Roger Zimmermann

Director of MMRL Lab

Assosciate Professor at NUS, Singapore

Biography


Generic placeholder image

Roger Zimmermann is an Associate Professor with the Computer Science Department at the National University of Singapore (NUS). He is also a deputy director with the Interactive and Digital Media Institute (IDMI) at NUS and co-director of the Centre of Social Media Innovations for Communities (COSMIC). He holds a Ph.D. and an M.S. degree in Computer Science from the University of Southern California (USC). Among his research interests are mobile video management, streaming media architectures, distributed systems, spatio-temporal data management and location-based services. He has co-authored seven patents and more than two-hundred peer-reviewed articles in the aforementioned areas. He received the Best Paper Awards at the IEEE International Symposium on Multimedia (ISM) 2012 and the ACM SIGSPATIAL International Workshop on GeoStreaming (IWGS) 2016. Roger is on the editorial boards of the IEEE Multimedia Communications Technical Committee (MMTC) R-Letter and the Springer International Journal of Multimedia Tools and Applications (MTAP). He is also an associate editor for the ACM Transactions on Multimedia journal (ACM TOMM) and he has been elected to serve as Secretary of ACM SIGSPATIAL for the term 1 July 2014 to 30 June 2017. He has served on the conference program committees of many leading conferences and as reviewer of many journals. Recently he was the general chair of the ACM Multimedia Systems 2014 and the IEEE ISM 2015 conferences, and TPC co-chair of the ACM TVX 2017 conference.

Abstract


Machine Learning for Image Processing in Healthcare


Many aspects of healthcare are undergoing rapid evolution and facing many challenges. Computer vision and image processing methods have progressed tremendously within the last few years. One of the reasons is the excellent performance that machine learning algorithms are achieving in many the fields of image processing, especially through deep learning techniques. There exists various application areas where computer-based image classification and object detection methods can make meaningful contributions. Yet, these data-intensive methods encounter a unique set of challenges in the medical domain – which often suffer from a scarcity of large public datasets and still require reliable analysis with high precision. This talk will present some recent work in the area of image analytics for cervical cancer screening in the context of low resource settings. The work is in collaboration with Dr. Pamela Tan from Singapore’s KK Hospital and MobileODT, a medical device and software-enabled services company. In this joint project, our group’s work focuses on machine learning algorithms for the medical analysis of cervix images acquired via unconventional consumer imaging devices like smartphones, based on their appearance and for the purpose of screening cervical cancer precursor lesions. The talk will present our methodology and some preliminary results.


Norbert Noury

Full Professor, University Lyon / INL CNRS 5270, Lyon, France

Biography


Generic placeholder image

Norbert Noury is an expert in the field of "smart health sensors", Ambient Assisted Living environments and Ubiquitous Health monitoring systems. He received the MSC Electronics (1985) and PhD Applied Physics (1992) from Grenoble University. After 8 years in various industrial companies (1985-93), he joined the Grenoble University (1993) where he initiated a new research activity on Health Smart Homes and wearable health sensors. Norbert Noury is a full Professor at the University of Lyon (2008). He guided 20 PhD students, authored over 200 scientific papers (H index 29), holds 15 patents in the field of Biomedical Engineering and is a recognized expert at the European Commission. He is also frequently involved in the organization committees of international scientific events (IEEE-EMBC2007, IEEE-Healthcom2010, pHealth2011, member steering Committee of the IEEE-Healthcom,…).

Abstract


Health Smart Homes...and Beyond


The main thesis in this presentation is that it is possible to allow for a more optimistic way of aging with the use of information technologies as “enablers”. With the collection of various information on the field, the use of data fusion mechanisms to produce higher level information on the trends (trajectory) of Health, with the production of adapted feedbacks which will motivate and accompany the person (coaching). This should also address the two other types of actors in home, that are the professionals and the informal carers. Therefore, the introduction of technologies in the personal home, and in the professional activities, must be questioned in terms of acceptations to avoid the risk of rejection. But the ICT are prone to be intrusive and will modify the relationships and interactions to others and to places. Therefore we must envisage the impact of the introduction of ICT in home, that is to select tools and metrics. The major source of information we can collect at home comes from the monitoring of the activities of the subject in terms of intensity and variety. Again we need some metrics. What do we need to measure and what do we do with the collected information? As a conclusion we want to consider Health Smart Homes not only as a problem to solve in itself, but furthermore as a more general paradigm of the more and more complex problems humans will have to solve in the near future. Obviously, more and more problems are a complex mix of different points of views from science, technology, usages, ergonomics, and sociology. From the ergonomic point of view we must be more prospective .What we learned is that we need to work together, to share our knowledges, to expand our own understanding, to be more integrative, also more flexibles and modests.


Timo Jamsa

Professor in Medical Technology, Faculty of Medecine, University of Oulu, Finland

President Elec of the European Alliance for Medical and Biological Engineering & Science (EAMBES)

Biography


Generic placeholder image

Dr Timo Jämsä, born in 1958, holds a Professorship in Medical Technology at the Faculty of Medicine, University of Oulu, since 2002. He received his MSc in Electrical Engineering from the Faculty of Technology, University of Oulu, and PhD from the Faculty of Medicine, University of Oulu, respectively. He has published over 140 original research articles in biomedical engineering and medical technologies. He has been involved in numerous research projects funded e.g. by EU, Academy of Finland and the Finnish Funding Agency for Innovation TEKES. His main research interests are musculoskeletal biomechanics, eHealth solutions, and technology applications for health promotion and independent living of aging people. The research highlights of his group include biomechanical risk factors of hip fractures, monitoring of health exercise, and technologies for fall detection. He has a number of regional, national and international academic activities. Currently he is President Elect of the European Alliance for Medical and Biological Engineering & Science (EAMBES). He is also member of the Administrative Council of the International Federation of Medical and Biological Engineering (IFMBE).

Abstract


Interpretation of Activity Data Collected with Accelerometers


Interpretation of activity data collected with accelerometers strongly depends on the purpose and health outcome in question and there is no single solution. Instead, a wide range of activity parameters have been presented, e.g. sensor location, sampling rate, acceleration range, pre-processing, analysis algorithm, and selected thresholds define the outcome. This talk overviews some approaches and algorithms for recording physical activity and sedentary behavior, applying data collected with an accelerometer on the wrist and waist in daily life.


Jutta Treviranus

Director of the Inclusive Design Research Centre (IDRC), Toronto, Canada

Professor in the faculty of Design at OCAD University, Toronto, Canada

Biography


Generic placeholder image

Jutta Treviranus is the Director of the Inclusive Design Research Centre (IDRC) and professor in the faculty of Design at OCAD University in Toronto. Jutta established the IDRC in 1993 (formerly the Adaptive Technology Resource Centre) as a center of expertise that proactively promotes the inclusive design of information and communication technologies, practices and policies. Jutta also heads the Inclusive Design Institute, a multi-university regional centre of expertise. Jutta founded an innovative graduate program in inclusive design at OCAD University. Together with Gregg Vanderheiden, Jutta co-directs Raising the Floor International. She leads international multi-partner research networks that have created broadly implemented innovations that support inclusion e.g., Fluid Project, FLOE, and many others. Jutta and her team have pioneered network-supported personalization as an approach to accessibility in the digital domain. She has played a leading role in developing accessibility legislation, standards and specifications internationally (including W3C WAI ATAG, IMS AccessForAll, ISO 24751, and AODA Information and Communication). She has chaired the Authoring Tool Accessibility Guidelines working group as part of the Web Accessibility Initiative of the World Wide Web Consortium. She is a member of numerous advisory boards globally concerned with accessible ICT, including Google, Amazon, Microsoft, NIST, UNDESA, CMHR and others. Jutta’s leadership in Inclusive Design has been recognized through numerous awards, including a Diamond Jubilee Medal and recognition as one of Canada’s top 45 over 45 by Zoomer Magazine. Jutta’s work has been attributed as the impetus for corporate adoption of more inclusive practices in large enterprise companies such as Microsoft.

Abstract


Teaching our Machines to be Smart, Not Prejudiced


Before an intelligent machine can be of help, it has to understand us. There is nothing more frustrating than negotiating with a machine that does not recognize our request, or that misunderstands our intent. Machine learning models and algorithms depend upon data analytics. Data analytics is biased toward dominant patterns, not outliers. People with disabilities and other minorities are outliers. Artificial intelligence has been heralded as a promising technology to assist individuals with disabilities. Intelligent machines have been envisioned as personal assistants, companions and smart environments to remind, prompt, guide, alert to risk and assist with daily functions. More urgently, intelligent machines are making a host of important decisions that affect our lives from predicting loan and credit worthiness, academic potential, terrorist intent, to future employment performance. Before the promise can be fully realized, and the prejudice averted, we must train our machines to be inclusive. This will benefit everyone. Intelligence that understands diversity and stretches to encompass the outliers is better at predicting risk and opportunity, more capable of processing the unexpected, more adaptable, and more dynamically resilient.