Gaurav Trivedi1, Charmgil Hong2, Esmaeel R Dadashzadeh3, Robert M Handzel4, Harry Hochheiser5, Shyam Visweswaran6. 1. Intelligent Systems Program, University of Pittsburgh, Pittsburgh, PA, United States; School of Computing and Information, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: trivedigaurav@pitt.edu. 2. School of Computing and Information, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: charmgil@cs.pitt.edu. 3. Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, PA, United States; Department of Surgery, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: dadashzadeher@upmc.edu. 4. Department of Surgery, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: handzelrm@upmc.edu. 5. Intelligent Systems Program, University of Pittsburgh, Pittsburgh, PA, United States; Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: harryh@pitt.edu. 6. Intelligent Systems Program, University of Pittsburgh, Pittsburgh, PA, United States; Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, PA, United States. Electronic address: shv3@pitt.edu.
Abstract
BACKGROUND: Radiologic imaging of trauma patients often uncovers findings that are unrelated to the trauma. These are termed as incidental findings and identifying them in radiology examination reports is necessary for appropriate follow-up. We developed and evaluated an automated pipeline to identify incidental findings at sentence and section levels in radiology reports of trauma patients. METHODS: We created an annotated dataset of 4,181 reports and investigated automated feature representations including traditional word and clinical concept (such as SNOMED CT) representations, as well as word and concept embeddings. We evaluated these representations by using them with traditional classifiers such as logistic regression and with deep learning methods such as convolutional neural networks (CNNs). RESULTS: The best performance was observed using word embeddings with CNNs with F1 scores of 0.66 and 0.52 at section and sentence levels respectively. The F1 score was statistically significantly higher for sections compared to sentences (Wilcoxon; Z < 0.001, p < 0.05). Compared to using words alone, the addition of SNOMED CT concepts did not improve performance. At the sentence level, the F1 score improved significantly from 0.46 to 0.52 when using pre-trained embeddings (Wilcoxon; Z < 0.001, p < 0.05). CONCLUSION: The results show that the best performance was achieved by using embeddings with CNNs at both sentence and section levels. This provides evidence that such a pipeline is capable of accurately identifying incidental findings in radiology reports in an automated manner.
BACKGROUND: Radiologic imaging of traumapatients often uncovers findings that are unrelated to the trauma. These are termed as incidental findings and identifying them in radiology examination reports is necessary for appropriate follow-up. We developed and evaluated an automated pipeline to identify incidental findings at sentence and section levels in radiology reports of traumapatients. METHODS: We created an annotated dataset of 4,181 reports and investigated automated feature representations including traditional word and clinical concept (such as SNOMED CT) representations, as well as word and concept embeddings. We evaluated these representations by using them with traditional classifiers such as logistic regression and with deep learning methods such as convolutional neural networks (CNNs). RESULTS: The best performance was observed using word embeddings with CNNs with F1 scores of 0.66 and 0.52 at section and sentence levels respectively. The F1 score was statistically significantly higher for sections compared to sentences (Wilcoxon; Z < 0.001, p < 0.05). Compared to using words alone, the addition of SNOMED CT concepts did not improve performance. At the sentence level, the F1 score improved significantly from 0.46 to 0.52 when using pre-trained embeddings (Wilcoxon; Z < 0.001, p < 0.05). CONCLUSION: The results show that the best performance was achieved by using embeddings with CNNs at both sentence and section levels. This provides evidence that such a pipeline is capable of accurately identifying incidental findings in radiology reports in an automated manner.
Authors: Charles DiMaggio; Patricia Ayoung-Chee; Matthew Shinseki; Chad Wilson; Gary Marshall; David C Lee; Stephen Wall; Shale Maulana; H Leon Pachter; Spiros Frangos Journal: Injury Date: 2016-04-22 Impact factor: 2.586
Authors: Peter L Elkin; David Froehling; Dietlind Wahner-Roedler; Brett Trusko; Gail Welsh; Haobo Ma; Armen X Asatryan; Jerome I Tokars; S Trent Rosenbloom; Steven H Brown Journal: AMIA Annu Symp Proc Date: 2008-11-06
Authors: Shang Gao; Michael T Young; John X Qiu; Hong-Jun Yoon; James B Christian; Paul A Fearn; Georgia D Tourassi; Arvind Ramanthan Journal: J Am Med Inform Assoc Date: 2018-03-01 Impact factor: 4.497
Authors: Gaurav Trivedi; Esmaeel R Dadashzadeh; Robert M Handzel; Wendy W Chapman; Shyam Visweswaran; Harry Hochheiser Journal: Appl Clin Inform Date: 2019-09-04 Impact factor: 2.342
Authors: Arlene Casey; Emma Davidson; Michael Poon; Hang Dong; Daniel Duma; Andreas Grivas; Claire Grover; Víctor Suárez-Paniagua; Richard Tobin; William Whiteley; Honghan Wu; Beatrice Alex Journal: BMC Med Inform Decis Mak Date: 2021-06-03 Impact factor: 2.796