Present: Peter Lucas, George Coghill, René Quiniou, Janet Thomas.
Apologies: Hidde de Jong.

Agenda
1. Welcome
2. Aims of the Biomedical Task Group
3. Domain Status Document (already sent around)
4. MONET Funding and organisation of scientific activities
5. Composition of the task group and establishing contacts
6. Future Task Group meetings
7. AOCB

2. Aims of the Biomedical Task Group
B1 - The BioMedical Domain Status Document is now complete and will be sent to the European Commission with the MONET First Annual Report at the end of this week.
B2 - The BioMedical roadmap version 1. There will be a meeting at the end of April 2003 to complete this task.
B4 - BioMedical workshops. George is dealing with several different people on this issue, including Derek Linkens. There are plans to possibly have an "open discussion" forum with three EUNITE and three MONET speakers. There are also plans to have a workshop at AIME in Cyprus in October 2003. Plan to submit a proposal on QR and MBS in BioMedicine - the deadline for this is the beginning of March 2003. Pedro Barahona who works in QR functional modelling in medicine, is the person responsible. Peter has contacted him, but is still waiting to hear back at present, which he will follow up.
B5 - This is already being looked into via links with EUNITE etc. Many people in the field are also already working on this. Also B1 will form the introduction to this work, providing a picture of where synergies lie. Thus, it will feed into it…

3. Domain Status Document
George firstly thanked Peter for all his outstanding work on the BioMedical Domain Status Document.
There was some discussion as to whether the task group could profit from this work - could it be used as the basis of for a survey paper, as it looks at what other people are doing in the field? Further work could be undertaken on this over the next twelve months and it may then be suitable for a journal publication, possibly AI in Medicine or Knowledge Engineering Review? Alternatively it could be used as the basis of an editorial to go with one of the proposed workshops. The first step would be to organise a workshop, then to publicise that this will be published in an official journal.

4. MONET Funding and organisation of scientific activities
There is the possibility of invited speakers for the workshop as funding should be available.
Outputs and Papers - standards are very high for AIME, but interested in papers from workshops, so should be in a reasonable position. The AI journal is also interested in BioMedical work, when there is emphasis on the modelling aspects.
Other special issues or activities - these will depend on what comes of the EUNITE/AIME/MONET collaborations. After the end of both workshops a call for papers for a special issue will be put forward, combining the work of both workshops.

5. Composition of the task group and establishing contacts
Several people who already work within the field were suggested to be approached regarding participation within the Task Group. George and Peter are going to contact these people to gauge their interest in joining the Task Group. These people are detailed in the following table:

6. Future Task Group meetings
The next Task Group Meeting will take place in Pavia, in June 2003, at a date to be finalised.
There will be further meetings at the workshops in July and October 2003.

7. Any Other Competent Business
There being no further business, the meeting was closed.

Attendee’s
George Coghill – University of Aberdeen
Peter Lucas – University of Nijmegen
Rene Quiniou – University of Rennes
Liliana Ironi – IAN-CNR, Pavia
Stefania Tentoni – IAN-CNR, Pavia
Iain Russell – MONET Project

Day One (6/5/04)

Overall Discussion

B1 - Has been produced and requires updating. PL spoke to Simon Parsons and he

feels that this would make a good Knowledge Engineering Review (KER) Article.

B3 - Has been produced and requires updating.

B4 - Done.

B5 - Synergies Document. We are not necessarily experts in all the fields we need to cover in this document, but we have enough of an understanding of the issues.

Plan for Project Extension - The Task Group would like to look for further work and also the continuation of the Bio-Medical Task Group after MONET2. At the April 2004 Project Review our Project Officer suggested that the ‘Quality of Life - Genomics’ call was an area we may look at. The group agree that they wish to continue the work of the Task Group after the period of MONET has ceased. There is the possibility of focusing on methodologies and showing that we cover the combination of the clinical and biological areas, i.e. show the areas and show the advantages of the methodologies that we propose.
We could ‘widen’ our view to take in ‘Model-based Approaches’ and not necessarily just model-based reasoning and then look at a particular problem and show the complementarities or combinatory advantages of these approaches.
AP 1: Iain to discuss this with Pepé – would a model-based approach to biomedical problems fit within a Framework 6 proposal. Would it make sense to combine the following communities to find the solutions to just a couple of problems / applications. [think ERA] Ask Pepe if we can invite a person(s) to speak at the Roadmapping meeting, from outside MBS&QR community, i.e. not a MONET Member.
List of complementary Communities:
  • Probabilistic Graphical Models
  • Machine Learning
  • Bio-Mathematics
  • Bio-Engineering (NOT Bio-Physics)
  • Fuzzy Logic
  • Medical Decision Making
The interaction of these communities is very important, it can lead to new ideas and the integration of methodologies can result in new and more powerful methods.
[Good area for Roadmap – maybe look into inviting people from these communities to Bio-Med RM Meeting].

Bio-Medical Domain Status Document

Peter will act as editor. The KER article (c. same length as current version) but. format will need to be changed from a ‘report’ to an ‘article’, however the structure makes sense and so should be kept. There are still areas that are not covered, or not covered in sufficient detail. We could split Section 2 into ‘methodologies’ and insert a new Section 3 ‘tasks’. We should focus very closely on the methodologies that have been applied in the bio-medical domain. The document should be completed with a ‘Discussion / Future Work’ section.
AP 2: Peter will email out request for input to the rest of the Task Group.
Next meetings: Aberdeen 13th & 14th September and a final one on 10th and 11th of January in Rennes. This will take place in December if the extension to the project is not granted.

Roadmap Document

Need to include people from healthcare and bio-medical industry. We should email specific people and ask them to attend or send representations. This input does not necessarily need to be from model-based experts, what we need is a view of where future research will be going in the healthcare and bio-medical domain as a whole.
Suggested contacts: Peter – Vitatron, Hugin, Imperial (Welcome), Rene – ELA Medical, (John Fox – Cancer Research / Infermed).
We should get feedback and then work on this document in Aberdeen and then send it out to Industrialists after the next major version is done and it is more industrially focused.

Synergies Document

A format has been proposed by George, this was developed into the following;
Section 1 - Introduction
Section 2 - Model Categorisations
Section 3 - Methodologies and Tasks
• Bayesian Networks and Qualitative Causal Models
• Fuzzy Systems and Qualitative Reasoning (also ‘Fuzzy For Diagnosis’)
• Qualitative Reasoning and Neural Nets
• Drug Design
• Qualitative Reasoning and Machine Learning (Inductive Logic Programming) -
include in Domain Status Doc as well
Section 4 - Discussion and Future Work

Day Two (7/5/04)

Experts to draft a brief of what will be included in the Synergies Document, which should focus on structure of sections and some content that gives us a clear idea of how much work there is to do. This should be emailed to Peter by Wednesday, 12th May 2004.

AP 3: Task Group members to send their text to peter by this wednesday, he will compile, add a draft introduction and distribute. Final text to be sent to Co-ordinating node by 21st June.

The Task Group will also consider if there are any events that would be useful for outreach activities, although the time scale is short. They will, however, think of who they could invite to speak at a TG Meeting or consider some conjoining with the ECAI workshop. This could be used to discuss continuation of the TG after the MONET2 period has finished.

AP 1:	Iain to discuss this with Pepé – would a model-based approach to bio-medical problems fit within a Framework 6 proposal. Would it make sense to combine the following communities to find the solutions to just a couple of problems / applications. [think ERA] Ask Pepe if we can invite a person(s) to speak at the Roadmapping meeting, from outside MBS&QR community.
AP 2:	Peter will email out request for input to the rest of the Task Group.
AP 3:	Task Group members to send their text to peter by this wed, he will compile, add a draft introduction and distribute.

Bio-Medical Task Group Meeting - Aberdeen, 13th September 2004

Attendees: George Coghill, Peter Lucas, Liliana Ironi, Iain Russell

Minutes

Discussion on definition of QR
The group tried to agree on how broad / narrow a definition was to be applied within the Documents from this Task Group.

Discussion of Documents
ACTION - use a non-sans serif font for these documents (times new roman). And cannot is one word. And rework rir's word doc from the latest pdf to ensure they are both the same.

Synergies document - needs completing. We are late on this, a document exists but more work needs to be done on it to make it a complete and coherent document. It has proven quite difficult to edit this document as it is on so many fields (which maybe worth mentioning in introduction). Synergies with Mathematic models are not very well developed in this document. Dependant on the definition of 'AI Approaches' there is actually very little that could not be covered by this document, so we should focus on approaches that have been 'developed' within the AI community and thus cut out many of the more traditional (mathematical, for example) approaches. In Pavia we tried to come up with a categorical definition of what we feel the document should cover. This is not an easy process, especially as we tried to give this a hierarchical structure and to link the approaches methodologically. One method for completing the document would be to list the different approaches then put in a final section which looks into the synergies / possible synergies in the past.

We should extend introduction and we should lose Section 2 as this attempts to come up with a general framework for the area and this is probably a couple of years work, not just a few months. We should then ask the people who worked on the document to complete their section now they can see then whole document.

ACTION - Peter Lucas will take out Section 2 put in header for final section 'Meta-synergies' and then send current version out to members for comment and work. Feedback by the end of September and to be submitted to the Commission at the end of October.

DSD needs to be discussed and rewritten. This is already a coherent document and needs to be revised with what we think is still valid. The TG will have this prepared for the end of December and then re-written (when fresh in mind) at Rene Meeting in January.

BM RM - Liliana has many comments. Also suggests tabulating some information for clarity. The TG worked through the Document, one thing we agreed upon was that this is a collection of perspectives and it would not assist the document to attempt to create a coherent document, but to show as many relevant perspectives as possible.

ACTION - Iain to look into turning section 4 into a table (to reduce replication) with explanations.

PGM Meeting
Look into amount of money that would be left over after Renne meeting and see if we can give some money to Peter (maybe two or three peoples travel / accommodation etc.)

ACTION - Contact René about Rennes Meeting dates / place. René states that he is still happy to host the meeting and suggests the period 2nd to 18th of January 2005.