- From: <helen.chen@agfa.com>
- Date: Wed, 27 Sep 2006 08:54:13 -0400
- To: VKASHYAP1@PARTNERS.ORG
- Cc: public-semweb-lifesci@w3.org
- Message-ID: <OFF6CD69B0.6B768FF3-ON852571F6.00466246-852571F6.0067D309@agfa.com>
Hi, Vipul Thanks for your "workaround" information. This is the email I received earlier from Samson. Great minds indeed come to the same place :) Jos had implemented a similar solution. One step further, you can have a triple to specify the N in the statement of (at least N out of M signs) so you don't have to be restricted to "at least 2" conditions. Helen ----- Forwarded by Helen Chen/AMPJB/AGFA on 09/27/2006 02:48 PM ----- Samson Tu <swt@stanford.edu> 09/26/2006 06:57 PM To Helen Chen/AMPJB/AGFA@AGFA cc Subject Re: Fw: CAP use case - Reasoning on Weighted Condition and Fuzzy Reasoning? Not sure whether this is going to help: A standard way to do counting with rule programming is to have a bunch of rules of the form: conditioni AND (count ?n) => ( retract (count ?n)) && assert (count ?n + 1)) and then use a lower priority rule (count ?n:&(?n >= 2)) => obtain chest X-ray. In the 80s people found that, if you need to manage "weights" and uncertainties in diagnosis, it's not a good idea to use rules. That's when Bayesian network become useful. SAmson helen.chen@agfa.com wrote: > > I am forwarding this discussion at ACPP to the HCLS list, hope to get > some input on this tread. > > ----- Forwarded by Helen Chen/AMPJB/AGFA on 09/26/2006 04:53 PM ----- > *Helen Chen/AMPJB/AGFA* > > 09/26/2006 08:30 AM > > > To > Chimezie Ogbuji > cc > cebarr01@yahoo.com, aziz@boxwala.com, sam.brandt@siemens.com, > THONGSERMEIER@PARTNERS.ORG, ogbujic@bio.ri.ccf.org, > davide@landcglobal.com, DAN.RUSSLER@ORACLE.COM > Subject > CAP use case > > > > > > > > > Hi, Chimezie > > As per our discussion last Tcon, here is the Guideline for Managing > Community Acquired Pneumonia (CAP). > > http://www.guideline.gov/summary/summary.aspx?doc_id=9398&nbr=005034&string=community+AND+acquired+AND+pneumonia > > > The algorithm is here : > > http://www.guideline.gov/algorithm/5034/NGC-5034_1.html > > Notice in the step 3, it says: > > 3: "Obtain chest X-ray, especially if patient has two or more of these > signs: > Temp > 100F > Pulse > 100 > Decreased breath sounds > Rales > Respiratory rate > 20 > > Now we are facing the new problem of modelling "two or more" facts of a > necessary condition for "order chest X-ray" in the knowledge base. > Furthermore, doctors will likely tell you that no only they need to > express "at least two or more", they also want to express "fact A > carries more weight or more indicative to a diagnosis than fact B". If > we were to model these "weighted condition", we are opening a whole can > of new worms, and I don't think any SW reasoners now can do reasoning on > this. > -- Samson Tu email: swt@stanford.edu Senior Research Scientist web: www.stanford.edu/~swt/ Stanford Medical Informatics phone: 1-650-725-3391 Stanford University fax: 1-650-725-7944
Received on Wednesday, 27 September 2006 12:54:33 UTC