New Webinar: AI and Data Science against COVID-19

The Finnish Center for Artificial Intelligence FCAI organises a webinar series together with Helsinki Centre for Data Science HiDATA. You are warmly welcome!

The webinar series sheds light on how artificial intelligence-based systems and data science could be of help while fighting against COVID-19. Come and listen to the top researchers at University of Helsinki and Aalto University and other related talks in the field of artificial intelligence and data science. The sessions are chaired by FCAI and HiDATA leading professors and the recordings of each webinar are below.

 

Part 1: Monday, 11 May 2020

 

HiDATA and FCAI AI & Data science against Covid-19 webinar series, part 1, 11 May 2020

Join us for a one-hour session chaired by prof. Petri Myllymäki, vice director of FCAI and the following talks:

  • Miika Koskinen: Clinical data repositories, applicability and opportunities for biomedical research

  • Antti Honkela: Privacy-preserving contact statistics collection for COVID-19 epidemic management

Miika Koskinen: Clinical data repositories, applicability and opportunities for biomedical research

Miika Koskinen (D.Sc., Docent) is a senior scientist working as the Head of Analytics at Helsinki Biobank, HUS Helsinki University Hospital. His background is in biomedical engineering, data science, signal processing and neuroscience. The current projects relate to systems and personalized medicine and opportunities of big data analysis and modelling.

Antti Honkela: Privacy-preserving contact statistics collection for COVID-19 epidemic management

Many European countries are planning to deploy contact tracing apps to help COVID-19 epidemic management. Privacy is an absolute requirement for such apps, and typically they would only share a list of anonymous identifiers of recent contacts when someone is diagnosed positive. We propose additional opt-in sharing of numbers of daily contacts of users under strict privacy guarantees from differential privacy. Such data would help epidemic modelling and monitoring the impact of non-pharmaceutical interventions needed for managing the epidemic.

Antti Honkela is Associate Professor of Data Science at the Department of Computer Science at the University of Helsinki. He is the coordinating professor of the Research Programme in Privacy-preserving and Secure AI at the Finnish Center for AI (FCAI). Prior to his current position, he was an Assistant Professor of Statistics with a joint appointment at the Faculty of Science and the Faculty of Medicine at the University of Helsinki. His research interests include machine learning and privacy, with applications especially in computational biology and medicine.

Part 2: Monday, 25 May 2020

Join us for a one-hour session at 2 p.m. chaired by prof. Simo Särkkä, leader of AI across fields in FCAI and the following talks:

  • Otto Seiskari: BLE and its alternatives for contact tracing from the point of view of indoor positioning
  • Alexander Törnroth: Sharing light on the work of the FAIA AI Task force

Otto Seiskari: BLE and its alternatives for contact tracing from the point of view of indoor positioning

Proximity detection - the technical problem of detecting when mobile devices are close to each other - is both the basic building block of many proposed digital contract tracing systems (e.g., Google/Apple, DP-3T, and TraceTogether) and an important concept in indoor positioning. The IndoorAtlas R&D team has a lot of experience on this topic, and this presentation highlights some of the technical and privacy-related challenges that are central for the feasibility of digital contact tracing, but rarely in the focus in public discussion. Link to pdf

Otto Seiskari, D.Sc. (Tech.), is the CTO at IndoorAtlas, a company providing sensor-fusion-based indoor location services to millions of mobile devices around the world. His background is in mathematics, data science, software engineering, and start-ups. Prior to joining IndoorAtlas, he co-founded and worked at Kaiku Health.

Alexander Törnroth: Sharing light on the work of the FAIA AI Task force

FAIA launched and completed an open call for all AI experts and organizations, with the aim to identify and build AI solutions which provides insights and concrete help in the ongoing fight against COVID-19. In his talk Alexander will share a progress update on two different projects: The Pulse, an open for all and free to use dashboard nowcasting the Finnish economy as well as a computer vision tool for physical distancing.

Alexander Törnroth is the lead of FAIA, short for Finland’s AI Accelerator, as well as Ecosystem and Partnership Lead at Silo.AI, the largest private AI Lab in the Nordics.

Part 3: Monday, 8 June 2020

Join us for a one-hour session at 3 p.m. chaired by Patrik Floréen, vice director of HiDATA and the following talks:

  • Pan Hui: Fear or Anticipation? Sentiment Analysis on Social Media during COVID-19 Pandemic
  • Miika Leminen: HUS and data during Covid-19

 

Pan Hui: Fear or Anticipation? Sentiment Analysis on Social Media during COVID-19 Pandemic

Social media have been become an effective platforms to share and discuss events ranging from local to global scale. This information discourse carries latent information that can be used to understand people’s opinions, stances, and sentiments at a given time. Such feedback can be used for multiple purposes including, but not limited to, event forecasting and a basis for designing public policies. In this talk, I will introduce our research on sentiment analysis of social media users during COVID-19 Pandemic. We will show how the lockdown, number of new infected cases, number of new death cases, and recovery cases would impact the sentiment of social media users. This research can help us better understand public reaction to pandemic scenarios, and further assist in designing policies that not only target public safety, but also reassure and comfort members of the public in difficult times.

Pan Hui is the Nokia Chair in Data Science at the University of Helsinki and Director of the HKUST-DT Systems and Media Lab at the Hong Kong University of Science and Technology. His research interest widely covers Data Science, Ubiquitous Computing, Mobile Computing, and Augmented Reality. He is an IEEE Fellow, an ACM Distinguished Scientist, and a member of the Academia Europaea. 

Miika Leminen: HUS and data during Covid-19

Leminen will present some insight about how data have been used during the Covid-19 pandemia in HUS University Hospital.

Miika Leminen has a background in both engineering and psychology, and has previously been working over a decade as a brain research laboratory engineer at University of Helsinki. Currently he works as a development manager in HUS IT department and leads unit called AI and Analytics development services.

Part 4: Monday, 15 June 2020

Join us for a one-hour ses­sion at 2. pm. chaired by Prof. Sasu Tarkoma, dir­ec­tor of HiDATA and the fol­low­ing talks:

Oguzhan Gencoglu: Computer Vision for Crowd Analysis

During pandemics, primary responsibility of risk management is not centralized to a single institution, but distributed across society. Therefore, adequate monitoring of crowds in public (e.g. for social distancing) emerges as a need. This talk will delve into Computer Vision and AI solutions for analyzing crowd behaviour. Real-life examples as well as privacy-preservation will also be discussed.

Oguzhan Gencoglu is the Co-founder and Head of AI at Top Data Science, a Helsinki-based AI startup that provides AI development as a service. With his team, he delivered more than 60 machine learning solutions in numerous industries for the past 4 years. Before that, he used to conduct machine learning research in several countries including USA, Czech Republic, Turkey, Denmark, and Finland.

Indre Zliobaite: Worlds in transition: lessons learned from concept drift research

Concept drift refers to a subdiscipline of machine learning where the world is assumed to be changing over time and predictive models need to continuously adapt in order not to lose accuracy. There are tradeoffs. Immediately after a shock there is not enough of new information to learn about the new state of the world, and yet many of the old rules may not apply anymore. Compromises must be sought selectively blending old and new rules. In this talk I will outline theoretical and empirical considerations from the machine learning perspective of when it is beneficial to adapt immediately after a shock, when to wait and what happens during times of transitions. Parallels with post-pandemic world at this stage can only be speculative.

Indrė Žliobaitė is a tenure track professor at Department of Computer Science, University of Helsinki, where she leads a research group on Data science and evolution. She is also in charge of the global database of fossil mammals, called NOW. Her research is interdisciplinary and has contributed to foundations of fairness-aware machine learning, machine learning with evolving data, as well as evolutionary theory.

 

The Finnish Center for Artificial Intelligence FCAI is a community of experts that brings together top talents in academia with industry and public sector to solve real-life problems using both existing and novel AI. FCAI is one of the six Academy of Finland Finnish flagships. Subscribe to FCAI newsletter.

The Helsinki Centre for Data Science HiDATA is a large multi-disciplinary network of researchers that solves data challenges working on both methods and applications supported by our Data Science infrastructure. HiDATA is a joint hub of the University of Helsinki and Aalto University. Subscribe to HiDATA newsletter.