Computing Research Challenges in Biomedicine

Last June, CRA and he National Institutes of Health jointly hosted a workshop motivated by the following two observations (from the 2004 NIH Roadmap):

The success of computational biology is shown by the fact that computation has become integral and critical to modern biomedical research.

Because computation is integral to biomedical research, its deficiencies have become significant limiters on the rate of progress of biomedical research.

It seems rational to conclude (as the attendees of the workshop concluded) that the productive synergies between the two fields can accelerate research in both, but only if the challenges are addressed through cooperative effort. So, the workshop attendees — leaders in computing and biomedicine, along with NIH Program Directors — aimed to address these challenges by developing a “list of focused recommendations and action items that would guide the NIH and computing communities in addressing current impediments to fully realizing effective collaborations at the interface between computing and biomedical research.” Those recommendations are now available (pdf) as a 14 page report.
The workshop participants ultimately came to agreement on six recommendations, which are listed in some detail in the report but that I’ll attempt to summarize here:

  • Recommendation 1: NIH, the National Science Foundation and Department of Energy Office of Science should support biomedicine and computing research collaborations by:
    • Initiating small, interdisciplinary planning grants that require conceptual proof-of-principal, but minimal or no preliminary results and that involve both computing and biomedical researchers as full partners;
    • creating (or expanding current programs) to fund computing and biomedicine research projects at the PI level, as well as larger collaborative projects with multiple PIs, that reflect the maturation of teams and projects from the small grants above;
    • establishing a cross-disciplinary, multiagency working group to identify, explore and recommend individual agency opportunities and define and coordinate joint agency programs.
  • Recommendation 2: Federal agencies should enhance support for “training at the interface.” These mechanisms would include summer schools for students, post-docs, and professors; increased emphasis on extant undergrad and grad training programs; and funding to transform existing “silo” disciplinary education into new, multidisciplinary structures that support the integration of computing and biomedicine.
  • Recommendation 3: NIH should create a cross-institute software program to create and maintain high-quality, well-engineered biomedical computing software, to assess the quality of existing software, and to create and support for repositories.
  • Recommendation 4: NIH should fund a number of large, distributed transformational centers — distinct from and somewhat orthogonal to the NIH National Centers for Biomedical Computing program — to act as “expeditions to the future.
  • Recommendation 5: NIH should invest in a range of computing research technologies (specified in detail in the report) that are motivated by current and future biomedical research and healthcare needs.
  • Recommendation 6: NIH, NSF, DOE and CRA should create a joint “Interface Task Force” (ITF) — perhaps using the Computing Community Consortium to involve the community — to recommend specific ways to support advances at the interface between computing and biomedicine.

The report includes much more detail for each of the recommendations, including a timeline for implementation and an estimated cost for each. The report also includes more detail on the particular computing research areas the participants thought deserved particular attention.
The whole thing is only 14 pages and is a quick read — well worth it.

  • CRA-NIH Computing Research Challenges in Biomedicine Workshop Recommendations.
    Update: (5/29/07) — Dan Reed has a lot more of the backstory for the report on his blog today.

  • Computing Research Challenges in Biomedicine