Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer
Abstract Background To determine whether deep learning reconstruction (DLR) could improve the image quality of rectal MR images, and to explore the discrimination of the TN stage of rectal cancer by different readers and deep learning classification models, compared with conventional MR images witho...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-07-01
|
| Series: | BMC Medical Imaging |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12880-025-01775-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849333996517851136 |
|---|---|
| author | Yuan Yuan Shengnan Ren Haidi Lu Fangying Chen Lei Xiang Ryan Chamberlain Chengwei Shao Jianping Lu Fu Shen Luguang Chen |
| author_facet | Yuan Yuan Shengnan Ren Haidi Lu Fangying Chen Lei Xiang Ryan Chamberlain Chengwei Shao Jianping Lu Fu Shen Luguang Chen |
| author_sort | Yuan Yuan |
| collection | DOAJ |
| description | Abstract Background To determine whether deep learning reconstruction (DLR) could improve the image quality of rectal MR images, and to explore the discrimination of the TN stage of rectal cancer by different readers and deep learning classification models, compared with conventional MR images without DLR. Methods Images of high-resolution T2-weighted, diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted imaging (CE-T1WI) from patients with pathologically diagnosed rectal cancer were retrospectively processed with and without DLR and assessed by five readers. The first two readers measured the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the lesions. The overall image quality and lesion display performance for each sequence with and without DLR were independently scored using a five-point scale, and the TN stage of rectal cancer lesions was evaluated by the other three readers. Fifty of the patients were randomly selected to further make a comparison between DLR and traditional denoising filter. Deep learning classification models were developed and compared for the TN stage. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) were used to evaluate the diagnostic performance of the proposed model. Results Overall, 178 patients were evaluated. The SNR and CNR of the lesion on images with DLR were significantly higher than those without DLR, for T2WI, DWI and CE-T1WI, respectively (p < 0.0001). A significant difference was observed in overall image quality and lesion display performance between images with and without DLR (p < 0.0001). The image quality scores, SNR, and CNR values of DLR image set were significantly larger than those of original and filter enhancement image sets (all p values < 0.05) for all the three sequences, respectively. The deep learning classification models with DLR achieved good discrimination of the TN stage, with area under the curve (AUC) values of 0.937 (95% CI 0.839–0.977) and 0.824 (95% CI 0.684–0.913) in the test sets, respectively. Conclusion Deep learning reconstruction and classification models could improve the image quality of rectal MRI images and enhance the diagnostic performance for determining the TN stage of patients with rectal cancer. |
| format | Article |
| id | doaj-art-50e7e558e7da445abc97231c33d16cb0 |
| institution | Kabale University |
| issn | 1471-2342 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Imaging |
| spelling | doaj-art-50e7e558e7da445abc97231c33d16cb02025-08-20T03:45:41ZengBMCBMC Medical Imaging1471-23422025-07-0125111310.1186/s12880-025-01775-1Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancerYuan Yuan0Shengnan Ren1Haidi Lu2Fangying Chen3Lei Xiang4Ryan Chamberlain5Chengwei Shao6Jianping Lu7Fu Shen8Luguang Chen9Department of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Nuclear Medicine, Shanghai Fourth People’s Hospital, School of Medicine, Tongji UniversityDepartment of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Research and Development, Subtle MedicalDepartment of Research and Development, Subtle MedicalDepartment of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Radiology, Changhai Hospital, Naval Medical UniversityDepartment of Radiology, Changhai Hospital, Naval Medical UniversityAbstract Background To determine whether deep learning reconstruction (DLR) could improve the image quality of rectal MR images, and to explore the discrimination of the TN stage of rectal cancer by different readers and deep learning classification models, compared with conventional MR images without DLR. Methods Images of high-resolution T2-weighted, diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted imaging (CE-T1WI) from patients with pathologically diagnosed rectal cancer were retrospectively processed with and without DLR and assessed by five readers. The first two readers measured the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the lesions. The overall image quality and lesion display performance for each sequence with and without DLR were independently scored using a five-point scale, and the TN stage of rectal cancer lesions was evaluated by the other three readers. Fifty of the patients were randomly selected to further make a comparison between DLR and traditional denoising filter. Deep learning classification models were developed and compared for the TN stage. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) were used to evaluate the diagnostic performance of the proposed model. Results Overall, 178 patients were evaluated. The SNR and CNR of the lesion on images with DLR were significantly higher than those without DLR, for T2WI, DWI and CE-T1WI, respectively (p < 0.0001). A significant difference was observed in overall image quality and lesion display performance between images with and without DLR (p < 0.0001). The image quality scores, SNR, and CNR values of DLR image set were significantly larger than those of original and filter enhancement image sets (all p values < 0.05) for all the three sequences, respectively. The deep learning classification models with DLR achieved good discrimination of the TN stage, with area under the curve (AUC) values of 0.937 (95% CI 0.839–0.977) and 0.824 (95% CI 0.684–0.913) in the test sets, respectively. Conclusion Deep learning reconstruction and classification models could improve the image quality of rectal MRI images and enhance the diagnostic performance for determining the TN stage of patients with rectal cancer.https://doi.org/10.1186/s12880-025-01775-1Rectal cancerMagnetic resonance imagingDeep learningReconstruction |
| spellingShingle | Yuan Yuan Shengnan Ren Haidi Lu Fangying Chen Lei Xiang Ryan Chamberlain Chengwei Shao Jianping Lu Fu Shen Luguang Chen Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer BMC Medical Imaging Rectal cancer Magnetic resonance imaging Deep learning Reconstruction |
| title | Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer |
| title_full | Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer |
| title_fullStr | Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer |
| title_full_unstemmed | Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer |
| title_short | Preoperative MRI-based deep learning reconstruction and classification model for assessing rectal cancer |
| title_sort | preoperative mri based deep learning reconstruction and classification model for assessing rectal cancer |
| topic | Rectal cancer Magnetic resonance imaging Deep learning Reconstruction |
| url | https://doi.org/10.1186/s12880-025-01775-1 |
| work_keys_str_mv | AT yuanyuan preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT shengnanren preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT haidilu preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT fangyingchen preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT leixiang preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT ryanchamberlain preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT chengweishao preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT jianpinglu preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT fushen preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer AT luguangchen preoperativemribaseddeeplearningreconstructionandclassificationmodelforassessingrectalcancer |