Spyridon (Spyros) Bakas, Ph.D.
Assistant Professor (Tenure Track)
Dept. of Pathology & Laboratory Medicine
Dept. of Radiology (joint appt)
Dept. of Bioengineering (secondary appt)
Perelman School of Medicine
University of Pennsylvania
Medical Research Analytics for Quantitative Integration (MeRAQI) lab - Director
Artificial Intelligence in Biomedical Imaging Lab (AIBIL)
Center for AI & Data Science for Integrated Diagnostics (AI2D) - Executive Faculty Committee
Center for Biomedical Image Computing & Analytics (CBICA)
Google Scholar | LinkedIn | Twitter | GitHub
Richards Medical Research Labs, Fl7, St.A702
3700 Hamilton Walk
Philadelphia, PA 19104
Email
Research Summary
My research interest focuses on the development, application, and benchmarking of advanced 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 involved in radio-patho-genomic research where correlations between quantitative imaging features and molecular characteristics can lead to highly accurate imaging biomarkers, towards enabling treatment selection models customized on an individual patient basis. I have also been leading federated machine learning efforts in healthcare towards facilitating expedited multi-institutional studies, while patient data are always retained within the acquiring institution. My work so far has spanned across the areas of image segmentation, feature extraction, statistical analysis, motion analysis, and machine learning techniques applied in magnetic-resonance (MR), digitized histopathology, and contrast-enhanced ultrasound (CEUS), imaging data. The ultimate aim of my research is clinical deployment, towards making diagnostic and treatment decisions more promptly, objectively, and precisely.
(for a complete up to date list of publications, please visit my Google scholar page)
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)
- The Federated Tumor Segmentation (FeTS) platform: An intuitive tool facilitating secure multi-institutional collaboration (U01CA242871, NIH/NCI/ITCR, PI: Bakas, Spyridon, 07/01/2019 - 06/30/2023)
- Populating the Penn Immune Health Report to Support Precision Medicine: Immunohistochemical Markers to Predict Response to Checkpoint Blockade in Non-Small Cell Lung Cancer (NSCLC) (Abramson Cancer Center, PI: Thompson, Jeffrey (Role: Co-I), 04/22/2020 - 04/21/2021)
- A pilot radiogenomics study evaluating MRI signatures of therapeutically targetable gene expression alterations in human glioblastoma (CTSA/ITMAT/TBIC, PIs: Bakas, Spyridon / Bagley, Stephen, 03/01/2019 - 02/28/2021)
- Democratization of AI-based Lesion Segmentation (Intramural CBICA seed grant, PI: Bakas, Spyridon, 02/03/2020 - 02/02/2021)
- Cancer imaging phenomics software suite: application to brain and breast cancer (1U24CA189523, NIH/NCI/ITCR, PI: Davatzikos, Christos, 09/01/2015 - 08/31/2020)
- Predicting brain tumor progression via multiparametric image analysis and modeling (R01NS042645, NIH/NINDS, PI: Davatzikos, Christos, 09/01/2014 - 05/31/2020)
- Refined Personalized Radiotherapy Target Volume Definition using Predictive Recurrence maps in Glioblastoma (Intramural CBICA seed grant, PI: Bakas, Spyridon, 06/01/2018 - 05/31/2019)
- In vivo surrogate markers of clinically-relevant molecular characteristics of glioblastoma, based on multivariate machine learning and clinically-acquired MRI (CTSA/ITMAT/TBIC, PIs: Bakas, Spyridon / Davatzikos, Christos, 02/01/2017 - 01/31/2019)
- In vivo predictive models of meningioma progression, (CTSA/ITMAT/TBIC, PIs: Dahmane, Nadia / Grady, Sean / Davatzikos, Christos, 02/01/2017 - 01/31/2019)
- Collaborative Funding
- Imaging Signatures of Genetic Mutations in Glioblastoma Using Machine Learning (R01NS042645, NIH/NINDS, PI: Davatzikos, Christos (Role: Co-I), 12/15/2019 - 11/30/2024)
-
Generalizable quantitative imaging and machine learning signatures in glioblastoma, for precision diagnostics and personalized treatment: the ReSPOND consortium (R01CA269948, NIH/NCI, PI: Davatzikos, Christos (Role: Co-I), 06/15/2022 - 05/15/2027)
- Plenary Keynote @ Bio-IT World Conference & Expo (May 17): "Federated Futures: How the Largest Federated Learning Effort in Medicine Will Inform Our Next Steps"
- Invited talk @ ASNR'23 Annual Meeting (May 3): "Big Data and Brain Tumors"
- Invited lecture @ University of Verona (Dept of CS): "Federated Learning and Clinical Deployment"
- Invited talk @ North American Imaging in MS Cooperative (NAIMS): "The Federated Tumor Segmentation (FeTS) initiative"
- Invited talk @ Dagstuhl Seminar 23022 "Inverse Biophysical Modeling and Machine Learning in Personalized Oncology"
- Nature Communications: New publication on the Largest Real-world Federated Learning study to-date, involving >6,300 cases from 71 healthcare sites, to quantify the rare cancer of glioblastoma
- CellPress Patterns: Co-organizing special issue: "Federated Learning and Digital Healthcare: New Discoveries, Advances, and Applications"
- RSNA 2022: Invited talk on: "Translational AI Science: Bringing Advances in Deep Learning into Clinical Practice"
- SNO 2022:
- 9 accepted abstracts & 3 thematic sessions - Congratulations team!
- Invited Young Investigator Mentor
- Digital Pathology and AI Workshop: Invited Talk on "Federated Learning and Clinical Deployment"
- Invited talk @ NIH’s Cancer Imaging Program staff meeting: The Federated Tumor Segmentation (FeTS) tool: An intuitive tool facilitating secure multi-site collaboration
- NeurIPS 2022: Program Committee of the Eye Gaze meets Machine Learning workshop
- Physics in Medicine & Biology: New publication on The Open Federated Learning (OpenFL) Library
- Physics in Medicine & Biology: New publication on The Federated Tumor Segmentation (FeTS) tool
- MICCAI 2022:
- Leading the organization of the:
- Co-organizing the:
- Workshop on Brain Lesions (BrainLes).
- Workshop on Distributed, Collaborative, and Federated Learning (DeCaF).
- Challenge on Cross-Modality Domain Adaptation for Medical Image Segmentation (CrossMoDA).
- Co-organizing the 1st Int. Neuro-Imaging, Neuro-Oncology, Neuro-Science (N3S2) summer school
- Machine Learning in Medical Imaging Consortium (MaLMIC) forum 2022: Invited talk on the Federated Tumor Segmentation (FeTS) initiative
- Nature Scientific Data: New paper releasing The UPenn-GBM data collection
- SNO-RANO 2022 Webinar: Invited talk: “AI-RANO: AI for Response Assessment in Neuro-Oncology"
- Histopathological Image Analysis (HIMA) 2021 workshop: Invited lecture on "How can we do it!" (link to course info)
- ASCO 2021 - Mid-Year RANO meeting: Invited talk: “AI-RANO: Federated Learning as a novel paradigm shift in multi-institutional collaborations"
- ISMRM 2021: Invited talk: “AI & NeuroOncology: Are we there yet?"
- NIH’s Quantitative Imaging Network Meeting: Invited talk: “The Cancer Imaging Phenomics Toolkit (CaPTk): An easy-to-use tool and a computational library for quantitative medical image analysis”
- SNO 2021: Invited Keynote: "Imaging Genomics Reveals Features Associated with Molecular Subtypes of Gliomas."
- MICCAI 2021:
- Member of the MICCAI 2021 Organizing Committee
- Leading the organization of the:
- RSNA/ASNR/MICCAI Brain Tumor Segmentation (BraTS) challenge 2021.
- 1st Federated Learning Challenge 2021 for Tumor Segmentation.
- Co-organizing the:
- Workshop on Brain Lesions (BrainLes).
- Workshop on Distributed and Collaborative Learning (DCL).
- Challenge on Quantification of Uncertainties in Biomedical Image Quantification (QUBIQ).
- Challenge on Cross-Modality Domain Adaptation for Medical Image Segmentation (CrossMoDA).
- EBM 2021 (Evidence Based Management of Cancers) conference: Invited Talk on: "AI in Clinical Practice - Challenges and Strategies to Make it Reliable"
- New Publication: Accurate and Robust Alignment of Differently Stained Histologic Images Based on Greedy Diffeomorphic Registration
- Nature Scientific Reports: New paper on a Federated Learning Simulation Study for Brain Tumor Segmentation.
- MED-NeurIPS 2020: Invited Keynote on "The Federated Tumor Segmentation (FeTS) Initiative"
- SNO 2020:
- Organizer and Chair of the "Artificial Intelligence and Neuro-Oncology Imaging" sunrise session (Postponed for SNO 2021 due to COVID-19)
- Medical Physics: New multi-institutional study on Reproducibility of Expert Annotations and Radiomic Features on the IvyGAP dataset.
- RSNA 2020: Invited talk on: "Identifying the Best Machine Learning Algorithms for Tumor Segmentation, Progression Assessment, and Overall Survival Prediction."
- iGLASS: I am honored to be invited to lead the Imaging Working Group of the Glioma Longitudinal AnalySiS (GLASS) consortium (position paper).
- NeuroImage: New paper on a Multi-institutional Evaluation of Deep Learning Methods for Brain Extraction on Brain Glioma MRI Scans, and a Robust Modality-agnostic Training Approach.
- Annual Review of Biomedical Engineering: New paper on Integrated Biophysical Tumor Growth Modeling and Image Analysis for Neuro-Oncology.
- Journal of Medical Imaging: New paper on Predicting GBM Patient Overall Survival Using Structural MRI.
- npj Digital Medicine: New (community/position) paper on The Future of Digital Health with Federated Learning.
- Tomography: New paper on Radiomic Feature Standardization on CT Digital Reference Object and Patient Data.
- MICCAI 2020:
- Member of the MICCAI 2020 Organizing Committee
- Leading the International Brain Tumor Segmentation (BraTS) challenge 2020.
- Co-organizing the:
- MICCAI: I am honored to be invited to join the MICCAI Board Challenge Working Group.
- NIH Funding: My U01 grant proposal on Federated Learning for Tumor Segmentation was awarded by NCI's ITCR program.
- SNO 2019: Organizing the 2nd 'Computational Neuro-Oncology' session in the Society for Neuro-Oncology Meeting.
- Frontiers in Computational Neuroscience: New paper on MRI Signatures of Transcriptomic GBM Subtypes.
- Nature Biomedical Engineering: New paper (perspective) on Machine Learning for Histopathology.
- Journal of Magnetic Resonance Imaging: New Review Article on Radiogenomics of Brain Tumors.
- MICCAI 2019:
- Leading the organization of the:
- Co-organizing the:
- Full-day Workshop on Brain Lesions (BrainLes) 2019.
- NIH Computational Precision Medicine Challenge.
- ISBI 2019: Our team won the 2nd rank in the Automatic Non-rigid Histological Image Registration challenge.
- Elsevier's Biomedical Signal Processing and Control: New paper on Longitudinal Prediction of Brain Tumor Segmentation.
- AACR 2019: Invited talk: "Cancer Imaging: Separating the Hype from the Hope in Radiomics and Machine Learning".
- Magnetic Resonance Imaging: New paper on Machine Learning MRI signatures for Precision Diagnostics.
- Ultrasound in Medicine & Biology: New paper on Evaluation of Indirect Motion Compensation Methods in CEUS.(Apr'19)
- Cancer Cell: New paper on EGFR ECD mutations in GBM presenting opportunities for therapeutic development.(09 Jul'18
- Neuro-Oncology: New publication on in vivo evaluation of EGFRvIII in glioblastoma via multiparametric MRI. (30 Mar'18)
- SNO 2018:
- Invited to moderate the Neuro-Imaging session in the Society for Neuro-Oncology Meeting, on Nov15.
- Organizing the 'Computational Neuro-Oncology' session in the Society for Neuro-Oncology Meeting, on Nov17.
- MICCAI 2018:
- Invited to act as Associate Program Chair for MICCAI 2018.
- Leading the organization of the International Brain Tumor Segmentation (BraTS) challenge 2018, on Sep16.
- Leading the organization the Full-day tutorial on Tools Allowing Clinical Translation of Image Computing ALgorithms (TACTICAL) 2018, on Sep20.
- Co-organizing the Full-day Workshop on Brain Lesions (BrainLes) 2018 and the, on Sep16.
- Co-organizing the Workshop and Challenges in Computational Precision Medicine, on Sep16.
- Co-organizing the Medical Segmentation Decathlon challenge, on Sep20.
- ISBI 2018: Co-organizing the half-day hands-on Tutorial on the Cancer Imaging Phenomics Toolkit (CaPTk).
- Nature Scientific Data: Our publication on providing expert segmentations and radiomic features for the TCGA-GBM and TCGA-LGG MRI collections is featured on the journal's front page. (5 Sep'17)
- 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:
- Leading the organisation of the International Brain Tumor Segmentation (BraTS) Challenge 2017.
- 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)
- JAMA Pediatrics: Our work on predicting the need for intervention in fetal ventriculomegaly initiated an Editorial. (Feb'17)