A W Olthof1, P Shouche2, E M Fennema3, F F A IJpma3, R H C Koolstra4, V M A Stirler3, P M A van Ooijen5, L J Cornelissen6. 1. Department of Radiology, Treant Health Care Group, Dr. G.H. Amshoffweg 1, Hoogeveen, the Netherlands; Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands. Electronic address: a.olthof@treant.nl. 2. Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands. 3. Department of Trauma Surgery, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands. 4. Department of Radiology, Treant Health Care Group, Dr. G.H. Amshoffweg 1, Hoogeveen, the Netherlands. 5. Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands; Machine Learning Lab, Data Science Center in Health (DASH),University Medical Center Groningen, University of Groningen, L.J. Zielstraweg 2, Groningen, the Netherlands. 6. Department of Radiation Oncology, University Medical Center Groningen, University of Groningen, Hanzeplein 1, Groningen, the Netherlands; COSMONiO Imaging BV, L.J. Zielstraweg 2, Groningen, the Netherlands.
Abstract
OBJECTIVES: To compare different Machine Learning (ML) Natural Language Processing (NLP) methods to classify radiology reports in orthopaedic trauma for the presence of injuries. Assessing NLP performance is a prerequisite for downstream tasks and therefore of importance from a clinical perspective (avoiding missed injuries, quality check, insight in diagnostic yield) as well as from a research perspective (identification of patient cohorts, annotation of radiographs). METHODS: Datasets of Dutch radiology reports of injured extremities (n = 2469, 33% fractures) and chest radiographs (n = 799, 20% pneumothorax) were collected in two different hospitals and labeled by radiologists and trauma surgeons for the presence or absence of injuries. NLP classification was applied and optimized by testing different preprocessing steps and different classifiers (Rule-based, ML, and Bidirectional Encoder Representations from Transformers (BERT)). Performance was assessed by F1-score, AUC, sensitivity, specificity and accuracy. RESULTS: The deep learning based BERT model outperforms all other classification methods which were assessed. The model achieved an F1-score of (95 ± 2)% and accuracy of (96 ± 1)% on a dataset of simple reports (n= 2469), and an F1 of (83 ± 7)% with accuracy (93 ± 2)% on a dataset of complex reports (n= 799). CONCLUSION: BERT NLP outperforms traditional ML and rule-base classifiers when applied to Dutch radiology reports in orthopaedic trauma.
OBJECTIVES: To compare different Machine Learning (ML) Natural Language Processing (NLP) methods to classify radiology reports in orthopaedic trauma for the presence of injuries. Assessing NLP performance is a prerequisite for downstream tasks and therefore of importance from a clinical perspective (avoiding missed injuries, quality check, insight in diagnostic yield) as well as from a research perspective (identification of patient cohorts, annotation of radiographs). METHODS: Datasets of Dutch radiology reports of injured extremities (n = 2469, 33% fractures) and chest radiographs (n = 799, 20% pneumothorax) were collected in two different hospitals and labeled by radiologists and trauma surgeons for the presence or absence of injuries. NLP classification was applied and optimized by testing different preprocessing steps and different classifiers (Rule-based, ML, and Bidirectional Encoder Representations from Transformers (BERT)). Performance was assessed by F1-score, AUC, sensitivity, specificity and accuracy. RESULTS: The deep learning based BERT model outperforms all other classification methods which were assessed. The model achieved an F1-score of (95 ± 2)% and accuracy of (96 ± 1)% on a dataset of simple reports (n= 2469), and an F1 of (83 ± 7)% with accuracy (93 ± 2)% on a dataset of complex reports (n= 799). CONCLUSION: BERT NLP outperforms traditional ML and rule-base classifiers when applied to Dutch radiology reports in orthopaedic trauma.
Authors: Yaguang Zheng; Victoria Vaughan Dickson; Saul Blecker; Jason M Ng; Brynne Campbell Rice; Gail D'Eramo Melkus; Liat Shenkar; Marie Claire R Mortejo; Stephen B Johnson Journal: JMIR Diabetes Date: 2022-05-16