Filters








15 Hits in 3.6 sec

CheXbert: Combining Automatic Labelers and Expert Annotations for Accurate Radiology Report Labeling Using BERT [article]

Akshay Smit, Saahil Jain, Pranav Rajpurkar, Anuj Pareek, Andrew Y. Ng, Matthew P. Lungren
2020 arXiv   pre-print
In this work, we introduce a BERT-based approach to medical image report labeling that exploits both the scale of available rule-based systems and the quality of expert annotations.  ...  We demonstrate superior performance of a biomedically pretrained BERT model first trained on annotations of a rule-based labeler and then finetuned on a small set of expert annotations augmented with automated  ...  Thanks to Alistair Johnson for support in the radiologist benchmark, to Jeremy Irvin for support in the CheXpert labeler, Alex Tamkin for helpful comments, and Yifan Peng for helpful feedback.  ... 
arXiv:2004.09167v3 fatcat:ibixaghabnakljrfyw2c7azs6y

Combining Automatic Labelers and Expert Annotations for Accurate Radiology Report Labeling Using BERT

Akshay Smit, Saahil Jain, Pranav Rajpurkar, Anuj Pareek, Andrew Ng, Matthew Lungren
2020 Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)   unpublished
We demonstrate superior performance of a biomedically pretrained BERT model first trained on annotations of a rulebased labeler and then fine-tuned on a small set of expert annotations augmented with automated  ...  In this work, we introduce a BERTbased approach to medical image report labeling that exploits both the scale of available rule-based systems and the quality of expert annotations.  ...  Thanks to Alistair Johnson for support in the radiologist benchmark, to Jeremy Irvin for support in the CheXpert labeler, Alex Tamkin for helpful comments, and Yifan Peng for helpful feedback.  ... 
doi:10.18653/v1/2020.emnlp-main.117 fatcat:kbdqwqwelfbyzldxb6xybh57jq

Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality

Hyun Gi Lee, Evan Sholle, Ashley Beecy, Subhi Al'Aref, Yifan Peng
2021 Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies  
Furthermore, they typically require a large amount of data and high-quality expert annotations for training.  ...  In this work, we present a novel method of using BERT-based hidden layer representations of clinical texts as covariates for proportional hazards models to predict patient survival outcomes.  ...  Pranav Rajpurkar and Dr. Matthew P Lungren for providing the ChexBert model.  ... 
doi:10.18653/v1/2021.naacl-main.358 pmid:35463193 pmcid:PMC9034454 fatcat:gfwmdfgl25heze7vjlxcl3oimq

Explaining Chest X-ray Pathologies in Natural Language [article]

Maxime Kayser, Cornelius Emde, Oana-Maria Camburu, Guy Parsons, Bartlomiej Papiez, Thomas Lukasiewicz
2022 arXiv   pre-print
Most deep learning algorithms lack explanations for their predictions, which limits their deployment in clinical practice.  ...  NLEs are human-friendly and comprehensive, and enable the training of intrinsically explainable models.  ...  Acknowledgments We thank Sarim Ather for useful discussions and feedback. M.K. is supported by the EPSRC Center for Doctoral Training in Health Data Science (EP/S02428X/1), and by Elsevier BV.  ... 
arXiv:2207.04343v1 fatcat:6ivnavxmkbh5jlr5ym3rbnpvly

Making the Most of Text Semantics to Improve Biomedical Vision–Language Processing [article]

Benedikt Boecking, Naoto Usuyama, Shruthi Bannur, Daniel C. Castro, Anton Schwaighofer, Stephanie Hyland, Maria Wetscherek, Tristan Naumann, Aditya Nori, Javier Alvarez-Valle, Hoifung Poon, Ozan Oktay
2022 arXiv   pre-print
and discourse characteristics in radiology reports.  ...  Multi-modal data abounds in biomedicine, such as radiology images and reports. Interpreting this data at scale is essential for improving clinical care and accelerating clinical research.  ...  , and Dr Matthew Lungren for their clinical input and data annotations provided to this study.  ... 
arXiv:2204.09817v4 fatcat:c72thidabfbgpgln7yjeoeu6ae

Learning Semi-Structured Representations of Radiology Reports [article]

Tamara Katic, Martin Pavlovski, Danijela Sekulic, Slobodan Vucetic
2021 arXiv   pre-print
This paper aims to present an approach for the automatic generation of semi-structured representations of radiology reports.  ...  We evaluated the proposed approach on the OpenI corpus of manually annotated chest x-ray radiology reports.  ...  Chexbert: combining automatic labelers and expert an- notations for accurate radiology report labeling using bert. arXiv preprint arXiv:2004.09167, 2020. 10.  ... 
arXiv:2112.10746v1 fatcat:sfp4gnodvjepnfga6nidllyxym

Improving Factual Completeness and Consistency of Image-to-Text Radiology Report Generation [article]

Yasuhide Miura, Yuhao Zhang, Emily Bao Tsai, Curtis P. Langlotz, Dan Jurafsky
2021 arXiv   pre-print
Neural image-to-text radiology report generation systems offer the potential to improve radiology reporting by reducing the repetitive process of report drafting and identifying possible medical errors  ...  We combine these with the novel use of an existing semantic equivalence metric (BERTScore). We further propose a report generation system that optimizes these rewards via reinforcement learning.  ...  Acknowledgements We would like to thank the anonymous reviewers and the members of the Stanford NLP Group for their very helpful comments that substantially improved this paper.  ... 
arXiv:2010.10042v2 fatcat:7tgoflxc7zburngk44g3ulvsye

Neural Natural Language Processing for Unstructured Data in Electronic Health Records: a Review [article]

Irene Li, Jessica Pan, Jeremy Goldwasser, Neha Verma, Wai Pan Wong, Muhammed Yavuz Nuzumlalı, Benjamin Rosand, Yixin Li, Matthew Zhang, David Chang, R. Andrew Taylor, Harlan M. Krumholz (+1 others)
2021 arXiv   pre-print
Well over half of the information stored within EHRs is in the form of unstructured text (e.g. provider notes, operation reports) and remains largely untapped for secondary use.  ...  Despite this central role, EHRs are notoriously difficult to process automatically.  ...  CheXbert overcomes this limitation by learning to label radiology reports using both annotations and existing rule-based systems.  ... 
arXiv:2107.02975v1 fatcat:nayhw7gadfdzrovycdkvzy75pi

Using BERT Models to Label Radiology Reports

John R Zech
2022
is a radiology resident at New York Presbyterian-Columbia and incoming musculoskeletal radiology fellow at NYU. He previously served on the Radiology: Artificial Intelligence trainee editorial board.  ...  He is currently pursuing a Radiological Society of North America Resident Research Grant-funded project to develop a deep learning-based tool to assist radiologists in detecting pediatric upper extremity  ...  First, they evaluate how these models perform with small training datasets (CheXbert had 1687 labeled reports and used 75% for training).  ... 
doi:10.1148/ryai.220124 pmid:35923380 pmcid:PMC9344203 fatcat:ukanxqg4hjbklgj6j6c6zdv2te

Government plans in the 2016 and 2021 Peruvian presidential elections: A natural language processing analysis of the health chapters

Rodrigo M. Carrillo-Larco, Manuel Castillo-Cara, Jesús Lovón-Melgarejo
2021 Wellcome Open Research  
Representations from Transformers (KeyBERT), and Rapid Automatic Keywords Extraction (Rake).  ...  NLP analysis could also be included in research of health policies and politics during general elections and provide informative summaries for the general population.  ...  .: Chexbert: combining automatic labelers and expert annotations for accurate radiology report labeling using bert . arXiv preprint arXiv: 2004.09167. 2020.  ... 
doi:10.12688/wellcomeopenres.16867.1 fatcat:q45k6j4qhrhzxptxyriqzmrpcy

Government plans in the 2016 and 2021 Peruvian presidential elections: A natural language processing analysis of the health chapters

Rodrigo M. Carrillo-Larco, Manuel Castillo-Cara, Jesús Lovón-Melgarejo
2021 Wellcome Open Research  
Representations from Transformers (KeyBERT), and Rapid Automatic Keywords Extraction (Rake).  ...  NLP analysis could also be included in research of health policies and politics during general elections and provide informative summaries for the general population.  ...  .: Chexbert: combining automatic labelers and expert annotations for accurate radiology report labeling using bert . arXiv preprint arXiv: 2004.09167. 2020.  ... 
doi:10.12688/wellcomeopenres.16867.2 fatcat:f4y6smg2jbamhlk2a26leg6lgu

Improving Factual Completeness and Consistency of Image-to-Text Radiology Report Generation

Yasuhide Miura, Yuhao Zhang, Emily Tsai, Curtis Langlotz, Dan Jurafsky
2021 Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies   unpublished
Neural image-to-text radiology report generation systems offer the potential to improve radiology reporting by reducing the repetitive process of report drafting and identifying possible medical errors  ...  We combine these with the novel use of an existing semantic equivalence metric (BERTScore). We further propose a report generation system that optimizes these rewards via reinforcement learning.  ...  Acknowledgements We would like to thank the anonymous reviewers and the members of the Stanford NLP Group for their very helpful comments that substantially improved this paper.  ... 
doi:10.18653/v1/2021.naacl-main.416 fatcat:qmm7zzojzrezvjmkv6kumttkhm

Referee report. For: Government plans in the 2016 and 2021 Peruvian presidential elections: A natural language processing analysis of the health chapters [version 2; peer review: 1 approved with reservations]

Glen Dario Rodriguez Rafael
2021
and expert annotations for accurate radiology report labeling using bert.  ...  .: Chexbert: combining automatic labelers with self-labelling. 2019.  ... 
doi:10.21956/wellcomeopenres.19245.r47104 fatcat:rxhyjvdluzddfn72old4ci6wxi

Referee report. For: Government plans in the 2016 and 2021 Peruvian presidential elections: A natural language processing analysis of the health chapters [version 4; peer review: 1 approved]

Glen Dario Rodriguez Rafael
2022
Representations from Transformers (KeyBERT), and Rapid Automatic Keywords Extraction (Rake).  ...  While clinical medicine has exploded, electronic health records for Natural Language Processing (NLP) analyses, public health, and health policy research have not yet adopted these algorithms.  ...  .: Chexbert: combining automatic labelers and expert annotations for accurate radiology report labeling using bert . arXiv preprint arXiv: 2004.09167. 2020.  ... 
doi:10.21956/wellcomeopenres.19477.r48096 fatcat:732lhlszhnaize3xw2mogfegaa

Referee report. For: Government plans in the 2016 and 2021 Peruvian presidential elections: A natural language processing analysis of the health chapters [version 3; peer review: 1 approved with reservations]

Glen Dario Rodriguez Rafael
2022
Representations from Transformers (KeyBERT), and Rapid Automatic Keywords Extraction (Rake).  ...  While clinical medicine has exploded, electronic health records for Natural Language Processing (NLP) analyses, public health, and health policy research have not yet adopted these algorithms.  ...  .: Chexbert: combining automatic labelers and expert annotations for accurate radiology report labeling using bert . arXiv preprint arXiv: 2004.09167. 2020.  ... 
doi:10.21956/wellcomeopenres.19368.r47709 fatcat:gaevssgy4fhxnnh54zuam4pqa4