in ctakes-drug-ner/src/main/java/org/apache/ctakes/drugner/ae/DrugMentionAnnotator.java [664:873]
private void executeFSMs(JCas jcas, List baseTokenList, List namedE,
List wordTokenList) throws AnnotatorProcessException
{
try
{
Set fractionTokenSet = iv_fractionFSM.execute(baseTokenList);
addAnnotations(jcas, fractionTokenSet, FractionStrengthAnnotation.type);
Set decimalTokenSet = iv_decimalFSM.execute(baseTokenList);
addAnnotations(jcas, decimalTokenSet, FractionStrengthAnnotation.type);
Set statusTokenSet = iv_statusFSM.execute(baseTokenList);
addAnnotations(jcas, statusTokenSet, DrugChangeStatusAnnotation.type);
//statusConfidence = true;
Set rangeTokenSet = iv_rangeFSM.execute(baseTokenList);
addAnnotations(jcas, decimalTokenSet, RangeStrengthAnnotation.type);
//Mayo SPM 2/20/2012 Changed due to separation of strength tokens
Set strengthTokenSet = iv_strengthUnitFSM.execute(baseTokenList, rangeTokenSet);
Iterator measurementTokenItr = strengthTokenSet.iterator();
int begin = 0, previous = 0;
while (measurementTokenItr.hasNext())
{
Iterator chkNE = namedE.iterator();
Iterator newNE = wordTokenList.iterator();
boolean neFound = false;
// MedicationMention ne = null;
WordToken we = null;
Object mt = (Object) measurementTokenItr.next();
if (mt instanceof StrengthUnitToken) {
// StrengthUnitToken mt = (StrengthUnitToken) measurementTokenItr.next();
int begSeg = ((StrengthUnitToken) mt).getStartOffset(), endSeg = ((StrengthUnitToken) mt).getEndOffset();
StrengthUnitAnnotation ma = new StrengthUnitAnnotation(jcas, begSeg, endSeg);
ma.addToIndexes();
} else {
int begSeg = ((StrengthUnitCombinedToken) mt).getStartOffset(), endSeg = ((StrengthUnitCombinedToken) mt).getEndOffset();
StrengthUnitAnnotation ma = new StrengthUnitAnnotation(jcas, begSeg, endSeg);
ma.setBegin(findTextualStringOffset(ma.getCoveredText()) + begSeg);
ma.addToIndexes();
}
//TODO: Does this need to be commented out? Created Strength
}
// Set decTokenSet = null;
// {
// decTokenSet = iv_strengthFSM.execute(baseTokenList, decimalTokenSet,
// fractionTokenSet);
// Iterator decTokenItr = decTokenSet.iterator();
//
// while (decTokenItr.hasNext())
// {
// StrengthToken mt = (StrengthToken) decTokenItr.next();
// StrengthAnnotation ma = new StrengthAnnotation(jcas, mt
// .getStartOffset(), mt.getEndOffset());
// ma.addToIndexes();
// // loadandAppend("./strengthTable.csv", ma.getCoveredText(),
// // true);
//
// }
// }
Set formTokenSet = iv_formFSM.execute(baseTokenList,
/* decimalTokenSet */new HashSet());
Iterator formTokenItr = formTokenSet.iterator();
while (formTokenItr.hasNext())
{
FormToken mt = (FormToken) formTokenItr.next();
FormAnnotation ma = new FormAnnotation(jcas, mt.getStartOffset(), mt.getEndOffset());
ma.addToIndexes();
// if (defaultNE != null && defaultNE.getCoveredText() != null)
// defaultNE.setForm(ma.getCoveredText());
}
// last run checks for numbers (text or actual) in front of dosage
// info and see how we align w/ the ne if they exist to determine
// confidence factoring
// if (!doseConfidence && decTokenSet != null)
// {
Set preTokenSet = iv_strengthFSM.execute(baseTokenList, strengthTokenSet, fractionTokenSet);
Iterator preTokenItr = preTokenSet.iterator();
while (preTokenItr.hasNext())
{
StrengthToken mt = (StrengthToken) preTokenItr.next();
int begOff = mt.getStartOffset();
int endOff = mt.getEndOffset();
StrengthAnnotation ma = new StrengthAnnotation(jcas, begOff, endOff);
Iterator subStrengthItr = FSUtil.getAnnotationsIteratorInSpan(jcas, StrengthUnitAnnotation.type, begOff, endOff);
if (subStrengthItr.hasNext() )
ma.setEnd(((StrengthUnitAnnotation)subStrengthItr.next()).getBegin());
ma.addToIndexes();
// loadandAppend("./strengthTable.csv", ma.getCoveredText(),
// true);
}
// }
Set doseTokenSet = iv_dosagesFSM.execute(baseTokenList, formTokenSet,
strengthTokenSet);
Iterator dosTokenItr = doseTokenSet.iterator();
Iterator formCheckItr = formTokenSet.iterator();
Iterator strengthCheckItr = strengthTokenSet.iterator();
boolean foundDosage = false;
int begSegDose = 0, endSegDose = 0;
while (dosTokenItr.hasNext())
{
DosageToken mt = (DosageToken) dosTokenItr.next();
begSegDose = mt.getStartOffset();
endSegDose = mt.getEndOffset();
DosagesAnnotation ma = new DosagesAnnotation(jcas, begSegDose,
endSegDose);
ma.addToIndexes();
}
Set suffixTokenSet = iv_suffixFSM.execute(baseTokenList, strengthTokenSet);
Iterator suffixTokenItr = suffixTokenSet.iterator();
while (suffixTokenItr.hasNext())
{
SuffixStrengthToken mt = (SuffixStrengthToken) suffixTokenItr.next();
SuffixStrengthAnnotation ma = new SuffixStrengthAnnotation(jcas, mt.getStartOffset(), mt.getEndOffset());
ma.addToIndexes();
}
// This needs to be handled differently. But since I'm not sure if this feature will be utilized
// I am going to leave 'as is' for now.
Set routeTokenSet = iv_routeFSM.execute(baseTokenList);
boolean foundRoute = false;
Iterator routeTokenItr = routeTokenSet.iterator();
int begSegRT = 0, endSegRT = 0;
while (routeTokenItr.hasNext())
{
RouteToken mt = (RouteToken) routeTokenItr.next();
begSegRT = mt.getStartOffset();
endSegRT = mt.getEndOffset();
RouteAnnotation ma = new RouteAnnotation(jcas, begSegRT, endSegRT);
ma.setIntakeMethod(Integer.valueOf(mt.getFormMethod()).toString());
ma.addToIndexes();
}
Set frequencyUnitTokenSet = iv_frequencyUnitFSM.execute(baseTokenList);
boolean foundFrequencyUnit = false;
Iterator frequencyUnitTokenItr = frequencyUnitTokenSet.iterator();
int begSegFUT = 0, endSegFUT = 0;
while (frequencyUnitTokenItr.hasNext())
{
FrequencyUnitToken fut = (FrequencyUnitToken) frequencyUnitTokenItr.next();
begSegFUT = fut.getStartOffset();
endSegFUT = fut.getEndOffset();
FrequencyUnitAnnotation ma = new FrequencyUnitAnnotation(jcas,
begSegFUT, endSegFUT);
ma.setPeriod(Double.valueOf( fut.getFrequencyUnitQuantity()).floatValue());
ma.addToIndexes();
}
// The frequencyFSM can take advantage of the frequencyUnit to
// establish conditions via the override
Set frequencyTokenSet = iv_frequencyFSM.execute(baseTokenList,
frequencyUnitTokenSet, rangeTokenSet);
boolean foundFrequency = false;
Iterator frequencyTokenItr = frequencyTokenSet.iterator();
int begSegFT = 0, endSegFT = 0;
while (frequencyTokenItr.hasNext())
{
FrequencyToken ft = (FrequencyToken) frequencyTokenItr.next();
begSegFT = ft.getStartOffset();
endSegFT = ft.getEndOffset();
FrequencyAnnotation ma = new FrequencyAnnotation(jcas, begSegFT,
endSegFT);
ma.addToIndexes();
}
/* Check again if confidence was found during frequency check */
Set durationTokenSet = iv_durationFSM.execute(baseTokenList,
rangeTokenSet);
Iterator durationTokenItr = durationTokenSet.iterator();
int begSegDU = 0, endSegDU = 0;
while (durationTokenItr.hasNext())
{
DurationToken du = (DurationToken) durationTokenItr.next();
begSegDU = du.getStartOffset();
endSegDU = du.getEndOffset();
DurationAnnotation ma = new DurationAnnotation(jcas, begSegDU, endSegDU);
// dm.setDurationElement(jcas, begSegDU, endSegDU);
ma.addToIndexes();
// loadandAppend("./frequencyTable.csv", ma.getCoveredText(),
// true);
}
} catch (Exception e)
{
throw new AnnotatorProcessException(e);
}
}