The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma
Purpose: This study aimed to evaluate the predictability of survival in patients with glioblastoma using a machine learning (ML) model developed with tissue analysis features obtained through preoperative post-contrast T1-weighted images(T1WI). Materials and Methods: The radiomic features of tumors...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cukurova University
2021-06-01
|
| Series: | Cukurova Medical Journal |
| Subjects: | |
| Online Access: | https://dergipark.org.tr/tr/download/article-file/1667809 |
| _version_ | 1797920525942521856 |
|---|---|
| author | Bozkurt Gülek Berna Bozkurt Duman Emre Bilgin Emin Demırel Okan Dılek |
| author_facet | Bozkurt Gülek Berna Bozkurt Duman Emre Bilgin Emin Demırel Okan Dılek |
| author_sort | Bozkurt Gülek |
| collection | DOAJ |
| description | Purpose: This study aimed to evaluate the predictability of survival in patients with glioblastoma using a machine learning (ML) model developed with tissue analysis features obtained through preoperative post-contrast T1-weighted images(T1WI).
Materials and Methods: The radiomic features of tumors were obtained from postcontrast T1WI of 60 glioblastoma patients. Radiomic properties, density, shape, and textural properties obtained from six matrices were included in the analysis. The patients' three- and six-month survival rates were recorded. Five different ML algorithms were applied to create predictive models [random forest, neural network, linear discriminant analysis(LDA), stochastic gradient descent (SGD), and support vector machine(SMV)].
Results: The mean survival time of the patients was 295.4 days, and the median value was 211.5 (17-1357) days. Among the models developed for three- and six-month survival prediction, the highest success was obtained from the LDA algorithm, in which the AUC values were calculated as 0.88 and 0.78, respectively.
Conclusion: Using ML techniques, the success of predicting imaging-based patient survival was very high. With the development and widespread adoption of these techniques, ML models will be useful in deciding on treatment according to survival prediction in glioblastoma. |
| first_indexed | 2024-04-10T14:02:24Z |
| format | Article |
| id | doaj.art-ad767af980f94a4a95e9d9c4c698624c |
| institution | Directory Open Access Journal |
| issn | 2602-3040 |
| language | English |
| last_indexed | 2024-04-10T14:02:24Z |
| publishDate | 2021-06-01 |
| publisher | Cukurova University |
| record_format | Article |
| series | Cukurova Medical Journal |
| spelling | doaj.art-ad767af980f94a4a95e9d9c4c698624c2023-02-15T16:10:11ZengCukurova UniversityCukurova Medical Journal2602-30402021-06-0146270671310.17826/cumj.90468848The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastomaBozkurt Gülek0Berna Bozkurt Duman1Emre Bilgin2Emin Demırel3Okan Dılek4Adana Şehir Eğitim ve Araştırma HastanesiAdana Şehir Eğitim ve Araştırma HastanesiAdana Şehir Eğitim ve Araştırma HastanesiEMİRDAĞ DEVLET HASTANESİADANA ŞEHİR EĞİTİM VE ARAŞTIRMA HASTANESİPurpose: This study aimed to evaluate the predictability of survival in patients with glioblastoma using a machine learning (ML) model developed with tissue analysis features obtained through preoperative post-contrast T1-weighted images(T1WI). Materials and Methods: The radiomic features of tumors were obtained from postcontrast T1WI of 60 glioblastoma patients. Radiomic properties, density, shape, and textural properties obtained from six matrices were included in the analysis. The patients' three- and six-month survival rates were recorded. Five different ML algorithms were applied to create predictive models [random forest, neural network, linear discriminant analysis(LDA), stochastic gradient descent (SGD), and support vector machine(SMV)]. Results: The mean survival time of the patients was 295.4 days, and the median value was 211.5 (17-1357) days. Among the models developed for three- and six-month survival prediction, the highest success was obtained from the LDA algorithm, in which the AUC values were calculated as 0.88 and 0.78, respectively. Conclusion: Using ML techniques, the success of predicting imaging-based patient survival was very high. With the development and widespread adoption of these techniques, ML models will be useful in deciding on treatment according to survival prediction in glioblastoma.https://dergipark.org.tr/tr/download/article-file/1667809glioblastomamachine learningmriglioblastommakine öğrenmesitexturemrg |
| spellingShingle | Bozkurt Gülek Berna Bozkurt Duman Emre Bilgin Emin Demırel Okan Dılek The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma Cukurova Medical Journal glioblastoma machine learning mri glioblastom makine öğrenmesi texture mrg |
| title | The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma |
| title_full | The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma |
| title_fullStr | The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma |
| title_full_unstemmed | The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma |
| title_short | The success of machine learning algorithms developed with radiomic features obtained from preoperative contrast-enhanced MRI in the prediction of short-term survival in patients with glioblastoma |
| title_sort | success of machine learning algorithms developed with radiomic features obtained from preoperative contrast enhanced mri in the prediction of short term survival in patients with glioblastoma |
| topic | glioblastoma machine learning mri glioblastom makine öğrenmesi texture mrg |
| url | https://dergipark.org.tr/tr/download/article-file/1667809 |
| work_keys_str_mv | AT bozkurtgulek thesuccessofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT bernabozkurtduman thesuccessofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT emrebilgin thesuccessofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT emindemırel thesuccessofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT okandılek thesuccessofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT bozkurtgulek successofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT bernabozkurtduman successofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT emrebilgin successofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT emindemırel successofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma AT okandılek successofmachinelearningalgorithmsdevelopedwithradiomicfeaturesobtainedfrompreoperativecontrastenhancedmriinthepredictionofshorttermsurvivalinpatientswithglioblastoma |