Login
Talks & Seminars
Title: Knowledge-Based Approaches for Artificial Intelligence in Healthcare
Prof. David Parry, Auckland University of Technology, New Zealand
Date & Time: August 24, 2018 15:30
Venue: Room # 109, 01st Floor, Department of Computer Science and Engineering, New CSE/CC Building
Abstract:
In the last 5 years there has been an explosion in interest in the use of machine learning approaches to produce "Artificial intelligence" (AI). There have been remarkable successes using deep learning approaches - for example learning the rules of Go, and becoming the best player the world, simply by learning from data [1]. However, AI in healthcare does not have to be either machine learning or case- based reasoning. There are often issues with the quality and availability of medical data. Different clinicians may have different opinions in terms of how data should be labelled. Issues of privacy, ownership and security of clinical data can also cause difficulty. Finally, the explanatory capacity of machine learning approaches may also be limited. There are more than 20 million articles available in PUBMED, and there are other large stores of curated knowledge available. Knowledge based Decision-support tools are able to both answer clinical queries but also give auditable explanations for the decisions. Traditionally these tools have required human input to select, rate and format available knowledge, and this has caused the knowledge acquisition "bottleneck". Recent work has looked at using semantic web techniques to find relevant knowledge [2]. This talk will discuss some of the approaches being used, their disadvantages and advantages and possible ways forward in the future. I will discuss some of the work I have being doing along with my PhD student, Jamal Zolhavarieh,[3] to develop automatic methods of identifying high quality knowledge and incorporating this knowledge into decision support systems for clinical applications. Finally I will look into the future and try to imagine an ecosystems of knowledge and data that could be used to support clinical decision making and bring us closer to AI in healthcare. 1. Silver, David, et al. "Mastering the game of Go without human knowledge." Nature 550.7676 (2017): 354. 2. Zolhavarieh, Seyedjamal, David Parry, and Quan Bai. "Issues Associated With the Use of Semantic Web Technology in Knowledge Acquisition for Clinical Decision Support Systems: Systematic Review of the Literature." Ed. Gunther Eysenbach. JMIR Medical Informatics 5.3 (2017): e18. PMC. Web. 26 July 2018. 3. Parry, D., and S. J. Zolhavarieh. "A knowledge quality assessment model for clinical decision support systems." MEDINFO 2017: Precision Healthcare through Informatics Proceedings of the 16th World Congress on Medical and Health Informatics. IOS Press, 2018.
Speaker Profile:
Originally from Wales, I've lived in New Zealand for the last 16 years. I started work as a Medical Physicist and moved across to software development for healthcare and non-healthcare applications. I worked at the Department of Information Science, University of Otago 1997-2000 on the world's first online teaching course in health informatics. From 2000 I've worked at AUT initially as a software developer and researcher in online learning and subsequently as an academic. in 2006 I spent three Months in Bhutan, working on a charity-funded project at the JDW national referral hospital. I also worked at the University of Portsmouth UK on their Telecare knowledge network project in 2006/7. I worked at UNSW in the Centre for Health Systems and Safety Research on sabbatical in2010. Back in New Zealand I became director of the AUT RFID applications laboratory (AURA), and a senior research lecturer above the bar. I was instrumental in developing the HINZ primer series - delivering workshops on health informatics to clinical staff around the country, funded by the National Health IT Board (RIP). I'm the academic leader for the Masters in Health Informatics Programme. I'm passionate about teaching and researching in the area of health IT and I'm also part of the Centre for ehealth.
List of Talks

Webmail

Username:
Password:
Faculty CSE IT
Forgot Password
    [+] Sitemap     Feedback