Perelman School of Medicine at the University of Pennsylvania

Section for Biomedical Image Analysis (SBIA)

Spyridon (Spyros) Bakas, Ph.D.


Postdoctoral Researcher

Section of Biomedical Image Analysis
Center for Biomedical Image Computing & Analytics
Department of Radiology
Perelman School of Medicine
University of Pennsylvania

Richards Medical Research Laboratories, Floor 7
3700 Hamilton Walk
Philadelphia, PA 19104
fax: (001) 215.573.1811
email: S.Bakas@uphs.upenn.edu

News

- World Molecular Imaging Society: Our work was top-ranked and selected for presentation at the Highlight Lecture of WMIC 2017. Title: Non-invasive Molecular and Prognostic Stratification of de novo Glioblastoma Patients through Multivariate Radiomic Analysis of Baseline Preoperative Multimodal Magnetic Resonance Imaging (14 Jul'17)
MICCAI 2017: I am leading the organisation of the International Brain Tumor Segmentation (BraTS) Challenge 2017.
- MICCAI 2017: I am co-organizing the full-day MICCAI workshop on Brain Lesions (BrainLes) 2017, on Sep14. (CfP)
Clinical Cancer Research: New publication on In vivo EGFRvIII detection in glioblastoma via MRI signature. (20 Apr'17)

Educational Qualifications

Ph.D. in Medical Image Computing & Analysis - Kingston University, London (UK)
M.Sc. in Vision, Imaging & Virtual Environments - University College London (UK)
B.Sc. (Hons) in Computer Science - Kingston University, London (UK)

Research Summary

My research interest focuses on the development and application of computational algorithms in oncological imaging, with the intention of improving the assessment, quantification and diagnosis of cancer in the current clinical practice. Eagerly embracing the concept of personalized/precision medicine, I am also involved in radiogenomic research where correlations between quantitative imaging features and genomic information lead to highly accurate imaging biomarkers, which can enable treatment selection models customized on an individual patient basis.

Publications

Journals

  • Spyridon Bakas, Hamed Akbari, Jared Pisapia, Maria Martinez-Lage, Martin Rozycki, Saima Rathore, Nadia Dahmane, Donald M. O’Rourke, Christos Davatzikos. "In vivo detection of EGFRvIII in glioblastoma via perfusion magnetic resonance imaging signature consistent with deep peritumoral infiltration: the φ index", Clin Cancer Res, April 20 2017. [Epub ahead of print] DOI: 10.1158/1078-0432.CCR-16-1871
    (Publisher's version)
  • Spyridon Bakas, Dimitrios Makris, Gordon J.A. Hunter, Cheng Fang, Paul S. Sidhu, Katerina Chatzimichail. "Automatic Identification of the Optimal Reference Frame for Segmentation and Quantification of Focal Liver Lesions in Contrast-Enhanced Ultrasound", Ultrasound in Medicine & Biology, 2017. [In Press] DOI: 10.1016/j.ultrasmedbio.2017.06.005
  • Jared M. Pisapia, Martin Rozycki, Hamed AkbariSpyridon Bakas, Jayesh P. Thawani, Julie S. Moldenhauer, Phillip B. Storm, Deborah M. Zarnow, Christos Davatzikos, Gregory G. Heuer. "Correlations of Atrial Diameter and Fronto-Occipital Horn Ratio with Ventricle Size in Fetal Ventriculomegaly", Journal of Neurosurgery: Pediatrics, 19(3):300-306, 2017. DOI: 10.3171/2016.9.PEDS16210
    (Publisher's version)
  • Spyridon Bakas, Katerina Chatzimichail, Gordon Hunter, Bastien Labbe, Paul S. Sidhu, Dimitrios Makris, "Fast Semi-Automatic Segmentation of Focal Liver Lesions in Contrast-Enhanced Ultrasound, Based on a Probabilistic Model", TCIV Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 5(5):329:338, 2017. DOI: 10.1080/21681163.2015.1029642
    (View Preprint)  —  (Publisher's Version)
  • Spyridon Bakas, Katerina Chatzimichail, Andreas Hoppe, Vasileios Galariotis, Gordon Hunter, Dimitrios Makris, "Histogram-based Motion Segmentation and Characterisation of Focal Liver Lesions in CEUS", Annals of the BMVA (Special Issue), 2012.
    (Published Version)

Close


Conferences

  1. Ke Zeng, Spyridon Bakas, Aristeidis Sotiras, Hamed Akbari, Martin Rozycki, Saima Rathore, Sarthak Pati, Christos Davatzikos. "Segmentation of Gliomas in Pre-Operative and Post-Operative Multimodal Magnetic Resonance Imaging Volumes Based on a Hybrid Generative-Discriminative Framework", Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, Springer, LNCS, 10154:184-194, 2017. DOI: 10.1007/978-3-319-55524-9_18
    (Athens, Greece, 17-21 Oct 2016)
  2. Ke Zeng, Spyridon Bakas, Aristeidis Sotiras, Hamed Akbari, Martin Rozycki, Saima Rathore, Sarthak Pati, Christos Davatzikos. "Segmentation of Gliomas in Pre-Operative and Post-Operative Multimodal Magnetic Resonance Imaging Volumes Based on a Hybrid Generative-Discriminative Framework", In Proceedings of the Multimodal Brain Tumor Image Segmentation Challenge held in conjunction with MICCAI 2016 (MICCAI-BRATS 2016).
    (Athens, Greece, 17-21 Oct 2016)
  3. Spyridon Bakas, Ke Zeng, Aristeidis Sotiras, Saima Rathore, Hamed Akbari, Bilwaj Gaonkar, Martin Rozycki, Sarthak Pati, Christos Davatzikos. "GLISTRboost: Combining Multimodal MRI Segmentation, Registration, and Biophysical Tumor Growth Modeling with Gradient Boosting Machines for Glioma Segmentation", Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, Springer, LNCS, 9556:144-155, 2016. DOI: 10.1007/978-3-319-30858-6_13
    (Technische Universität München (T.U.M.), Munich, Germany, 5-9 Oct 2015)
    (Publisher's Version)
  4. Spyridon Bakas, Ke Zeng, Aristeidis Sotiras, Saima Rathore, Hamed Akbari, Bilwaj Gaonkar, Martin Rozycki, Sarthak Pati, Christos Davatzikos. "Segmentation of Gliomas in Multimodal Magnetic Resonance Imaging Volumes Based on a Hybrid Generative-Discriminative Framework", In Proceedings of the Multimodal Brain Tumor Image Segmentation Challenge held in conjunction with MICCAI 2015 (MICCAI-BRATS 2015).
    (Technische Universität München (T.U.M.), Munich, Germany, 5-9 Oct 2015)
  5. Spyridon Bakas, Katerina Chatzimichail, Paul SidhuDimitrios Makris. "Automatic Identification and Localisation of Potential Malignancies in Contrast-Enhanced Ultrasound Liver Scans Using Spatio-Temporal Features", 6th MICCAI Workshop on Abdominal Imaging - Computational and Clinical Applications, Springer, LNCS, 8676:13-22, 2014. DOI: 10.1007/978-3-319-13692-9_2
    (Massachusetts Institute of Technology (M.I.T.), Boston, U.S.A., 14-18 Sep 2014)
    (View)  —  (Publisher's Version)
  6. Spyridon Bakas, Bastien Labbe, Gordon J.A. HunterPaul Sidhu, Katerina Chatzimichail, Dimitrios Makris. "Fast Segmentation of Focal Liver Lesions in Contrast-Enhanced Ultrasound Data", In Proceedings of the 18th Annual Conference on Medical Image Understanding and Analysis (MIUA), 73-78, 2014. DOI: 10.13140/2.1.1769.0400
    (London, U.K., 9-11 Jul 2014)
    (View)  —  (Publisher's Version)
  7. Spyridon Bakas, Bastien Labbe, Gordon J.A. Hunter. "Making the Best Use of Fifty (or more) Shades of Gray: Intelligent Contrast Optimisation for Image Segmentation in False-Colour Video", 10th International Conference on Intelligent Environments - IE'14, IEEE, 218-221, 2014. DOI: 10.1109/IE.2014.41
    (Shanghai, China, 30 Jun-4 Jul 2014)
    (View)  —  (Publisher's Version)
  8. Spyridon BakasPaul SidhuMaria SellarsGordon J.A. HunterDimitrios Makris, Katerina Chatzimichail. "Non-Invasive Offline Characterisation of Contrast-Enhanced Ultrasound Evaluations of Focal Liver Lesions: Dynamic Assessment Using a New Tracking Method", 20th European Congress of Radiology, European Society of Radiology, 2014. DOI: 10.1594/ecr2014/C-1378
    (Vienna, Austria, 6-10 Mar 2014)
    (View)  —  (Publisher's Version)
  9. Spyridon BakasGordon J.A. Hunter, Celia Thiebaud, Dimitrios Makris. "Spot the Best Frame: Towards Intelligent Automated Selection of the Optimal Frame for Initialisation of Focal Liver Lesion Candidates in Contrast Enhanced Ultrasound Video Sequences",9th International Conference on Intelligent Environments - IE'13, IEEE, 196-203, 2013. DOI: 10.1109/IE.2013.20
    (Athens, Greece, 18-19 Jul 2013)
    (View)  —  (Publisher's version)
  10. Spyridon Bakas, Katerina Chatzimichail, Andreas Hoppe, Vasileios Galariotis, Gordon HunterDimitrios Makris. "Focal Liver Lesion Tracking in CEUS for Characterisation based on Dynamic Behaviour", Springer, Advances in Visual Computing, 8th International Symposium on Visual Computing (ISVC), LNCS, 7431:32-41, 2012. DOI: 10.1007/978-3-642-33179-4_4
    (Crete, Greece, 16-18 July 2012)
    (View)  —  (Publisher's version)
  11. Spyridon Bakas, Katerina Chatzimichail, Awen Autret, Andreas Hoppe, Vasileios Galariotis, Dimitrios Makris. "Localisation and characterisation of focal liver lesions using contrast-enhanced ultrasonographic visual cues", In Proceedings of the 15th Annual Conference on Medical Image Understanding and Analysis (MIUA), 2011. DOI: 10.13140/2.1.1474.1280
    (London, U.K., 14 - 15 July 2011)
    (View)  —  (Publisher's version)

Close


Abstracts

  1. Sarthak Pati, Spyridon Bakas, Aristeidis Sotiras, Ratheesh Kalarot, Patmaa Sridharan, Mark Bergman, Saima Rathore, Hamed Akbari, Paul Yushkevich, Taki Shinohara, Yong Fan, Despina Kontos, Ragini Verma, Christos Davatzikos, "Cancer Imaging Phenomics Toolkit (CaPTk): A Radio(geno)mics Software Platform Leveraging Quantitative Imaging Analytics for Computational Oncology", 2017
    (Accepted in the Quantitative Imaging Reading Room Showcase, in the 103rd Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA), Chicago, Illinois, U.S.A., 26 November – 1 December 2017)
  2. Hamed AkbariSpyridon Bakas, Xiao Da, Donald M. O'RourkeChristos Davatzikos, "Non-invasive Molecular and Prognostic Stratification of de novo Glioblastoma Patients through Multivariate Radiomic Analysis of Baseline Preoperative Multimodal Magnetic Resonance Imaging", 2017
    (Accepted for Oral Presentation in 10th World Molecular Imaging Congress (WMIC), Philadelphia, Pennsylvania, U.S.A., 13 – 16 September 2017)
  3. Gaurav Shukla, Spyridon Bakas, Saima Rathore, Hamed Akbari, Aristeidis Sotiras, Christos Davatzikos, "Radiomic Features From Multi-Institutional Glioblastoma MRI Offer Additive Prognostic Value to Clinical and Genomic Markers: Focus on TCGA-GBM Collection", 2017
    (To be presented in the 59st Annual Meeting of the American Society for Radiation Oncology (ASTRO), San Diego, California, U.S.A., 24-27 September 2017)
  4. Spyridon Bakas, Zev A. Binder, Hamed Akbari, Maria Martinez-Lage, Martin Rozycki, Jennifer J.D. Morrissette, Nadia Dahmane, Donald M. O'Rourke, Christos Davatzikos, "Highly-expressed wild-type EGFR and EGFRvIII mutant glioblastomas have similar MRI signature, consistent with deep peritumoral infiltration", Neuro-Oncology, 18(Suppl.6):VI125-VI126, 2016, DOI: 10.1093/neuonc/now212.523
    (Oral Presentation in the 21st Annual Scientific Meeting of the Society for Neuro-Oncology (SNO), Scottsdale, Arizona, U.S.A., 17-20 November 2016)
  5. Zev A. Binder, Spyridon Bakas, E. Paul Wileyto, Hamed Akbari, Saima Rathore, Martin Rozycki, Jennifer J.D. Morrissette, Maria Martinez-Lage, Nadia Dahmane, Christos DavatzikosDonald M. O'Rourke, "Extracellular EGFR289 activating mutations confer poorer survival and exhibit radiographic signature of enhanced motility in primary glioblastoma", Neuro-Oncology, 18(Suppl.6):VI105-VI106, 2016, DOI: 10.1093/neuonc/now212.441 
    (Presented in the 21st Annual Scientific Meeting of the Society for Neuro-Oncology (SNO), Scottsdale, Arizona, U.S.A., 17-20 November 2016)
  6. Saima Rathore, Hamed Akbari, Martin Rozycki, Spyridon BakasChristos Davatzikos, "Imaging pattern analysis reveals three distinct phenotypic subtypes of GBM with different survival rates", Neuro-Oncology, 18(Suppl.6):VI128, 2016, DOI: 10.1093/neuonc/now212.532 
    (Presented in the 21st Annual Scientific Meeting of the Society for Neuro-Oncology (SNO), Scottsdale, Arizona, U.S.A., 17-20 November 2016)
  7. Jared M. Pisapia, Martin Rozycki, Hamed Akbari, Hannah E. Goldstein, Spyridon Bakas, Julie S. Moldenhauer, Phillip B. Storm, Deborah M. Zarnow, Richard C. E. Anderson, Gregory G. Heuer, Christos Davatzikos, "Fetal Ventriculomegaly: External validity of a fetal MRI-based model to predict the need for post-natal CSF diversion", 2016.
    (Oral Presentation in the 44th Annual Meeting of the International Society for Pediatric Neurosurgery, Kobe, Japan)
  8. Jared M. Pisapia, Hamed Akbari, Spyridon Bakas, Martin Rozycki, Julie S. Moldenhauer, Phillip B. Storm, Deborah M. Zarnow, Gregory G. Heuer, Christos Davatzikos, "Fetal Ventriculomegaly: Predicting Need for Post-Natal Shunt using Machine Learning and Fetal MRI", 2015.
    (Oral Presentation in the 44th Annual Meeting of the AANS/CNS Section on Pediatric Neurological Surgery, Seattle, Washington, U.S.A., 8-11 December 2015)
  9. Jared M. Pisapia, Hamed Akbari, Spyridon Bakas, Martin Rozycki, Julie S. Moldenhauer, Jayesh P. Thawani, Phillip B. Storm, Deborah M. Zarnow, Christos Davatzikos, Gregory G. Heuer. "Fetal Ventriculomegaly: Ventricle Size Correlates Better with Atrial Diameter than with Fronto-Occipital Horn Ratio", 2015.
    (Presented in the 44th Annual Meeting of the AANS/CNS Section on Pediatric Neurological Surgery, Seattle, Washington, U.S.A., 8-11 December 2015)
  10. Hamed Akbari, Spyridon Bakas, Martin Rozycki, Xiao Da, Jared Pisapia, Michel BilelloDonald M. O'RourkeChristos Davatzikos "Non-Invasive Determination of Epidermal Growth Factor Receptor Variant III Expression in Glioblastoma through Analysis of Multi-Parametric Magnetic Resonance Imaging", 2015
    (Oral Presentation in 101st Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA), Chicago, Illinois, U.S.A., 29 November – 4 December 2015)
  11. Spyridon Bakas, Hamed Akbari, Jared Pisapia, Martin Rozycki, Donald M. O'Rourke, Christos Davatzikos, "Identification of Imaging Signatures of the Epidermal Growth Factor Receptor Variant III (EGFRvIII) in Glioblastoma", Neuro-Oncology, 17(Suppl.5):V154, 2015, DOI: 10.1093/neuonc/nov225.05
    (Oral Presentation in the 20th Annual Scientific Meeting of the Society for Neuro-Oncology (SNO), San Antonio, Texas, U.S.A., 19-22 November 2015)
  12. Spyridon BakasChristos Davatzikos, "Predictive Modeling in Brain Cancer Using Machine Learning, Imaging and Genomics: Using Informatics to Obtain Rich Characterizations of Aggressive Brain Tumors, in Support of Precision Medicine", Invited talk in MICCAI NIH Workshop on Computational Precision Medicine II, 2015.
    (Presented in the Technische Universität München (T.U.M.), Munich, Germany, 5-9 Oct 2015)
  13. Spyridon BakasBilwaj GaonkarChristos Davatzikos, "Improved Prediction of Post-surgical Survival in Glioblastoma Through Integrative Genomic and Epigenomic Analysis", The Cancer Genome Atlas' 4th Annual Scientific Symposium, 2015. DOI: 10.13140/RG.2.1.1156.1841
    (Presented in the National Institutes of Health (NIH) Campus, Bethesda, MD, U.S.A., 11-12 May 2015)

Close


Data

  1. Spyridon Bakas, Hamed Akbari, Aristeidis Sotiras, Michel Bilello, Martin Rozycki, Justin Kirby, John Freymann, Keyvan Farahani, Christos Davatzikos. "Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-GBM collection", The Cancer Imaging Archive, 2017.    DOI: 10.7937/K9/TCIA.2017.KLXWJJ1Q
  2. Spyridon Bakas, Hamed Akbari, Aristeidis Sotiras, Michel Bilello, Martin Rozycki, Justin Kirby, John Freymann, Keyvan Farahani, Christos Davatzikos. "Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-LGG collection", The Cancer Imaging Archive, 2017.    DOI: 10.7937/K9/TCIA.2017.GJQ7R0EF

Close


Media

Close


Theses

  • "Computer-Aided Localisation, Segmentation and Quantification of Focal Liver Lesions in Contrast-Enhanced Ultrasound", Kingston University - Faculty of Science, Engineering & Computing (SEC) - Digital Imaging Research Centre (DIRC) - BioImaging Group (BIG)
    (Ph.D. Thesis - 2014)
  • "Discriminative vs. Generative Methods for Face Detection", University College London (UCL) - Department of Computer Science (CS) - Vision, Imaging & Virtual Environments
    (M.Sc. Thesis - 2007)
  • "Three-Dimensional Reconstruction from Two-Dimensional Representations", Kingston University - Faculty of Technology - School of Computing and Information Systems
    (B.Sc. Dissertation - 2005)

Close



Software

  • Peritumoral Heterogeneity Index (PHI) Estimator is a lightweight tool, built towards the following goals:
        - Perform quantitative pattern analysis of the spatial heterogeneity of peritumoral perfusion imaging dynamics, retrieved from Dynamic Susceptibility Contrast Magnetic Resonance Imaging (DSC-MRI) data.
        - Evaluate the imaging biomarker of the Epidermal Growth Factor Receptor variant III (EGFRvIII) mutation status, in individual patients diagnosed with glioblastoma.
     
  • GLISTRboost: Boosted GLioma Image SegmenTation and Registration describes a hybrid generative-discriminative method for segmenting low- and high-grade gliomas in multimodal MRI volumes. Note that this is the winning method of the Multimodal Brain Tumor Image Segmentation (BRATS) Challenge, held in conjunction with the Medical Image Computing and Computer Assisted Intervention (MICCAI) conference in Technische Universitaet Muenchen (TUM) in Munich (Germany) - October 2015
     
  • GLioma Image SegmenTation and Registration (GLISTR) is a software package designed for jointly segmenting multimodal MRI brain scans of glioma patients and registering these scans to a normal, healthy atlas under an Expectation-Maximisation framework..
     
  • Cancer and Phenomics Toolkit (CapTk) showcases some of the highlight applications from the Center for Biomedical Image Computing and Analytics (CBICA) along with advanced visualization and interactive capabilities to make it a complete radiological tool.
    It provides the initialization front-end for:
        - GLISTRboost (The best performing method during BRATS 2015 challenge)
        - GLioma Image SegmenTation and Registration (GLISTR)
        - Pre-Operative and post-Recurrence brain Tumor Registration (PORTR)
    and entails the functionality of:
        - Evaluating the status of the EGFRvIII mutant in pre-operative scans of patients with brain tumors
        - Producing parametric probability maps of infiltration/recurrence of brain tumors
        - Performing a segmentation of brain tumors using a geodesic segmentation
     
  • Brain Tumor Viewer (BTV) is a lightweight viewer, built for fast and simple interaction with MRI image volumes. BTV is primarily designed for visualizing multimodal MRI brain volumes and initializing seed-points for the following software packages:
        - GLioma Image SegmenTation and Registration (GLISTR)
        - Pre-Operative and post-Recurrence brain Tumor Registration (PORTR)
    After initializing these seed-points, a user can generate execution scripts through BTV's graphical user interface for various platforms and custom input parameters.