eprintid: 10179572 rev_number: 6 eprint_status: archive userid: 699 dir: disk0/10/17/95/72 datestamp: 2023-10-25 11:26:41 lastmod: 2023-10-25 11:26:41 status_changed: 2023-10-25 11:26:41 type: article metadata_visibility: show sword_depositor: 699 creators_name: Casey, Arlene creators_name: Davidson, Emma creators_name: Grover, Claire creators_name: Tobin, Richard creators_name: Grivas, Andreas creators_name: Zhang, Huayu creators_name: Schrempf, Patrick creators_name: O'Neil, Alison Q creators_name: Lee, Liam creators_name: Walsh, Michael creators_name: Pellie, Freya creators_name: Ferguson, Karen creators_name: Cvoro, Vera creators_name: Wu, Honghan creators_name: Whalley, Heather creators_name: Mair, Grant creators_name: Whiteley, William creators_name: Alex, Beatrice title: Understanding the performance and reliability of NLP tools: a comparison of four NLP tools predicting stroke phenotypes in radiology reports ispublished: pub divisions: UCL divisions: B02 divisions: DD4 keywords: brain radiology, electronic health records, natural language processing, stroke phenotype note: © 2023 Casey, Davidson, Grover, Tobin, Grivas, Zhang, Schrempf, O’Neil, Lee, Walsh, Pellie, Ferguson, Cvero, Wu, Whalley, Mair, Whiteley and Alex. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/). abstract: BACKGROUND: Natural language processing (NLP) has the potential to automate the reading of radiology reports, but there is a need to demonstrate that NLP methods are adaptable and reliable for use in real-world clinical applications. METHODS: We tested the F1 score, precision, and recall to compare NLP tools on a cohort from a study on delirium using images and radiology reports from NHS Fife and a population-based cohort (Generation Scotland) that spans multiple National Health Service health boards. We compared four off-the-shelf rule-based and neural NLP tools (namely, EdIE-R, ALARM+, ESPRESSO, and Sem-EHR) and reported on their performance for three cerebrovascular phenotypes, namely, ischaemic stroke, small vessel disease (SVD), and atrophy. Clinical experts from the EdIE-R team defined phenotypes using labelling techniques developed in the development of EdIE-R, in conjunction with an expert researcher who read underlying images. RESULTS: EdIE-R obtained the highest F1 score in both cohorts for ischaemic stroke, ≥93%, followed by ALARM+, ≥87%. The F1 score of ESPRESSO was ≥74%, whilst that of Sem-EHR is ≥66%, although ESPRESSO had the highest precision in both cohorts, 90% and 98%. For F1 scores for SVD, EdIE-R scored ≥98% and ALARM+ ≥90%. ESPRESSO scored lowest with ≥77% and Sem-EHR ≥81%. In NHS Fife, F1 scores for atrophy by EdIE-R and ALARM+ were 99%, dropping in Generation Scotland to 96% for EdIE-R and 91% for ALARM+. Sem-EHR performed lowest for atrophy at 89% in NHS Fife and 73% in Generation Scotland. When comparing NLP tool output with brain image reads using F1 scores, ALARM+ scored 80%, outperforming EdIE-R at 66% in ischaemic stroke. For SVD, EdIE-R performed best, scoring 84%, with Sem-EHR 82%. For atrophy, EdIE-R and both ALARM+ versions were comparable at 80%. CONCLUSIONS: The four NLP tools show varying F1 (and precision/recall) scores across all three phenotypes, although more apparent for ischaemic stroke. If NLP tools are to be used in clinical settings, this cannot be performed "out of the box." It is essential to understand the context of their development to assess whether they are suitable for the task at hand or whether further training, re-training, or modification is required to adapt tools to the target task. date: 2023 date_type: published official_url: https://doi.org/10.3389/fdgth.2023.1184919 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 2098631 doi: 10.3389/fdgth.2023.1184919 medium: Electronic-eCollection lyricists_name: Wu, Honghan lyricists_id: HWWUX46 actors_name: Flynn, Bernadette actors_id: BFFLY94 actors_role: owner funding_acknowledgements: 216767/Z/19/Z [Wellcome Trust]; R484/0516 [The Dunhill Medical Trust]; CAF/17/01 [Chief Scientist Office] full_text_status: public publication: Frontiers in Digital Health volume: 5 article_number: 1184919 event_location: Switzerland citation: Casey, Arlene; Davidson, Emma; Grover, Claire; Tobin, Richard; Grivas, Andreas; Zhang, Huayu; Schrempf, Patrick; ... Alex, Beatrice; + view all <#> Casey, Arlene; Davidson, Emma; Grover, Claire; Tobin, Richard; Grivas, Andreas; Zhang, Huayu; Schrempf, Patrick; O'Neil, Alison Q; Lee, Liam; Walsh, Michael; Pellie, Freya; Ferguson, Karen; Cvoro, Vera; Wu, Honghan; Whalley, Heather; Mair, Grant; Whiteley, William; Alex, Beatrice; - view fewer <#> (2023) Understanding the performance and reliability of NLP tools: a comparison of four NLP tools predicting stroke phenotypes in radiology reports. Frontiers in Digital Health , 5 , Article 1184919. 10.3389/fdgth.2023.1184919 <https://doi.org/10.3389/fdgth.2023.1184919>. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10179572/1/fdgth-05-1184919.pdf