Characterization of patients with advanced chronic pancreatitis using natural language processing of radiology reports

Abstract

Study aim To develop and apply a natural language processing algorithm for characterization of patients diagnosed with chronic pancreatitis in a diverse integrated U.S. healthcare system. Methods Retrospective cohort study including patients initially diagnosed with chronic pancreatitis (CP) within a regional integrated healthcare system between January 1, 2006 and December 31, 2015. Imaging reports from these patients were extracted from the electronic medical record system and split into training, validation and implementation datasets. A natural language processing (NLP) algorithm was first developed through the training dataset to identify specific features (atrophy, calcification, pseudocyst, cyst and main duct dilatation) from free-text radiology reports. The validation dataset was applied to validate the performance by comparing against the manual chart review. The developed algorithm was then applied to the implementation dataset. We classified patients with calcification(s) or �2 radiographic features as advanced CP. We compared etiology, comorbid conditions, treatment parameters as well as survival between advanced CP and others diagnosed during the study period. Results 6,346 patients were diagnosed with CP during the study period with 58,085 radiology studies performed. For individual features, NLP yielded sensitivity from 88.7% to 95.3%, specificity from 98.2% to 100.0%. A total of 3,672 patients met cohort inclusion criteria: 1,330 (36.2%) had evidence of advanced CP. Patients with advanced CP had increased frequency of smoking (57.8% vs. 43.0%), diabetes (47.6% vs. 35.9%) and underweight body mass index (6.6% vs. 3.6%), all p<0.001. Mortality from pancreatic cancer was higher in advanced CP (15.3/1,000 person-year vs. 2.8/1,000, p<0.001). Underweight BMI (HR 1.6,

Associations between body mass index and all‐cause mortality: A systematic review and meta‐analysis

“…There were 16 studies [144][145][146][147][148][149][150][151][152][153][154][155][156][157][158][159] (425,072 participants) with diabetes, pancreatitis, kidney disease, or hepatic disease populations (Table S3): 1; Figure 7). The heterogeneity was very large (I 2 100% and 100%, respectively).…”

Electronic case report forms generation from pathology reports by ARGO, automatic record generator for onco-hematology

“…In comparison with other applications in oncology, ARGO confirmed super-imposable performances in data field detection 6 , 11 , 14 , 16 , 21 , 22 , while overcoming some limitations. For instance, in the work by Nguyen et al, each metric decreases as the number of classes describing a certain data field increases 11 .…”

“…Natural Language Processing (NLP) is a consolidated technique to extract essential unstructured data from text, e.g. from diagnostic and prognostic notes 6 – 10 , widely adopted also in onco-hematology 11 – 16 . REDCap (Research Electronic Data CAPture) is a recognized platform of electronic case report forms (eCRFs) enabling rapid, high-quality and standardized annotation of data 17 , 18 .…”

Clinical and Imaging Predictors of Pancreatic Cancer in Patients Hospitalized for Acute Pancreatitis

“…We then applied a previously developed natural language processing algorithm for identification of pancreatic parenchymal atrophy as well as main duct dilatation given the previous association of these features with development of pancreatic cancer. [11][12][13][14] Briefly, the algorithm identified duct dilatation from the imaging reports based on the presence of terms such as "dilated", "dilation", and "dilatation", while excluding phrases such as "bile duct", "common duct", and "nondilated". 13 Patients who were identified with pancreatic cancer based on imaging reports were also excluded from the study.…”

“…[11][12][13][14] Briefly, the algorithm identified duct dilatation from the imaging reports based on the presence of terms such as "dilated", "dilation", and "dilatation", while excluding phrases such as "bile duct", "common duct", and "nondilated". 13 Patients who were identified with pancreatic cancer based on imaging reports were also excluded from the study.…”