Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 3 de 3
Filter
1.
Appl Clin Inform ; 7(4): 1051-1068, 2016 11 09.
Article in English | MEDLINE | ID: mdl-27826610

ABSTRACT

BACKGROUND: Important information to support healthcare quality improvement is often recorded in free text documents such as radiology reports. Natural language processing (NLP) methods may help extract this information, but these methods have rarely been applied outside the research laboratories where they were developed. OBJECTIVE: To implement and validate NLP tools to identify long bone fractures for pediatric emergency medicine quality improvement. METHODS: Using freely available statistical software packages, we implemented NLP methods to identify long bone fractures from radiology reports. A sample of 1,000 radiology reports was used to construct three candidate classification models. A test set of 500 reports was used to validate the model performance. Blinded manual review of radiology reports by two independent physicians provided the reference standard. Each radiology report was segmented and word stem and bigram features were constructed. Common English "stop words" and rare features were excluded. We used 10-fold cross-validation to select optimal configuration parameters for each model. Accuracy, recall, precision and the F1 score were calculated. The final model was compared to the use of diagnosis codes for the identification of patients with long bone fractures. RESULTS: There were 329 unique word stems and 344 bigrams in the training documents. A support vector machine classifier with Gaussian kernel performed best on the test set with accuracy=0.958, recall=0.969, precision=0.940, and F1 score=0.954. Optimal parameters for this model were cost=4 and gamma=0.005. The three classification models that we tested all performed better than diagnosis codes in terms of accuracy, precision, and F1 score (diagnosis code accuracy=0.932, recall=0.960, precision=0.896, and F1 score=0.927). CONCLUSIONS: NLP methods using a corpus of 1,000 training documents accurately identified acute long bone fractures from radiology reports. Strategic use of straightforward NLP methods, implemented with freely available software, offers quality improvement teams new opportunities to extract information from narrative documents.


Subject(s)
Fractures, Bone/diagnostic imaging , Natural Language Processing , Quality Improvement , Radiology , Research Report , Child , Clinical Decision-Making , Documentation , Emergency Medicine , Humans
2.
Pediatrics ; 123(1): 338-45, 2009 Jan.
Article in English | MEDLINE | ID: mdl-19117901

ABSTRACT

OBJECTIVE: Aspiration pneumonia is the most common cause of death in children with neurologic impairment who have gastroesophageal reflux disease. Fundoplications and gastrojejunal feeding tubes are frequently employed to prevent aspiration pneumonia in this population. Which of these approaches is more effective in preventing aspiration pneumonia and/or improving survival is unknown. The objective of this study was to compare outcomes for children with neurologic impairment and gastroesophageal reflux disease after either a first fundoplication or a first gastrojejunal feeding tube. PATIENTS AND METHODS: This was a retrospective, observational cohort study of children with neurologic impairment who had either a fundoplication or gastrojejunal feeding tube between January 1997 and December 2005 at a tertiary care children's hospital. Main outcome measures were postprocedure aspiration pneumonia-free survival and mortality. Propensity analyses were used to control for bias in treatment assignment and prognostic imbalances. RESULTS: Of the 366 children with neurologic impairment and gastroesophageal reflux disease, 43 had a first gastrojejunal feeding tube and 323 underwent a first fundoplication. Median length of follow-up was 3.4 years. Children who received a first fundoplication had similar rates of aspiration pneumonia and mortality after the procedure compared with those who had a first gastrojejunal feeding tube, when adjusting for the treatment assignment using propensity scores. CONCLUSIONS: Aspiration pneumonia and mortality are not uncommon events after either a first fundoplication or a first gastrojejunal feeding tube for the management of gastroesophageal reflux disease in children with neurologic impairment. Neither treatment option is clearly superior in preventing the subsequent aspiration pneumonia or improving overall survival for these children. This complex clinical scenario needs to be studied in a prospective, multicenter, randomized control trial to evaluate definitively whether 1 of these 2 management options is more beneficial.


Subject(s)
Enteral Nutrition/mortality , Fundoplication/mortality , Gastroesophageal Reflux/mortality , Nervous System Diseases/mortality , Pneumonia, Aspiration/mortality , Pneumonia, Aspiration/prevention & control , Child, Preschool , Cohort Studies , Enteral Nutrition/methods , Female , Follow-Up Studies , Fundoplication/methods , Gastroesophageal Reflux/complications , Gastroesophageal Reflux/surgery , Humans , Infant , Male , Nervous System Diseases/complications , Nervous System Diseases/surgery , Pneumonia, Aspiration/surgery , Retrospective Studies , Survival Rate/trends
3.
J Hosp Med ; 2(3): 165-73, 2007 May.
Article in English | MEDLINE | ID: mdl-17549766

ABSTRACT

BACKGROUND: Children with neurological impairment (NI) commonly have gastroesophageal reflux disease (GERD) treated with a fundoplication. The impact of this procedure on quality of life is poorly understood. OBJECTIVES: To examine the quality of life of children with NI who have received a fundoplication for GERD and of their caregivers. METHODS: The study was a prospective cohort study of children with NI and GERD who underwent a fundoplication at a children's hospital between January 1, 2005, and July 7, 2006. Quality of life of the children was assessed with the Child Health Questionnaire (CHQ) and of the caregivers with the Short-Form Health Survey Status (SF-36) and Parenting Stress Index (PSI), both at baseline and 1 month after fundoplication. Functional status was assessed using the WeeFIM. Repeated-measures analyses were performed. RESULTS: Forty-four of the 63 parents (70%) were enrolled. The median WeeFIM score was 31.2 versus the age-normal score of 83 (P = .001). Compared with the baseline scores, mean CHQ scores improved over 1 month in the domains of bodily pain (32.8 vs. 47.5, P = .01), role limitations-physical (30.6 vs. 56.6, P = .01), mental health (62.7 vs. 70.6, P = .01), family limitation of activities (43.3 vs. 55.1, P = .03), and parental time (43.0 vs. 55.3, P = .03). The parental SF-36 domain of vitality improved from baseline over 1 month (41.3 vs. 48.2, P = .001), but there were no changes from baseline in Parenting Stress scores. CONCLUSIONS: Parents reported that the quality of life of children with NI who receive a fundoplication for GERD was improved from baseline in several domains 1 month after surgery. The quality of life and stress of caregivers did not improve in nearly all domains, at least in the short term.


Subject(s)
Caregivers , Fundoplication , Gastroesophageal Reflux/surgery , Nervous System Diseases/complications , Quality of Life , Case-Control Studies , Child, Preschool , Fundoplication/adverse effects , Gastroesophageal Reflux/complications , Humans , Infant , Prospective Studies , Stress, Psychological , Utah
SELECTION OF CITATIONS
SEARCH DETAIL
...