An investigation of machine learning methods in delta-radiomics feature analysis.
dc.contributor.author | Chang, Yushi | |
dc.contributor.author | Lafata, Kyle | |
dc.contributor.author | Sun, Wenzheng | |
dc.contributor.author | Wang, Chunhao | |
dc.contributor.author | Chang, Zheng | |
dc.contributor.author | Kirkpatrick, John P | |
dc.contributor.author | Yin, Fang-Fang | |
dc.contributor.editor | Li, Taoran | |
dc.date.accessioned | 2020-03-24T13:19:02Z | |
dc.date.available | 2020-03-24T13:19:02Z | |
dc.date.issued | 2019-01 | |
dc.date.updated | 2020-03-24T13:18:59Z | |
dc.description.abstract | PURPOSE:This study aimed to investigate the effectiveness of using delta-radiomics to predict overall survival (OS) for patients with recurrent malignant gliomas treated by concurrent stereotactic radiosurgery and bevacizumab, and to investigate the effectiveness of machine learning methods for delta-radiomics feature selection and building classification models. METHODS:The pre-treatment, one-week post-treatment, and two-month post-treatment T1 and T2 fluid-attenuated inversion recovery (FLAIR) MRI were acquired. 61 radiomic features (intensity histogram-based, morphological, and texture features) were extracted from the gross tumor volume in each image. Delta-radiomics were calculated between the pre-treatment and post-treatment features. Univariate Cox regression and 3 multivariate machine learning methods (L1-regularized logistic regression [L1-LR], random forest [RF] or neural networks [NN]) were used to select a reduced number of features, and 7 machine learning methods (L1-LR, L2-LR, RF, NN, kernel support vector machine [KSVM], linear support vector machine [LSVM], or naïve bayes [NB]) was used to build classification models for predicting OS. The performances of the total 21 model combinations built based on single-time-point radiomics (pre-treatment, one-week post-treatment, and two-month post-treatment) and delta-radiomics were evaluated by the area under the receiver operating characteristic curve (AUC). RESULTS:For a small cohort of 12 patients, delta-radiomics resulted in significantly higher AUC than pre-treatment radiomics (p-value<0.01). One-week/two-month delta-features resulted in significantly higher AUC (p-value<0.01) than the one-week/two-month post-treatment features, respectively. 18/21 model combinations were with higher AUC from one-week delta-features than two-month delta-features. With one-week delta-features, RF feature selector + KSVM classifier and RF feature selector + NN classifier showed the highest AUC of 0.889. CONCLUSIONS:The results indicated that delta-features could potentially provide better treatment assessment than single-time-point features. The treatment assessment is substantially affected by the time point for computing the delta-features and the combination of machine learning methods for feature selection and classification. | |
dc.identifier | PONE-D-19-10297 | |
dc.identifier.issn | 1932-6203 | |
dc.identifier.issn | 1932-6203 | |
dc.identifier.uri | ||
dc.language | eng | |
dc.publisher | Public Library of Science (PLoS) | |
dc.relation.ispartof | PloS one | |
dc.relation.isversionof | 10.1371/journal.pone.0226348 | |
dc.title | An investigation of machine learning methods in delta-radiomics feature analysis. | |
dc.type | Journal article | |
duke.contributor.orcid | Kirkpatrick, John P|0000-0002-4019-0350 | |
duke.contributor.orcid | Yin, Fang-Fang|0000-0002-2025-4740|0000-0003-1064-2149 | |
pubs.begin-page | e0226348 | |
pubs.issue | 12 | |
pubs.organisational-group | School of Medicine | |
pubs.organisational-group | Duke Cancer Institute | |
pubs.organisational-group | Radiation Oncology | |
pubs.organisational-group | Duke | |
pubs.organisational-group | Institutes and Centers | |
pubs.organisational-group | Clinical Science Departments | |
pubs.organisational-group | Neurosurgery | |
pubs.organisational-group | Duke Kunshan University Faculty | |
pubs.organisational-group | Duke Kunshan University | |
pubs.organisational-group | Staff | |
pubs.organisational-group | Physics | |
pubs.organisational-group | Trinity College of Arts & Sciences | |
pubs.publication-status | Published | |
pubs.volume | 14 |
Files
Original bundle
- Name:
- An investigation of machine learning methods in delta-radiomics feature analysis.pdf
- Size:
- 1.25 MB
- Format:
- Adobe Portable Document Format