(Un)seeing things hierarchically!

In his outstanding book “The War of Art”, Steven Pressfield says the following about the artist, which is applicable to anyone who seeks to produce a work of value:

“For the artist to define himself hierarchically is fatal. Let’s examine why. First, let’s look at what happens in a hierarchical orientation. An individual who defines himself by his place in a pecking order will:

1) Compete against all others in the order, seeking to elevate his station by advancing against those above him, while defending his place against those beneath.

2) Evaluate his happiness/success/achievement by his rank within the hierarchy, feeling most satisfied when he’s high and most miserable when he’s low.

3) Act toward others based upon their rank in the hierarchy, to the exclusion of all other factors.

4) Evaluate his every move solely by the effect it produces on others. He will act for others, dress for others, speak for others, think for others.

In the hierarchy, the artist faces outward. Meeting someone new he asks himself, What can this person do for me? How can this person advance my standing?

In the hierarchy, the artist looks up and looks down. The one place he can’t look is that place he must: within.”

Presentations at 2018 SNMMI Annual Meeting

Thirteen accepted works by our group and collaborators (8 oral and 5 posters) are being presented at the 2018 Annual Meeting of the Society of Nuclear Medicine & Molecular Imaging (SNMMI) in Philadelphia June 23-26:


  • A. Rahmim, K. P. Bak-Fredslund, S. Ashrafinia, C. R. Schmidtlein, R. M. Subramaniam, A. Morsing, S. Keiding, J. Horsager, and O. L. Munk
    Quantification of colorectal liver metastases using FDG PET volumetric and heterogeneity features for improved prediction of clinical outcome
  • A. Rahmim, S. Ashrafinia, S. Rowe, C. R. Schmidtlein, M. H. Vendelbo, T. El-Galaly, L. C. Gormsen, and O. L. Munk
    Quantification of lymphoma using FDG PET heterogeneity features for improved prediction of clinical outcome
  • S. Ashrafinia, P. Dalaie, R. Yan, P. Ghazi, C. Marcus, M. Taghipour, P. Huang, M. G. Pomper, T. Schindler, and A. Rahmim
    Radiomics analysis of clinical myocardial perfusion SPECT to predict coronary artery calcification
  • S. Ashrafinia, P. Dalaie, R. Yan, P. Huang, Martin G. Pomper, T. Schindler, and A. Rahmim
    Application of texture and radiomics analysis to clinical myocardial perfusion SPECT imaging
  • H. Leung, W. Marashdeh, S. Ashrafinia, A. Rahmim, M. G. Pomper, and A. K. Jha
    A deep-learning-based fully automated segmentation approach to delineate tumors in FDG PET images of lung cancer patients
  • S. Klyuzhin, N. Shenkov, A. Rahmim, and V. Sossi
    Use of deep convolutional neural networks to predict Parkinson’s disease progression from DaTscan SPECT images
  • D. Du, W. Lv, Q. Yuan, Q. Wang, Q. Feng, W. Chen, A. Rahmim, and L. Lu
    Machine learning methods for optimal differentiation of recurrence versus inflammation from post-therapy nasopharyngeal 18F-FDG PET/CT images
  • X. Hong, W. Lv, Q. Yuan, Q. Wang, Q. Feng, W. Chen, A. Rahmim, and L. Lu
    Prediction of local recurrence and distant metastasis using radiomics analysis of pretreatment nasopharyngeal 18F-FDG PET/CT images
  • Y. Gao, M. Bilgel, S. Ashrafinia, Lijun Lu, Olivier Rousset, Susan Resnick, Dean F. Wong, Arman Rahmim
    Evaluation of non-local methods with and without anatomy information for improved quantitative amyloid PET imaging
  • A. Rahmim, M. A. Lodge, N. A. Karakatsanis, V. Y. Panin, Y. Zhou, A. McMillan, S. Cho, H. Zaidi, M. E. Casey, R. L. Wahl
    Dynamic whole-body PET imaging: principles, potentials and applications
  • W. Lv, Q. Yuan, Q. Wang, J. Ma, Q. Feng, W. Chen, A. Rahmim, and L. Lu
    Prognostic potentials of radiomics analysis on the PET and CT components of PET/CT complementary to clinical parameters in patients with nasopharyngeal carcinoma
  • L. Lu, P. Wang, J. Ma, Q. Feng, A. Rahmim, and W. Chen
    Generalized factor analysis incorporating alpha-divergence and kinetics-based clustering: application to dynamic myocardial perfusion PET imaging
  • Y. Li, A. Rahmim, and L. Lu
    Direct Bayesian parametric image reconstruction from dynamic myocardial perfusion PET data


Our lab is moving to Canada!

After 13 years at Johns Hopkins University, our lab is moving to Vancouver in July! Our lab will be jointly affiliated with the Departments of Radiology and Physics & Astronomy at the University of British Columbia (UBC), as well as the BC Cancer Agency (BCCA). We look forward to playing an active role in concerted and integrated efforts to improve research, education and clinical practice at UBC and BCCA. Hopkins, as an amazing institute and community, has been extremely gracious to us, to which we’re indebted, and we surely hope and plan to pursue significant collaborations and joint efforts.

Special Abstract Citation!

Our abstract was cited in the highlights of the 30th Annual Congress of the European Association of Nuclear Medicine (EANM), Vienna 2017. This was for an unusual reason, though we’ve been doing this for years now!

It reads:

Shiri et al. [46] presented a most unusual and scientifically highly interesting paper. These authors sought to predict lung metastases in patients with soft tissue sarcoma applying advanced machine learning to radiomic features. The unusual aspect, however, was the collaboration between Iranian universities and universities in the US, showing that science is above politics.”


Parametric PET imaging is finally a product!

Finally, there is a vendor product (FlowMotion Multiparametric PET by Siemens) that enables dynamic whole-body PET including parametric imaging. This is very rewarding given that our group was the earliest to propose and work on this framework, including close collaboration with Siemens. This is clearly an enabling technology, and it remains to be seen whether it will add significant value to routine clinical imaging. Hopefully more and more centers will try and explore potential benefits from this technology.

Saeed Ashrafinia awarded the 2018 Bradley-Alavi Student Fellowship

Congratulations to Saeed Ashrafinia who has been awarded the 2018 Society of Nuclear Medicine and Molecular Imaging (SNMMI) Bradley-Alavi Student Fellowship!

Saeed, an Electrical & Computer Engineering PhD candidate in the lab, is actively pursuing research in quantitative PET and SPECT imaging. The awarded fellowship, entitled, “Radiomics Analysis of Clinical Myocardial Perfusion SPECT Images to Identify Subclinical Coronary Artery Disease” proposes to translate radiomics analyses (which has been largely absent in SPECT) to the domain of clinical cardiac imaging.

Bradley-Alavi Fellows are named in honor of the late Stanley E. Bradley, Professor of Medicine at Columbia University College of Physicians and Surgeons and a prominent researcher in the fields of renal physiology and liver disease, and Abass Alavi, M.D., Professor and Director of Research Education at the Department of Radiology at the University of Pennsylvania.