Interesting links/news Workshop/conference

Lecture on Use of Dynamic PET Imaging in the Clinic

Lecture at the University of Groningen on May 8, 2019 , entitled: “Does Dynamic PET Imaging have a Future in Clinical Oncologic Practice?”

Interesting links/news

Grand Round Talk on Radiomics and its Relationship to Machine Learning

Here’s a grand round talk delivered to UBC Department of Radiology, entitled: “What is Radiomics? What is Radiogenomics? And What is Their Relationship to Machine Learning and Deep Learning?” (Oct 17, 2018).

Conference works

11 Presentations at 2019 EANM Annual Meeting

Eleven accepted works by our team and collaborators (5 oral; 6 posters) are being presented at the 2019 Annual Congress of the European Association of Nuclear Medicine (EANM), taking place in Barcelona on October 12-16:

  • X. Hou, W. Lv, J-M. Buregaurd, A. Celler, and A. Rahmim
    Dose distribution radiomics: a new paradigm for assessment of radioligand therapy
  • W. Lv, S. Ashrafinia, J. Ma, L. Lu, and A. Rahmim
    Multi-level multi-modality fusion radiomics: application to PET and CT imaging for improved prognostication of head and neck cancer
  • S. Ashrafinia, P. Dalaie, M. S. Sadaghiani, T. H. Schindler, M. G. Pomper, and A. Rahmim
    Standardized radiomics of clinical myocardial perfusion stress SPECT images to determine coronary artery calcification score
  • I. Shiri, P. Ghafarian, P. Geramifar, K. H. Leung, M. Oveisi, A. Rahmim, and M. R. Ay
    Deep direct attenuation correction of brain PET images using emission data and deep convolutional encoder-decoder for application to PET/MR and dedicated brain PET scanners
  • I. Shiri, G. Hajianfar, S. Ashrafinia, E. Jenabi, M. Oveisi, and A. Rahmim
    Radiogenomics analysis of PET/CT images in lung cancer patients: Conventional radiomics versus deep learning
  • R. Ataya, C. F. Uribe, R. Coope, A. Rahmim, F. Bénard
    Variable density 3D-grids for non-uniform activity distributions in PET and SPECT phantoms: a proof of concept
  • Y. Zhu and A. Rahmim
    MR-guided partial volume correction of 3D PET images using a split Bregman optimized parallel level set framework
  • C. Miller, A. Rahmim, and A. Celler
    Dual-isotope peptide receptor radionuclide therapies with 177Lu and 90Y: is quantitative imaging possible?
  • C. F. Uribe, N. Colpo, E. Rousseau, F. Lacroix-Poisson, D. Wilson, A. Rahmim, and F. Bénard
    Regularized reconstruction improves signal-to-noise and quantification for 18F- PSMA PET/CT imaging
  • S. Rezaei, P. Ghafarian, A. K. Jha, A. Rahmim, S. Sarkar, and M. R. Ay
    Joint compensation for motion and partial volume effects in PET/CT images of lung cancer patients: impact on quantification for different image reconstruction methods
  • H. Vosoughi, P. Geramifar, M. Hajizade, F. Emami, A. Rahmim, and M. Momennezhad
    Optimized PET reconstructions: can they be harmonized as well?
Interesting links/news Workshop/conference

An Interview on Radiomics

Here’s a podcast interview at the annual meeting of SNMMI on radiomics:

Conference works

Presentations at 2019 SNMMI Annual Meeting

The published abstracts can now be found here:

Conference works

Presentations at 2019 SNMMI Annual Meeting

Eight accepted works by our group and collaborators (4 oral; 4 posters) are being presented at the 2019 Annual Meeting of the Society of Nuclear Medicine & Molecular Imaging (SNMMI) in Anaheim, June 22-25:

  • K. H. Leung, S. Ashrafinia, M. S. Sadaghiani, P. Dalaie, R. Tulbah, Y. Yin, R. VanDenBerg, J. P. Leal, M. A. Gorin, Y. Du, M. G. Pomper, S. P. Rowe, and A. Rahmim
    A fully automated deep-learning based method for lesion segmentation in 18F-DCFPyL PSMA PET images of patients with prostate cancer
  • Y. Zhu, Y. Gao, O. Rousset, D. F. Wong, and A. Rahmim
    Post-reconstruction MRI-guided enhancement of PET images using parallel level set method with Bregman iteration
  • J. Kim, S. Seo, S. Ashrafinia, A. Rahmim, V. Sossi, and I. S. Klyuzhin
    Training of deep convolutional neural nets to extract radiomic signatures of tumors
  • P. E. Bravo, B. Fuchs, A. K Tahari, D. Pryma, J. Dubroff, and A. Rahmim
    Quantitative renal PET imaging with Rubidium-82 can discriminate individuals with different degrees of renal impairment
  • S. Ashrafinia, M. S. Sadaghiani, P. Dalaie, R. Tulbah, Y. Yin, K. H. Leung, R. VanDenBerg, J. P. Leal, M. A. Gorin, M. G. Pomper, A. Rahmim, and S. P. Rowe
    Characterization of segmented 18F-DCFPyL PET/CT lesions in the context of PSMA-RADS structured reporting
  • I. Shiri, K. H. Leung, P. Ghafarian, P. Geramifar, M. Oveisi, M. R. Ay, and A. Rahmim
    HiResPET: high resolution PET image generation using deep convolution encoder decoder network
  • I. Shiri, K H. Leung, P. Geramifar, P. Ghafarian, M. Oveisi, M. Reza Ay, and A. Rahmim
    PSFNET: ultrafast generation of PSF-modelled-like PET images using deep convolutional neural network
  • I. Shiri, K. H. Leung, P. Ghafarian, P. Geramifar, M. Oveisi, M. R. Ay, and A. Rahmim
    Simultaneous attenuation correction and reconstruction of PET images using deep convolutional encoder decoder networks from emission data


NSERC Discovery Grant Awarded for Quantitative Oncological PET Imaging

We are pleased to announce being awarded an NSERC Discovery Grant. Our proposal (funded for 5 years; $250,000) is entitled, “Quantitative Oncological PET Image Generation and Analysis”. Our aims are to explore: (i) novel data acquisition methods in PET imaging, (ii) advanced 3D and 4D image reconstruction methods for improved image quality and/or dose reduction, integrating advanced models, dynamic as well as motion information; and (iii) advanced radiomics / AI-based image processing towards improved clinical task performance. The grant, aside from its scientific dimensions, emphasizes high-quality training of the next generation of scientist and experts, which is a very important mission of our team.

Interesting links/news Team Members

Visiting PhD Candidate from University of Munich

We are excited to have Julia Brosch visiting us from LMU (Ludwig Maximilian University of Munich). Julia obtained her BSc (physics) in 2014 (thesis: multi-phase tracer kinetics in Lutetium-177-DOTATATE therapy of NETs for optimized kidney dosimetry) and MSc (medical physics) in 2017 (thesis: dosimetric approaches for Yttrium-90 SIRT based on quantitative SPECT and PET images). Julia started her PhD in November 2017, and is part of the research training group GRK2274. Her overall research involves Monte Carlo based dosimetry for radionuclide therapy with focus on Lutetium-177-PSMA therapy. During her stay with us, she aims to learn how to use GATE for simulation of 3D dose distributions for Lu-177-PSMA, and to investigate differences in doses to organs at risk using different dosimetric approaches. We are already very impressed by Julia’s enthusiasm and depth of knowledge.

Awards/Grants Interesting links/news

CIHR Project Grant Awarded for Theranostic Imaging of Prostate Cancer

We are pleased to announce being awarded a CIHR project grant. Our proposal (funded for 4 years; $631,124) is entitled, “Quantitative PSMA Targeted Imaging of Prostate Cancer Patients”. We aim to improve assessment of disease for prostate cancer patients in the context of our ongoing clinical trials involving prostate-specific membrane antigen (PSMA) radioligand therapy (also known as radiopharmaceutical therapy). We will pursue advanced PSMA PET data acquisition (particularly dynamic whole-body imaging), as well as improved image reconstruction and enhancement. Our efforts will also involve automated deep-learning based segmentation of PET images, as well as predictive modeling of prostate cancer using radiomics and machine learning methods.


Presentations at 2018 IEEE Medical Imaging Conference

Eight works by our team and collaborators (2 oral and 6 posters) were recently presented at the 2018 IEEE Medical Imaging Conference in Sydney, November 14-17:

  • M. R. Salmanpour, M. Shamsaee, A. Saberi Manesh, S. Setayeshi, E. Taherinezhad, I. S. Klyuzhin, J. Tang, V. Sossi, and A. Rahmim
    Machine learning methods for optimal prediction of outcome in Parkinson’s disease
  • K. H. Leung, M. R. Salmanpour, A. S. Manesh, I. S. Klyuzhin, V. Sossi, A. K. Jha, M. G. Pomper, Y. Du, and A. Rahmim
    Using deep-learning to predict outcome of patients with Parkinson’s disease
  • Y. Gao, H. Zhang, Y. Zhu, M. Bilgel, O. Rousset, S. Resnick, D. F. Wong, L. Lu, and A. Rahmim
    Voxel-based partial volume correction of amyloid PET images incorporating non-local means regularization
  • I. Shiri, H. Maleki, G. Hajianfar, H. Abdollahi, S. Ashrafinia, M. Ghelich Oghli, M. Oveisi, and A. Rahmim
    PET/CT radiomic sequencer for prediction of EGFR and KRAS mutation status in NSCLC patients
  • M. P. Adams, B. Yang, A. Rahmim, and J. Tang
    Prediction of outcome in Parkinson’s disease patients from DAT SPECT images using a convolutional neural network
  • J. -C. Cheng, C. W. J. Bevington, A. Rahmim, I. S. Klyuzhin, J. Matthews, R. Boellaard, V. and Sossi
    Dynamic PET reconstruction utilizing a spatiotemporal 4D de-noising kernel
  • H. Li, L. Lu, S. Cao, J. Gong, Q. Feng, A. Rahmim, and W. Chen
    Dual-modality joint reconstruction of PET-MRI incorporating a cross-guided prior
  • M. A. Lodge, J. Sunderland, and A. Rahmim
    About measurement of PET spatial resolution