Name: Martin Vallières,PhD
Level at MPU: Postdoctoral Fellow
Email: martin.carrier-vallieres [at] mail.mcgill.ca
Website: github.com/mvallieres/radiomics
Supervisor(s): Dr. Jan Seuntjens
Research interests: PET/MR imaging, Texture analysis, Tumour outcome prediction, Radiomics
Research summary
The combination of Positron Emission Tomography (PET) and Magnetic Resonance (MR) imaging is nowadays playing an increasing role in monitoring tumour characteristics that are essential to the efficiency of treatments. In fact, specific imaging features of tumours (e.g. textures) can be extracted to quantify intratumoural heterogeneity and evaluate how patients will respond to treatments. This information is thereafter used to adapt treatments based on patients' predicted failure risk. Our research aims at developing PET/MR image analysis and simulation techniques for optimal prediction of treatment outcomes in soft-tissue sarcomas (e.g. lung metastases development), an aggressive type of cancer that develops in connective tissues of the body. For this purpose, novel image analysis techniques based on the texture features of images are developed to study the information embedded in joint PET/MR images of tumours. PET/MR imaging simulations are then used to investigate the physical factors affecting these features in the acquisition process of PET/MR images. The simulation pipeline allowd to optimize PET/MR imaging acquisition parameters in order to produce images with the optimal texture features needed to provide optimal prediction of tumour outcomes (i.e. optimal assessment of tumour aggressiveness). These tools should impact our understanding of tumour progression and treatment response in soft-tissue sarcoma cancer, potentially improving patient survival. The proposed methodology could afterwards be extended to other types of cancer and help more patients to overcome this deadly disease.
Key publications
- Vallières, M., Laberge, S., Diamant, A., & El Naqa, I. (2017). Enhancement of multimodality texture-based prediction models via optimization of PET and MR image acquisition protocols: a proof of concept. Physics in Medicine and Biology.
- Zhou, H., Vallières, M., Bai, H.X. et al. (2016). MR imaging features predict survival and molecular markers in diffuse lower-grade gliomas. Neuro-Oncology (IN PRESS)
- Vallières, M., Freeman, C.R., Skamene, S.R. et al. (2015). A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities. Phys. Med. Biol., 60(14), 5471-5496. doi:10.1088/0031-9155/60/14/5471
- Hatt, M., Majdoub, M., Vallières, M. et al. (2015). 18F-FDG PET Uptake Characterization Through Texture Analysis: Investigating the Complementary Nature of Heterogeneity and Functional Tumor volume in a Multi-Cancer Site Patient Cohort. J. Nuc. Med., 56(1), 38-44. doi:10.2967/jnumed.114.144055
Awards
(2015) McGill University - Faculty of Medicine
(2012-2015) NSERC Graduate Scholarship Doctoral Award (CGS D)
(2013) AAPM 2013 Science Council Session Award
(2011) FRSQ - Bourse de formation de maîtrise
(2010) NSERC - Graduate Scholarship Master Award (CGS M)