridge CB2 0QQ, UKDepartment of Radiology, University of Cambridge, Cambridge CB2 0QQ, UKProstate Cancer (PCa) is the most common oncological disease in Western men. Even though a growing effort has been carried out by the scientific community in recent years, accurate and reliable automated PCa detection methods on multiparametric Magnetic Resonance Imaging (mpMRI) are still a compelling issue. In this work, a Deep Neural Network architecture is developed for the task of classifying clinically significant PCa on non-contrast-enhanced MR images. In particular, we propose the use of Conditional Random Fields as a Recurrent Neural Network (CRF-RNN) to enhance the classification performance of XmasNet, a Convolutional Neural Network (CNN) architecture specifically tailored to the PROSTATEx17 Challenge. The devised approach builds a hybrid end-to-end trainable network, CRF-XmasNet, composed of an initial CNN component performing feature extraction and a CRF-based probabilistic graphical model component for structured prediction, without the need for two separate training procedures. Experimental results show the suitability of this method in terms of classification accuracy and training time, even though the high-variability of the observed results must be reduced before transferring the resulting architecture to a clinical environment. Interestingly, the use of CRFs as a separate postprocessing method achieves significantly lower performance with respect to the proposed hybrid end-to-end approach. The proposed hybrid end-to-end CRF-RNN approach yields excellent peak performance for all the CNN architectures taken into account, but it shows a high-variability, thus requiring future investigation on the integration of CRFs into a CNN.https://www.mdpi.com/2076-3417/10/1/338prostate cancer detectionmagnetic resonance imagingconvolutional neural networksconditional random fieldsrecurrent neural networks |
| spellingShingle | Paulo Lapa Mauro Castelli Ivo Gonçalves Evis Sala Leonardo Rundo A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI Applied Sciences prostate cancer detection magnetic resonance imaging convolutional neural networks conditional random fields recurrent neural networks |
| title | A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI |
| title_full | A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI |
| title_fullStr | A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI |
| title_full_unstemmed | A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI |
| title_short | A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI |
| title_sort | hybrid end to end approach integrating conditional random fields into cnns for prostate cancer detection on mri |
| topic | prostate cancer detection magnetic resonance imaging convolutional neural networks conditional random fields recurrent neural networks |
| url | https://www.mdpi.com/2076-3417/10/1/338 |
| work_keys_str_mv | AT paulolapa ahybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT maurocastelli ahybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT ivogoncalves ahybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT evissala ahybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT leonardorundo ahybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT paulolapa hybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT maurocastelli hybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT ivogoncalves hybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT evissala hybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri AT leonardorundo hybridendtoendapproachintegratingconditionalrandomfieldsintocnnsforprostatecancerdetectiononmri |