A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs
Abstract We aimed to apply a potent deep learning network, NAFNet, to predict adverse pathology events and biochemical recurrence-free survival (bRFS) based on pre-treatment MRI imaging. 514 prostate cancer patients from six tertiary hospitals throughout China from 2017 and 2021 were included. A tot...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-12-01
|
| Series: | npj Precision Oncology |
| Online Access: | https://doi.org/10.1038/s41698-023-00481-x |
| _version_ | 1797388533342666752 |
|---|---|
| author | Wei-jie Gu Zheng Liu Yun-jie Yang Xuan-zhi Zhang Liang-yu Chen Fang-ning Wan Xiao-hang Liu Zhang-zhe Chen Yun-yi Kong Bo Dai |
| author_facet | Wei-jie Gu Zheng Liu Yun-jie Yang Xuan-zhi Zhang Liang-yu Chen Fang-ning Wan Xiao-hang Liu Zhang-zhe Chen Yun-yi Kong Bo Dai |
| author_sort | Wei-jie Gu |
| collection | DOAJ |
| description | Abstract We aimed to apply a potent deep learning network, NAFNet, to predict adverse pathology events and biochemical recurrence-free survival (bRFS) based on pre-treatment MRI imaging. 514 prostate cancer patients from six tertiary hospitals throughout China from 2017 and 2021 were included. A total of 367 patients from Fudan University Shanghai Cancer Center with whole-mount histopathology of radical prostatectomy specimens were assigned to the internal set, and cancer lesions were delineated with whole-mount pathology as the reference. The external test set included 147 patients with BCR data from five other institutes. The prediction model (NAFNet-classifier) and integrated nomogram (DL-nomogram) were constructed based on NAFNet. We then compared DL-nomogram with radiology score (PI-RADS), and clinical score (Cancer of the Prostate Risk Assessment score (CAPRA)). After training and validation in the internal set, ROC curves in the external test set showed that NAFNet-classifier alone outperformed ResNet50 in predicting adverse pathology. The DL-nomogram, including the NAFNet-classifier, clinical T stage and biopsy results, showed the highest AUC (0.915, 95% CI: 0.871–0.959) and accuracy (0.850) compared with the PI-RADS and CAPRA scores. Additionally, the DL-nomogram outperformed the CAPRA score with a higher C-index (0.732, P < 0.001) in predicting bRFS. Based on this newly-developed deep learning network, NAFNet, our DL-nomogram could accurately predict adverse pathology and poor prognosis, providing a potential AI tools in medical imaging risk stratification. |
| first_indexed | 2024-03-08T22:42:13Z |
| format | Article |
| id | doaj.art-48b1d9b2679a40e3970ca7ebef6362be |
| institution | Directory Open Access Journal |
| issn | 2397-768X |
| language | English |
| last_indexed | 2024-03-08T22:42:13Z |
| publishDate | 2023-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Precision Oncology |
| spelling | doaj.art-48b1d9b2679a40e3970ca7ebef6362be2023-12-17T12:05:27ZengNature Portfolionpj Precision Oncology2397-768X2023-12-017111010.1038/s41698-023-00481-xA deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIsWei-jie Gu0Zheng Liu1Yun-jie Yang2Xuan-zhi Zhang3Liang-yu Chen4Fang-ning Wan5Xiao-hang Liu6Zhang-zhe Chen7Yun-yi Kong8Bo Dai9Department of Urology, Fudan University Shanghai Cancer CenterDepartment of Urology, Fudan University Shanghai Cancer CenterDepartment of Urology, Fudan University Shanghai Cancer CenterDepartment of Urology, Fudan University Shanghai Cancer CenterDepartment of Foundation Model, MEGVII TechnologyDepartment of Urology, Fudan University Shanghai Cancer CenterDepartment of Oncology, Shanghai Medical College, Fudan UniversityDepartment of Oncology, Shanghai Medical College, Fudan UniversityDepartment of Oncology, Shanghai Medical College, Fudan UniversityDepartment of Urology, Fudan University Shanghai Cancer CenterAbstract We aimed to apply a potent deep learning network, NAFNet, to predict adverse pathology events and biochemical recurrence-free survival (bRFS) based on pre-treatment MRI imaging. 514 prostate cancer patients from six tertiary hospitals throughout China from 2017 and 2021 were included. A total of 367 patients from Fudan University Shanghai Cancer Center with whole-mount histopathology of radical prostatectomy specimens were assigned to the internal set, and cancer lesions were delineated with whole-mount pathology as the reference. The external test set included 147 patients with BCR data from five other institutes. The prediction model (NAFNet-classifier) and integrated nomogram (DL-nomogram) were constructed based on NAFNet. We then compared DL-nomogram with radiology score (PI-RADS), and clinical score (Cancer of the Prostate Risk Assessment score (CAPRA)). After training and validation in the internal set, ROC curves in the external test set showed that NAFNet-classifier alone outperformed ResNet50 in predicting adverse pathology. The DL-nomogram, including the NAFNet-classifier, clinical T stage and biopsy results, showed the highest AUC (0.915, 95% CI: 0.871–0.959) and accuracy (0.850) compared with the PI-RADS and CAPRA scores. Additionally, the DL-nomogram outperformed the CAPRA score with a higher C-index (0.732, P < 0.001) in predicting bRFS. Based on this newly-developed deep learning network, NAFNet, our DL-nomogram could accurately predict adverse pathology and poor prognosis, providing a potential AI tools in medical imaging risk stratification.https://doi.org/10.1038/s41698-023-00481-x |
| spellingShingle | Wei-jie Gu Zheng Liu Yun-jie Yang Xuan-zhi Zhang Liang-yu Chen Fang-ning Wan Xiao-hang Liu Zhang-zhe Chen Yun-yi Kong Bo Dai A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs npj Precision Oncology |
| title | A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs |
| title_full | A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs |
| title_fullStr | A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs |
| title_full_unstemmed | A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs |
| title_short | A deep learning model, NAFNet, predicts adverse pathology and recurrence in prostate cancer using MRIs |
| title_sort | deep learning model nafnet predicts adverse pathology and recurrence in prostate cancer using mris |
| url | https://doi.org/10.1038/s41698-023-00481-x |
| work_keys_str_mv | AT weijiegu adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT zhengliu adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT yunjieyang adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT xuanzhizhang adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT liangyuchen adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT fangningwan adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT xiaohangliu adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT zhangzhechen adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT yunyikong adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT bodai adeeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT weijiegu deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT zhengliu deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT yunjieyang deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT xuanzhizhang deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT liangyuchen deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT fangningwan deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT xiaohangliu deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT zhangzhechen deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT yunyikong deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris AT bodai deeplearningmodelnafnetpredictsadversepathologyandrecurrenceinprostatecancerusingmris |