3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study
Abstract Purposes The presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients. Methods This retrospective stu...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2025-01-01
|
| Series: | Insights into Imaging |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13244-024-01896-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571571622903808 |
|---|---|
| author | Jie Bao Litao Zhao Xiaomeng Qiao Zhenkai Li Yanting Ji Yueting Su Libiao Ji Junkang Shen Jiangang Liu Jie Tian Ximing Wang Hailin Shen Chunhong Hu |
| author_facet | Jie Bao Litao Zhao Xiaomeng Qiao Zhenkai Li Yanting Ji Yueting Su Libiao Ji Junkang Shen Jiangang Liu Jie Tian Ximing Wang Hailin Shen Chunhong Hu |
| author_sort | Jie Bao |
| collection | DOAJ |
| description | Abstract Purposes The presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients. Methods This retrospective study included a bi-parametric MRI of 1567 consecutive male patients from six centers (Centers 1–6) between Jan 2015 and Dec 2020. Deep learning models with double channel attention modules based on MRI (AttenNet) for predicting PCa and csPCa were constructed separately. Each model was first pretrained using 1144 PI-RADS 1–2 and 4–5 images and then retrained using 238 PI-RADS 3 images from three training centers (centers 1–3), and tested using 185 PI-RADS 3 images from the other three testing centers (centers 4–6). Results Our AttenNet models achieved excellent prediction performances in testing cohort of center 4–6 with the area under the receiver operating characteristic curves (AUC) of 0.795 (95% CI: [0.700, 0.891]), 0.963 (95% CI: [0.915, 1]) and 0.922 (95% CI: [0.810, 1]) in predicting PCa, and the corresponding AUCs were 0.827 (95% CI: [0.703, 0.952]) and 0.926 (95% CI: [0.846, 1]) in predicting csPCa in testing cohort of center 4 and center 5. In particular, 71.1% to 92.2% of non-csPCa patients were identified by our model in three testing cohorts, who can spare from invasive biopsy or RP procedure. Conclusions Our model offers a noninvasive screening clinical tool to re-stratify risks in PI-RADS 3 patients, thereby reducing unnecessary invasive biopsies and improving the effectiveness of biopsies. Critical relevance statement The deep learning model with MRI can help to screen out csPCa in PI-RADS category 3. Key Points AttenNet models included channel attention and soft attention modules. 71.1–92.2% of non-csPCa patients were identified by the AttenNet model. The AttenNet models can be a screen clinical tool to re-stratify risks in PI-RADS 3 patients. Graphical Abstract |
| format | Article |
| id | doaj-art-2bb2ae340e304e0583142fb062f416a2 |
| institution | Kabale University |
| issn | 1869-4101 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Insights into Imaging |
| spelling | doaj-art-2bb2ae340e304e0583142fb062f416a22025-02-02T12:27:55ZengSpringerOpenInsights into Imaging1869-41012025-01-0116111310.1186/s13244-024-01896-13D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter studyJie Bao0Litao Zhao1Xiaomeng Qiao2Zhenkai Li3Yanting Ji4Yueting Su5Libiao Ji6Junkang Shen7Jiangang Liu8Jie Tian9Ximing Wang10Hailin Shen11Chunhong Hu12Department of Radiology, The First Affiliated Hospital of Soochow UniversitySchool of Engineering Medicine, Beihang UniversityDepartment of Radiology, The First Affiliated Hospital of Soochow UniversityDepartment of Radiology, Suzhou Kowloon Hospital, Shanghai Jiaotong University School of MedicineDepartment of Radiology, The First Affiliated Hospital of Soochow UniversityDepartment of Radiology, The People’s Hospital of TaizhouDepartment of Radiology, Changshu No.1 People’s HospitalDepartment of Radiology, The Second Affiliated Hospital of Soochow UniversitySchool of Engineering Medicine, Beihang UniversitySchool of Engineering Medicine, Beihang UniversityDepartment of Radiology, The First Affiliated Hospital of Soochow UniversityDepartment of Radiology, Suzhou Kowloon Hospital, Shanghai Jiaotong University School of MedicineDepartment of Radiology, The First Affiliated Hospital of Soochow UniversityAbstract Purposes The presence of clinically significant prostate cancer (csPCa) is equivocal for patients with prostate imaging reporting and data system (PI-RADS) category 3. We aim to develop deep learning models for re-stratify risks in PI-RADS category 3 patients. Methods This retrospective study included a bi-parametric MRI of 1567 consecutive male patients from six centers (Centers 1–6) between Jan 2015 and Dec 2020. Deep learning models with double channel attention modules based on MRI (AttenNet) for predicting PCa and csPCa were constructed separately. Each model was first pretrained using 1144 PI-RADS 1–2 and 4–5 images and then retrained using 238 PI-RADS 3 images from three training centers (centers 1–3), and tested using 185 PI-RADS 3 images from the other three testing centers (centers 4–6). Results Our AttenNet models achieved excellent prediction performances in testing cohort of center 4–6 with the area under the receiver operating characteristic curves (AUC) of 0.795 (95% CI: [0.700, 0.891]), 0.963 (95% CI: [0.915, 1]) and 0.922 (95% CI: [0.810, 1]) in predicting PCa, and the corresponding AUCs were 0.827 (95% CI: [0.703, 0.952]) and 0.926 (95% CI: [0.846, 1]) in predicting csPCa in testing cohort of center 4 and center 5. In particular, 71.1% to 92.2% of non-csPCa patients were identified by our model in three testing cohorts, who can spare from invasive biopsy or RP procedure. Conclusions Our model offers a noninvasive screening clinical tool to re-stratify risks in PI-RADS 3 patients, thereby reducing unnecessary invasive biopsies and improving the effectiveness of biopsies. Critical relevance statement The deep learning model with MRI can help to screen out csPCa in PI-RADS category 3. Key Points AttenNet models included channel attention and soft attention modules. 71.1–92.2% of non-csPCa patients were identified by the AttenNet model. The AttenNet models can be a screen clinical tool to re-stratify risks in PI-RADS 3 patients. Graphical Abstracthttps://doi.org/10.1186/s13244-024-01896-1Deep learningMRIClinically significant prostate cancerPI-RADS |
| spellingShingle | Jie Bao Litao Zhao Xiaomeng Qiao Zhenkai Li Yanting Ji Yueting Su Libiao Ji Junkang Shen Jiangang Liu Jie Tian Ximing Wang Hailin Shen Chunhong Hu 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study Insights into Imaging Deep learning MRI Clinically significant prostate cancer PI-RADS |
| title | 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study |
| title_full | 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study |
| title_fullStr | 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study |
| title_full_unstemmed | 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study |
| title_short | 3D-AttenNet model can predict clinically significant prostate cancer in PI-RADS category 3 patients: a retrospective multicenter study |
| title_sort | 3d attennet model can predict clinically significant prostate cancer in pi rads category 3 patients a retrospective multicenter study |
| topic | Deep learning MRI Clinically significant prostate cancer PI-RADS |
| url | https://doi.org/10.1186/s13244-024-01896-1 |
| work_keys_str_mv | AT jiebao 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT litaozhao 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT xiaomengqiao 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT zhenkaili 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT yantingji 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT yuetingsu 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT libiaoji 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT junkangshen 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT jiangangliu 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT jietian 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT ximingwang 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT hailinshen 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy AT chunhonghu 3dattennetmodelcanpredictclinicallysignificantprostatecancerinpiradscategory3patientsaretrospectivemulticenterstudy |