Sotirios A Tsaftaris, PhD

Chair in Machine Learning and Computer Vision @ University of Edinburgh

My name is Sotirios A. Tsaftaris but most people call me Sotos (in Greek Σωτήριος Α. Τσαυτάρης or Σώτος). I am currently with the University of Edinburgh and the Institute of Digital Communications as a Chair (Full Professor) in Machine Learning and Computer Vision. I hold the Canon Medical/Royal Academy of Engineering Research Chair in Healthcare AI. I am also a Fellow of the Alan Turing Institute.

My research interests are machine learning, deep learning, medical image analysis (medical image computing), image processing, and distributed computing. Core research application domains are computer aided diagnosis in medicine and computer vision in agriculture (e.g. plant phenotyping, precision agriculture, agritech and indoor farming).

Within my web pages, you will find an extended Biography and CV, descriptions of my various research interests as well as a list of publications.

I would like to acknowledge the sponsors and supporters of my research program.

Recent key news

  • A DREAMy MICCAI 2020 Tutorial: I will be giving a tutorial on disentangled representation learning with Alison from Canon Medical at MICCAI 2020 (Peru). See here [DREAM 2020 PAGE]
  • Area chair for CVPR 2021: Will serve as area chair for CVPR 2021 (Nashville, TN).
  • Area chair for MICCAI 2020: Will serve as area chair for MICCAI 2020 (Peru).
  • Invited speaker for AI4AH 2020: I am deeply honoured to give an invited talk at the AI for Affordable Healthcare Workshop in Addis Ababa (Ethiopia) in conjunction with ICLR (a premier machine learning conference).
  • Ilkay became an Assistant Professor: Proud to see an ex-student/postdoc make the next step in his career (Ilkay started at Istanbul Technical University as assistant professor.)
  • Tutorial chair for ECCV 2020: Serve for ECCV 2020 (Glasgow) as Tutorial Chair.
  • Keynote speaker for STACOM 2019: I gave the keynote talk at the best and longest running workshop on cardiac image analysis, the STACOM 2019 at MICCAI (China).
  • Excellent progress of the team: It has been a good 2019 as of now, with 3 papers in MICCAI (2 in workshops), 2 in MIDL, 3 in CVPR (workshop).
  • Valerio became a Lecturer: Proud to see an ex-student/postdoc make the next step in his career (Valerio started at Napier as lecturer.)
  • Paper in MedIA: Agis's excellent work on disentangled learning is finally live in MedIA.
  • Tian shortlisted for best paper award at MIDL 2019: Tian was a runner up for the best paper award at MIDL for his work on pathology factorisation.
  • Paper in Science Translational Medicine for BOLD MRI: After almost 8 years of work, our paper setting the new way to do BOLD in the heart is out in Science Translational Medicine.
  • Recruiting NOW PhD students and Postdocs: I am growing the team. If you have a strong profile please email me and attach a CV.
  • Won prestigious RAEng fellowship: I became a Canon Medical/Royal Academy of Engineering Research Chair in Healthcare AI. Over 5 years this endowed position will provide with the opportunity to look into distinct challenges in using and deploying AI in healthcare settings. It is thanks to the generous support of Canon Medical Research Europe, the Academy and the School of Engineering. [PROJECT PAGE]
  • CVPPP goes to Long Beach: The 5th CVPPP workshop will occur in conjunction with CVPR 2019 the premier Computer Vision Conference. (I am one of the co-organizers.)
  • Open data make impact: Our plant data have made a significant impact in progressing the state of the art. See paper and associated website.
  • £17M for Healthcare and AI: Our Scottish-wide bid for Artificial Intelligence in Healthcare with 17 partners across Scotland has been approved. More forthcoming at the website
  • UDRC 3: The Edinburgh-lead consortium for EPSRC-Dstl University Defence Research Collaboration 3, won funding of £4M. I am leading the workpackage on generative models.
  • Training session at MICCAI 2018: Together with Trucco, Konukoglu and Kainz organised a MICCAI 2018 training session on how to be a good reviewer (Spain, Sept 2018)
  • Associate Editor for IEEE TMI: AE for IEEE Transactions on Medical Imaging. (April 2018)
  • Area Chair for MICCAI 2018: Invited to serve as area chair MICCAI 2018 (Spain, Sept 2018)
  • Keynote at Int. Conf. on Artificial Neural Networks ICANN 2018 (Rhodes, Oct 2018)
  • My PhD student, Ilkay Oksuz, is now a Doctor. Congratulations Ilkay. He did a fantastic job in joint segmentation and registration of cardiac MRI. I am sure he will do great also at King's College London as a postdoc.
  • I am now also a Turing Fellow with the Alan Turing Institute.
  • New NIH (USA) project funded: I have received an NIH grant together with Prof. Dharmakumar to develop a new non-invasive non-contrast infarct MR imaging approach. My team is in charge of the analysis algorithms.
  • Two new UK projects funded: I have received an EPSRC first grant to develop multimodal deep learning methods for cardiac MRI processing [CardiacAI PROJECT PAGE]. I have also received a grant from BBSRC (as part of the Grand Challenges Research Fund) to develop affordable phenotyping of roots with colleagues in Ethiopia. [Chickpea GCRF PROJECT Website]
  • After 5 years in the making, 8 if you count from the start of writing the proposal, finally Phenotiki our affordable plant phenotyping solution is now available.