scholarly journals Automated Identification of Surveillance Colonoscopy in Inflammatory Bowel Disease Using Natural Language Processing

2012 ◽  
Vol 58 (4) ◽  
pp. 936-941 ◽  
Author(s):  
Jason K. Hou ◽  
Mimi Chang ◽  
Thien Nguyen ◽  
Jennifer R. Kramer ◽  
Peter Richardson ◽  
...  
2019 ◽  
Author(s):  
Jason Ken Hou ◽  
Christopher C. Taylor ◽  
Ergin Soysal ◽  
Shubhada Sansgiry ◽  
Peter Richardson ◽  
...  

Abstract Background: Although practice guidelines recommend colorectal cancer surveillance for inflammatory bowel disease (IBD) patients, the natural history of patient with dysplasia is poorly described. Assembling large cohorts of IBD patients with dysplasia is difficult as administrative codes are lacking. The aim of this study was to use natural language processing (NLP) in a large electronic health records (EHR) to identify IBD patients with colonic dysplasia. Methods: We conducted a retrospective cohort study using administrative data from the national Veterans Health Administration (VHA) Corporate Data Warehouse for patients with IBD. Full-text histopathology reports from patients who underwent colonoscopy in the VHA were obtained and a validation cohort was created using a random sample of 2000 reports. An NLP algorithm to identify the presence and grade of dysplasia was developed and performance tested in a validation cohort. The final NLP algorithm was applied to the entire IBD cohort to identify all cases of colonic dysplasia. Results: We identified a total of 44,099 Veterans with IBD, with 22,431 colonoscopy related histopathology reports. NLP had an accuracy of 97.1% for detection of low grade dysplasia, with a precision of 87%, recall of 96.6%, and F- measure of 91.5%. When applied to the entire cohort, a total of 1,762 cases of colonic dysplasia were identified. Conclusions: NLP accurately identifies colonic low-grade dysplasia in IBD patients from a national EHR. NLP can be used to identify large cohorts of IBD patients with dysplasia to further study the natural history and outcomes of colonic dysplasia in patients with IBD.


2020 ◽  
Vol 14 (Supplement_1) ◽  
pp. S355-S355
Author(s):  
M I Calvo Moya ◽  
I Omella Usieto ◽  
M I Vera Mendoza ◽  
V Matallana Royo ◽  
I Gonzalez Partida ◽  
...  

Abstract Background Current therapeutic goals in inflammatory bowel disease (IBD) include not only the mere absence of symptoms but also the resolution of endoscopic lesions, so-called mucosal healing (MH), which has been related to better outcomes. Data regarding the achievement of MH with vedolizumab (VDZ) in real-life clinical practice is still scarce. Methods Retrospective cohort study was carried out in a tertiary hospital between January 2015 and April 2019 including patients with a basal colonoscopy showing activity and who achieved clinical remission under treatment with VDZ, defined by partial Mayo score <2 for ulcerative colitis (UC) and Harvey–Bradshaw Index score (HBI) <4 for Crohn’s disease (CD). Surveillance colonoscopy was performed along with the follow-up according to clinical practice. In UC patients, MH was defined as Mayo Endoscopic Subscore (MES) = 0; the endoscopic response was defined by a decrease in MES ≥1 point. In CD, MH was defined by achievement SES-CD = 0–3 or Rutgeerts index i0; the endoscopic response was defined by a decrease of SES-CD of 50% or Rutgeerts index <i2 with at least 1 point of decease compared with baseline. Results In total, 118 patients treated with VDZ were analysed, but only 45 met inclusion criteria with a median follow-up of 21 (IQR: 14–19) months. Surveillance colonoscopy was performed after a median time of 12 months (IQR:9–17) of treatment. MH achieved in 33/45 patients (73%): 17/23 CD patients (74%) and 16/22 UC patients (73%). The endoscopic response was achieved in 9 of the remaining 12 patients: 3/6 CD patients and 6/6 UC patients. Only 3 (7%) of patients included showed no endoscopic benefit at the time of surveillance endoscopy. In multivariate analysis, probability of not achieving MH was 75% in patients previously treated with immunosuppressants (ISS) (HR 0.25, 0.11–0.55 IC95; p = 0.001) and 60% in patients previously treated with anti-TNFα (HR 0.40, 0.18–0.90 95% CI; p = 0.026). Type of IBD, concomitant ISS, corticosteroid use at induction, baseline endoscopy score or duration of disease before VDZ treatment were not associated with the achievement of MH. Conclusion In our experience, most of the patients who achieve clinical remission with VDZ also achieve MH. Refractory patients were less likely to achieve MH despite having achieved clinical remission.


2020 ◽  
Vol 10 (8) ◽  
pp. 2824
Author(s):  
Yu-Hsiang Su ◽  
Ching-Ping Chao ◽  
Ling-Chien Hung ◽  
Sheng-Feng Sung ◽  
Pei-Ju Lee

Electronic medical records (EMRs) have been used extensively in most medical institutions for more than a decade in Taiwan. However, information overload associated with rapid accumulation of large amounts of clinical narratives has threatened the effective use of EMRs. This situation is further worsened by the use of “copying and pasting”, leading to lots of redundant information in clinical notes. This study aimed to apply natural language processing techniques to address this problem. New information in longitudinal clinical notes was identified based on a bigram language model. The accuracy of automated identification of new information was evaluated using expert annotations as the reference standard. A two-stage cross-over user experiment was conducted to evaluate the impact of highlighting of new information on task demands, task performance, and perceived workload. The automated method identified new information with an F1 score of 0.833. The user experiment found a significant decrease in perceived workload associated with a significantly higher task performance. In conclusion, automated identification of new information in clinical notes is feasible and practical. Highlighting of new information enables healthcare professionals to grasp key information from clinical notes with less perceived workload.


2019 ◽  
Vol 4 (12) ◽  
pp. 971-983 ◽  
Author(s):  
Marietta Iacucci ◽  
Rosanna Cannatelli ◽  
Gian Eugenio Tontini ◽  
Remo Panaccione ◽  
Silvio Danese ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document