Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics
Abstract Purpose To predict the 1p/19q molecular status of Lower-grade glioma (LGG) patients nondestructively, this study developed a deep learning (DL) approach using radiomic to provide a potential decision aid for clinical determination of molecular stratification of LGG. Methods The study retros...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-07-01
|
| Series: | BMC Cancer |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12885-025-14454-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849332893667557376 |
|---|---|
| author | Fengda Li Zeyi Li Hong Xu Gang Kong Ze Zhang Kaiyuan Cheng Longyuan Gu Lei Hua |
| author_facet | Fengda Li Zeyi Li Hong Xu Gang Kong Ze Zhang Kaiyuan Cheng Longyuan Gu Lei Hua |
| author_sort | Fengda Li |
| collection | DOAJ |
| description | Abstract Purpose To predict the 1p/19q molecular status of Lower-grade glioma (LGG) patients nondestructively, this study developed a deep learning (DL) approach using radiomic to provide a potential decision aid for clinical determination of molecular stratification of LGG. Methods The study retrospectively collected images and clinical data of 218 patients diagnosed with LGG between July 2018 and July 2022, including 155 cases from The Cancer Imaging Archive (TCIA) database and 63 cases from a regional medical centre. Patients' clinical data and MRI images were collected, including contrast-enhanced T1-weighted images and T2-weighted images. After pre-processing the image data, tumour regions of interest (ROI) were segmented by two senior neurosurgeons. In this study, an Ensemble Convolutional Neural Network (ECNN) was proposed to predict the 1p/19q status. This method, consisting of Variational Autoencoder (VAE), Information Gain (IG) and Convolutional Neural Network (CNN), is compared with four machine learning algorithms (Random Forest, Decision Tree, K-Nearest Neighbour, Gaussian Neff Bayes). Fivefold cross-validation was used to evaluate and calibrate the model. Precision, recall, accuracy, F1 score and area under the curve (AUC) were calculated to assess model performance. Results Our cohort comprises 118 patients diagnosed with 1p/19q codeletion and 100 patients diagnosed with 1p/19q non-codeletion. The study findings indicate that the ECNN method demonstrates excellent predictive performance on the validation dataset. Our model achieved an average precision of 0.981, average recall of 0.980, average F1-score of 0.981, and average accuracy of 0.981. The average area under the curve (AUC) for our model is 0.994, surpassing that of the other four traditional machine learning algorithms (AUC: 0.523–0.702). This suggests that the model based on the ECNN algorithm performs well in distinguishing the 1p/19q molecular status of LGG patients. Conclusion The deep learning model based on conventional MRI radiomic integrates VAE and IG methods. Compared with traditional machine learning algorithms, it shows the best performance in the prediction of 1p/19q molecular co-deletion status. It may become a potentially effective tool for non-invasively and effectively identifying molecular features of lower-grade glioma in the future, providing an important reference for clinicians to formulate individualized diagnosis and treatment plans. |
| format | Article |
| id | doaj-art-5729fc30e68c42e9938d33de5ca12b5c |
| institution | Kabale University |
| issn | 1471-2407 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Cancer |
| spelling | doaj-art-5729fc30e68c42e9938d33de5ca12b5c2025-08-20T03:46:04ZengBMCBMC Cancer1471-24072025-07-0125111210.1186/s12885-025-14454-9Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomicsFengda Li0Zeyi Li1Hong Xu2Gang Kong3Ze Zhang4Kaiyuan Cheng5Longyuan Gu6Lei Hua7Department of Neurosurgery, Changshu Hospital Affiliated to Soochow UniversitySchool of Computer Science, Nanjing University of Posts and TelecommunicationsDepartment of Neurosurgery, Changshu Hospital Affiliated to Soochow UniversityDepartment of Neurosurgery, Changshu Hospital Affiliated to Soochow UniversitySchool of Internet of Things, Nanjing University of Posts and TelecommunicationsDepartment of Neurosurgery, Changshu Hospital Affiliated to Soochow UniversityDepartment of Neurosurgery, Ji’an Central People’s HospitalDepartment of Neurosurgery, Affiliated Hospital of Xuzhou Medical UniversityAbstract Purpose To predict the 1p/19q molecular status of Lower-grade glioma (LGG) patients nondestructively, this study developed a deep learning (DL) approach using radiomic to provide a potential decision aid for clinical determination of molecular stratification of LGG. Methods The study retrospectively collected images and clinical data of 218 patients diagnosed with LGG between July 2018 and July 2022, including 155 cases from The Cancer Imaging Archive (TCIA) database and 63 cases from a regional medical centre. Patients' clinical data and MRI images were collected, including contrast-enhanced T1-weighted images and T2-weighted images. After pre-processing the image data, tumour regions of interest (ROI) were segmented by two senior neurosurgeons. In this study, an Ensemble Convolutional Neural Network (ECNN) was proposed to predict the 1p/19q status. This method, consisting of Variational Autoencoder (VAE), Information Gain (IG) and Convolutional Neural Network (CNN), is compared with four machine learning algorithms (Random Forest, Decision Tree, K-Nearest Neighbour, Gaussian Neff Bayes). Fivefold cross-validation was used to evaluate and calibrate the model. Precision, recall, accuracy, F1 score and area under the curve (AUC) were calculated to assess model performance. Results Our cohort comprises 118 patients diagnosed with 1p/19q codeletion and 100 patients diagnosed with 1p/19q non-codeletion. The study findings indicate that the ECNN method demonstrates excellent predictive performance on the validation dataset. Our model achieved an average precision of 0.981, average recall of 0.980, average F1-score of 0.981, and average accuracy of 0.981. The average area under the curve (AUC) for our model is 0.994, surpassing that of the other four traditional machine learning algorithms (AUC: 0.523–0.702). This suggests that the model based on the ECNN algorithm performs well in distinguishing the 1p/19q molecular status of LGG patients. Conclusion The deep learning model based on conventional MRI radiomic integrates VAE and IG methods. Compared with traditional machine learning algorithms, it shows the best performance in the prediction of 1p/19q molecular co-deletion status. It may become a potentially effective tool for non-invasively and effectively identifying molecular features of lower-grade glioma in the future, providing an important reference for clinicians to formulate individualized diagnosis and treatment plans.https://doi.org/10.1186/s12885-025-14454-9Adult gliomaMagnetic resonance imagingRadiomicsDeep learning1p/19q co-deletion |
| spellingShingle | Fengda Li Zeyi Li Hong Xu Gang Kong Ze Zhang Kaiyuan Cheng Longyuan Gu Lei Hua Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics BMC Cancer Adult glioma Magnetic resonance imaging Radiomics Deep learning 1p/19q co-deletion |
| title | Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics |
| title_full | Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics |
| title_fullStr | Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics |
| title_full_unstemmed | Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics |
| title_short | Prediction of 1p/19q state in glioma by integrated deep learning method based on MRI radiomics |
| title_sort | prediction of 1p 19q state in glioma by integrated deep learning method based on mri radiomics |
| topic | Adult glioma Magnetic resonance imaging Radiomics Deep learning 1p/19q co-deletion |
| url | https://doi.org/10.1186/s12885-025-14454-9 |
| work_keys_str_mv | AT fengdali predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT zeyili predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT hongxu predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT gangkong predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT zezhang predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT kaiyuancheng predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT longyuangu predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics AT leihua predictionof1p19qstateingliomabyintegrateddeeplearningmethodbasedonmriradiomics |