dc.contributor.author | Blackledge, MD | |
dc.contributor.author | Winfield, JM | |
dc.contributor.author | Miah, A | |
dc.contributor.author | Strauss, D | |
dc.contributor.author | Thway, K | |
dc.contributor.author | Morgan, VA | |
dc.contributor.author | Collins, DJ | |
dc.contributor.author | Koh, D-M | |
dc.contributor.author | Leach, MO | |
dc.contributor.author | Messiou, C | |
dc.date.accessioned | 2019-11-12T11:08:28Z | |
dc.date.issued | 2019-10-10 | |
dc.identifier.citation | Frontiers in oncology, 2019, 9 pp. 941 - ? | |
dc.identifier.issn | 2234-943X | |
dc.identifier.uri | https://repository.icr.ac.uk/handle/internal/3404 | |
dc.identifier.eissn | 2234-943X | |
dc.identifier.doi | 10.3389/fonc.2019.00941 | |
dc.description.abstract | Background: Multi-parametric MRI provides non-invasive methods for response assessment of soft-tissue sarcoma (STS) from non-surgical treatments. However, evaluation of MRI parameters over the whole tumor volume may not reveal the full extent of post-treatment changes as STS tumors are often highly heterogeneous, including cellular tumor, fat, necrosis, and cystic tissue compartments. In this pilot study, we investigate the use of machine-learning approaches to automatically delineate tissue compartments in STS, and use this approach to monitor post-radiotherapy changes. Methods: Eighteen patients with retroperitoneal sarcoma were imaged using multi-parametric MRI; 8/18 received a follow-up imaging study 2-4 weeks after pre-operative radiotherapy. Eight commonly-used supervised machine-learning techniques were optimized for classifying pixels into one of five tissue sub-types using an exhaustive cross-validation approach and expert-defined regions of interest as a gold standard. Final pixel classification was smoothed using a Markov Random Field (MRF) prior distribution on the final machine-learning models. Findings: 5/8 machine-learning techniques demonstrated high median cross-validation accuracies (82.2%, range 80.5-82.5%) with no significant difference between these five methods. One technique was selected (Naïve-Bayes) due to its relatively short training and class-prediction times (median 0.73 and 0.69 ms, respectively on a 3.5 GHz personal machine). When combined with the MRF-prior, this approach was successfully applied in all eight post-radiotherapy imaging studies and provided visualization and quantification of changes to independent STS sub-regions following radiotherapy for heterogeneous response assessment. Interpretation: Supervised machine-learning approaches to tissue classification in multi-parametric MRI of soft-tissue sarcomas provide quantitative evaluation of heterogeneous tissue changes following radiotherapy. | |
dc.format | Electronic-eCollection | |
dc.format.extent | 941 - ? | |
dc.language | eng | |
dc.language.iso | eng | |
dc.publisher | FRONTIERS MEDIA SA | |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0 | |
dc.title | Supervised Machine-Learning Enables Segmentation and Evaluation of Heterogeneous Post-treatment Changes in Multi-Parametric MRI of Soft-Tissue Sarcoma. | |
dc.type | Journal Article | |
dcterms.dateAccepted | 2019-09-06 | |
rioxxterms.versionofrecord | 10.3389/fonc.2019.00941 | |
rioxxterms.licenseref.uri | https://creativecommons.org/licenses/by/4.0 | |
rioxxterms.licenseref.startdate | 2019-01 | |
rioxxterms.type | Journal Article/Review | |
dc.relation.isPartOf | Frontiers in oncology | |
pubs.notes | No embargo | |
pubs.organisational-group | /ICR | |
pubs.organisational-group | /ICR/Primary Group | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions/Radiotherapy and Imaging | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions/Radiotherapy and Imaging/Computational Imaging | |
pubs.organisational-group | /ICR/Primary Group/Royal Marsden Clinical Units | |
pubs.organisational-group | /ICR | |
pubs.organisational-group | /ICR/Primary Group | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions/Radiotherapy and Imaging | |
pubs.organisational-group | /ICR/Primary Group/ICR Divisions/Radiotherapy and Imaging/Computational Imaging | |
pubs.organisational-group | /ICR/Primary Group/Royal Marsden Clinical Units | |
pubs.publication-status | Published | |
pubs.volume | 9 | |
pubs.embargo.terms | No embargo | |
icr.researchteam | Computational Imaging | |
dc.contributor.icrauthor | Blackledge, Matthew | |
dc.contributor.icrauthor | Collins, David | |