Hi Pei, It looks like someone rewrote the section of code I am referencing, so the patch may not be good for trunk. I wonder if the problem is still persisting though.
I did verify on my end that it solve the problem I was encountering. It is unfortunate that I didn't have time sooner to resolve the issue. Kim Ebert 1.801.669.7342 Perfect Search Corp http://www.perfectsearchcorp.com/ On 04/17/2013 08:15 AM, Chen, Pei wrote: > > Hi Kim, > > This actually looks like a pretty good bug fix. > > It seems like you already have the code/patch, should we just commit > it to trunk once you verify the fix? > > > > --Pei > > > > *From:*Kim Ebert [mailto:[email protected]] > *Sent:* Monday, April 15, 2013 4:41 PM > *To:* Deepal Dhariwal > *Cc:* [email protected]; [email protected] > *Subject:* Re: Regarding inconsistency in UMLS Dictionary Lookup Annotator > > > > Hi Deepal, > > I have a patch for cTAKES 2.5 to resolve the problem of batch > processing returning fewer results as processing continues due to the > permutation list becoming corrupt. > > https://issues.apache.org/jira/browse/CTAKES-146 > > The patch should still apply to > ctakes-dictionary-lookup/src/main/java/org/apache/ctakes/dictionary/lookup/algorithms/FirstTokenPermutationImpl.java > > Thanks, > > Kim Ebert > 1.801.669.7342 > Perfect Search Corp > http://www.perfectsearchcorp.com/ > > On 04/15/2013 01:53 PM, Deepal Dhariwal wrote: > > Hello Kim, > > > > Yeah when I processed record individually blood pressure was > annotated as a medical concept. Can you kindly explain the reason > behind this difference in behavior. > > > > Thanks so much. > > > > Regards > > Deepal Dhariwal > > > > > > On Mon, Apr 15, 2013 at 3:39 PM, Kim Ebert > <[email protected] > <mailto:[email protected]>> wrote: > > Hi Deepal, > > Trying running the records individually and see if it identifies > the terms correctly; if so I may have an idea of the problem. > > Thanks, > > > > Kim Ebert > > 1.801.669.7342 <tel:1.801.669.7342> > > Perfect Search Corp > > http://www.perfectsearchcorp.com/ > > On 04/15/2013 01:36 PM, Deepal Dhariwal wrote: > > Hello Kim, > > > > The records were processed in a batch using UIMA Collection > processing Engine. > > The parameters for UIMA CPE were as follows: > > Descriptor: UIMA File System Collection Reader > > Analysis Engine: cTAKES Aggregate UMLS Plaintext Processor > > CAS Consumer: XMI Writer CAS Consumer > > > > Thanks > > Deepal Dhariwal > > > > On Mon, Apr 15, 2013 at 3:31 PM, Kim Ebert > <[email protected] > <mailto:[email protected]>> wrote: > > Hi Deepal, > > Was either record processed in a batch of records? Or where they > processed as individual records? > > Kim Ebert > 1.801.669.7342 <tel:1.801.669.7342> > Perfect Search Corp > http://www.perfectsearchcorp.com/ > > > On 04/15/2013 12:49 PM, Deepal Dhariwal wrote: > > Hello all, > > > > I have 2 medical records with the following content: > > > > Record 1 : > > He went to the hospital Emergency Room where his blood > pressure was > > 120 / 80 , heart rate 94. > > > > Record 2: > > Weight 197 lbs, up to 6 lbs, blood pressure 128/88, heart > rate 70 . > > > > In the first record blood pressure is getting annotated as a > Clinical > > Entity Mention and a Medical Concept while in second record > blood > > pressure is not getting annotated as a clinical entity. > > > > Could someone shed some light on why the difference in > behavior ? I > > have attached the screenshots for reference. > > > > Regards > > Deepal Dhariwal > > > > > > > > >
