Natural Language Processing and Machine Learning to Enable Clinical Decision Support for Treatment of Pediatric Pneumonia

Pneumonia is the most frequent cause of infectious disease-related deaths in children worldwide. Clinical decision support (CDS) applications can guide appropriate treatment, but the system must first recognize the appropriate diagnosis. To enable CDS for pediatric pneumonia, we developed an algorithm integrating natural language processing (NLP) and random forest classifiers to identify potential pediatric pneumonia from radiology reports. We deployed the algorithm in the EHR of a large children's hospital using real-time NLP. We describe the development and deployment of the algorithm, and evaluate our approach using 9-months of data gathered while the system was in use. Our model, trained on individual radiology reports, had an AUC of 0.954. The intervention, evaluated on patient encounters that could include multiple radiology reports, achieved a sensitivity, specificity, and positive predictive value of0.899, 0.949, and 0.781, respectively.

[1]  Meng Zhang,et al.  The Clinical Utility of Chest Radiography for Identifying Pneumonia: Accounting for Diagnostic Uncertainty in Radiology Reports. , 2019, AJR. American journal of roentgenology.

[2]  Derek J. Williams,et al.  Pediatric Community-Acquired Pneumonia in the United States , 2017, Infectious Disease Clinics of North America.

[3]  M. Neuman,et al.  Does This Child Have Pneumonia?: The Rational Clinical Examination Systematic Review , 2017, JAMA.

[4]  Ramkiran Gouripeddi,et al.  Enhancing Comparative Effectiveness Research With Automated Pediatric Pneumonia Detection in a Multi-Institutional Clinical Repository: A PHIS+ Pilot Study , 2017, Journal of medical Internet research.

[5]  Peter K Lindenauer,et al.  Epidemiology of pediatric hospitalizations at general hospitals and freestanding children's hospitals in the United States. , 2016, Journal of hospital medicine.

[6]  Frank E Harrell,et al.  Predicting Severe Pneumonia Outcomes in Children , 2016, Pediatrics.

[7]  Dominik Aronsky,et al.  Impact of an Electronic Clinical Decision Support Tool for Emergency Department Patients With Pneumonia. , 2015, Annals of emergency medicine.

[8]  Anami R Patel,et al.  Community-acquired pneumonia requiring hospitalization among U.S. children. , 2015, The New England journal of medicine.

[9]  Vincent Liu,et al.  Automated identification of pneumonia in chest radiograph reports in critically ill patients , 2013, BMC Medical Informatics and Decision Making.

[10]  A. Donald,et al.  Supplemental Material to , 2013 .

[11]  M. Sheng,et al.  Radiological findings in 210 paediatric patients with viral pneumonia: a retrospective case study. , 2012, The British journal of radiology.

[12]  John S. Bradley,et al.  The Management of Community-Acquired Pneumonia in Infants and Children Older Than 3 Months of Age: Clinical Practice Guidelines by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America , 2011, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America.

[13]  Peter L. Elkin,et al.  Detection of pneumonia using free-text radiology reports in the BioSense system , 2011, Int. J. Medical Informatics.

[14]  Wendy W. Chapman,et al.  ConText: An algorithm for determining negation, experiencer, and temporal status from clinical reports , 2009, J. Biomed. Informatics.

[15]  S. Trent Rosenbloom,et al.  NLP-based Identification of Pneumonia Cases from Free-Text Radiological Reports , 2008, AMIA.

[16]  V. Novack,et al.  Disagreement in the interpretation of chest radiographs among specialists and clinical outcomes of patients hospitalized with suspected pneumonia. , 2006, European journal of internal medicine.

[17]  Carol Friedman,et al.  Extracting Information on Pneumonia in Infants Using Natural Language Processing of Radiology Reports , 2003, BioNLP@ACL.

[18]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[19]  Wendy W. Chapman,et al.  A Simple Algorithm for Identifying Negated Findings and Diseases in Discharge Summaries , 2001, J. Biomed. Informatics.

[20]  Peter J. Haug,et al.  Combining decision support methodologies to diagnose pneumonia , 2001, AMIA.

[21]  Peter J. Haug,et al.  Automatic Identification of Patients Eligible for a Pneumonia Guideline: Comparing the Diagnostic Accuracy of Two Decision Support Models , 2001, MedInfo.

[22]  Peter J. Haug,et al.  Research Paper: Automatic Detection of Acute Bacterial Pneumonia from Chest X-ray Reports , 2000, J. Am. Medical Informatics Assoc..

[23]  Peter J. Haug,et al.  Automatic identification of patients eligible for a pneumonia guideline , 2000, AMIA.

[24]  Peter J. Haug,et al.  Comparing expert systems for identifying chest x-ray reports that support pneumonia , 1999, AMIA.

[25]  Peter J. Haug,et al.  An integrated decision support system for diagnosing and managing patients with community-acquired pneumonia , 1999, AMIA.

[26]  M Korppi,et al.  Radiological diagnosis of pneumonia in children. , 1996, Annals of medicine.

[27]  L M Lau,et al.  A natural language understanding system combining syntactic and semantic techniques. , 1994, Proceedings. Symposium on Computer Applications in Medical Care.