Return to 2012 Open Space/Unconference discussions
Disscussion
Scribe:Louis Gross
Participants: Susan Volman Georgy Vavashev Scott Berceli Marc Garbay Steve Marcus Dalin Tang Mike Bindschlager
In all educational issues it is appropriate to take into account at least 3 components:
Pedagogy
Audience
Content
From Marc Garbey - some projects ongoing at Univ of Houston PUF UHouston and several French institution Joint international masters Vietnam Int Univ and U Houston
Will growth of massive online courses affect education in multiscale modeling - Coursera at Stanford and similar projects at MIT and elsewhere
Distributed teaching as an alternative is to have a customized course with the right students and instructor for that course - for example computational surgery in liver disease. Or in agent-based models of livers, etc. and ensure students have the appropriate interdisciplinary background to benefit from the course. These build a common base of knowledge first that arises specific to the topics important for the project. Then the course proceeds to carry out the group research project. Students can be in different countries. This can be an effective mechanism to introduce students to a research lab before they arrive there. So the pedagogy here is to base the class on what project is most appropriate then go backward to see what is needed to address it and then focus the content on the targeted project.
Systems science in K-12 - what is being done currently in this area - Shodor Foundation and the BioQuest projecthave carried out some activities in this regard. Is there a blueprint on content for systems thinking at different educational levels?.
There are numerous reports on undergrad and graduate education in biology and associated integration with interdisciplinary education: NRC BIO2010 report, HHMI/AAMC report Scientific Foundations for Preparing Future Physicians, Vision and Change report
Is there is resistance among physicians to evidence-based science? Some experience indicates that pre-med students often want to know what research was impacting and not the detailed processes of the research itself. They view it as a professional program. Those really interested in research, aside from the limited exposure in the Med School program, go into MD/PhD programs, which are limited and highly selective. Med School trains people on how to operate in the medical spectrum - it is very skill based, factual and the socialization to be part of the complex health care system. There is some emphasis on how to carry out a diagnosis but cannot learn skills from all the specialties - that's what a residency is for. Oxford handbook on clinical diagnosis is brilliant compilation. Continuous education of MDs is going on.
KO8's track record is not very good for encouraging the awardees to go on to do continuing research. One potential reason is financial in that there is a salary cut associated with research/
Any well-constructed statistical algorithm can probably beat 80-90% of MDs in a diagnosis but the best MDs will always be better than the algorithms
Is it too late by the time they are MDs to encourage the type of research thinking that is needed to inform predictive medicine and the associated modeling? An argument is that no - there are 10% of MDs who came into it from computational sort of backgrounds so they are able to think quantitatively. Where do biomedical engineering programs fit into this?
Conclusions: We should work to create systematic methods and opportunities where people can interact across disciplines and learn the vocabulary and conceptual background for effective communication. There are a variety of NIH-supported programs that do this in some ways. There are five programs in computational neuroscience - Emory and Georgia Tech have one of these. There are NIH Training grants - and training programs in imaging that bring in people to 3 centers and T90/R90 mechanisms. There are additional ways for computational folks to get involved in medical analysis including K25s which can bring mathematicians into biology.
There are multiple potential routes to entice those with strong quantitative background to focus efforts in biology and those with strong biology backgrounds to develop the conceptual quantitative foundations to collaborate effectively with computational scientists. Note that NIH Pre-Doc traineeships from NIGMS specifically require "ensuring that students have the appropriate quantitative training to pursue cutting-edge research and providing opportunities for exposure to topics related to human health, physiology and disease. Describe what the training program does to ensure that students have appropriate quantitative graduate training." It is likely appropriate for IMAG/MSMC to be involved in assisting in deciding how to make this requirement an effective one.
Disscussion
Scribe:
Participants:
Disscussion
Scribe:
Participants: