Inferring Clinical Correlations from EEG Reports with Deep Neural Learning. Goodwin, T. R. & Harabagiu, S. M. In Proceedings of the American Medical Informatics Association Annual Symposium, November, 2017. ![Inferring Clinical Correlations from EEG Reports with Deep Neural Learning [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper Slides abstract bibtex "Successful diagnosis and management of neurological dysfunction relies on proper communication between the neurologist and the primary physician (or other specialists). Because this communication is documented within medical records, the ability to automatically infer the clinical correlations for a patient from his or her medical records would provide an important step towards enabling health care systems to automatically identify patients requiring additional follow-up as well as flagging any unexpected clinical correlations for review. In this paper, we present a Deep Section Recovery Model (DSRM) which applies deep neural learning on a large body of EEG reports in order to infer the expected clinical correlations for a patient from the information in a given EEG report by (1) automatically extracting word- and report- level features from the report and (2) inferring the most likely clinical correlations and expressing those clinical correlations in natural language. We evaluated the performance of the DSRM by removing the clinical correlation sections from EEG reports and measuring how well the model could recover that information from the remainder of the report. The DSRM obtained a 17% improvement over the top-performing baseline, highlighting not only the power of the DSRM but also the promise of automatically recognizing unexpected clinical correlations in the future."
@InProceedings{goodwin2017inferring,
title = {Inferring Clinical Correlations from EEG Reports with Deep Neural Learning},
author = {Goodwin, Travis R. and Harabagiu, Sanda M.},
booktitle = {Proceedings of the American Medical Informatics Association Annual Symposium},
shortbooktitle={AMIA '17},
location= {Washington, DC, USA},
year = {2017},
month = nov,
url_Paper = {papers/amia_2017a.pdf},
url_Slides = {papers/amia_2017a_slides.pdf},
abstract={"Successful diagnosis and management of neurological dysfunction relies on proper communication between the neurologist and
the primary physician (or other specialists). Because this communication is documented within medical records, the ability to
automatically infer the clinical correlations for a patient from his or her medical records would provide an important step towards
enabling health care systems to automatically identify patients requiring additional follow-up as well as flagging any unexpected
clinical correlations for review. In this paper, we present a Deep Section Recovery Model (DSRM) which applies deep neural
learning on a large body of EEG reports in order to infer the expected clinical correlations for a patient from the information in a
given EEG report by (1) automatically extracting word- and report- level features from the report and (2) inferring the most likely
clinical correlations and expressing those clinical correlations in natural language. We evaluated the performance of the DSRM
by removing the clinical correlation sections from EEG reports and measuring how well the model could recover that information
from the remainder of the report. The DSRM obtained a 17\% improvement over the top-performing baseline, highlighting not
only the power of the DSRM but also the promise of automatically recognizing unexpected clinical correlations in the future."},
keywords = {deep learning, natural language processing, natural language generation, medical informatics, electroencephalogram reports}
}
Downloads: 0
{"_id":"uEztNoJdDTncJkp3e","bibbaseid":"goodwin-harabagiu-inferringclinicalcorrelationsfromeegreportswithdeepneurallearning-2017","downloads":0,"creationDate":"2017-12-12T18:20:31.344Z","title":"Inferring Clinical Correlations from EEG Reports with Deep Neural Learning","author_short":["Goodwin, T. R.","Harabagiu, S. M."],"year":2017,"bibtype":"inproceedings","biburl":"http://www.hlt.utdallas.edu/~travis/publications.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Inferring Clinical Correlations from EEG Reports with Deep Neural Learning","author":[{"propositions":[],"lastnames":["Goodwin"],"firstnames":["Travis","R."],"suffixes":[]},{"propositions":[],"lastnames":["Harabagiu"],"firstnames":["Sanda","M."],"suffixes":[]}],"booktitle":"Proceedings of the American Medical Informatics Association Annual Symposium","shortbooktitle":"AMIA '17","location":"Washington, DC, USA","year":"2017","month":"November","url_paper":"papers/amia_2017a.pdf","url_slides":"papers/amia_2017a_slides.pdf","abstract":"\"Successful diagnosis and management of neurological dysfunction relies on proper communication between the neurologist and the primary physician (or other specialists). Because this communication is documented within medical records, the ability to automatically infer the clinical correlations for a patient from his or her medical records would provide an important step towards enabling health care systems to automatically identify patients requiring additional follow-up as well as flagging any unexpected clinical correlations for review. In this paper, we present a Deep Section Recovery Model (DSRM) which applies deep neural learning on a large body of EEG reports in order to infer the expected clinical correlations for a patient from the information in a given EEG report by (1) automatically extracting word- and report- level features from the report and (2) inferring the most likely clinical correlations and expressing those clinical correlations in natural language. We evaluated the performance of the DSRM by removing the clinical correlation sections from EEG reports and measuring how well the model could recover that information from the remainder of the report. The DSRM obtained a 17% improvement over the top-performing baseline, highlighting not only the power of the DSRM but also the promise of automatically recognizing unexpected clinical correlations in the future.\"","keywords":"deep learning, natural language processing, natural language generation, medical informatics, electroencephalogram reports","bibtex":"@InProceedings{goodwin2017inferring,\ntitle = {Inferring Clinical Correlations from EEG Reports with Deep Neural Learning},\nauthor = {Goodwin, Travis R. and Harabagiu, Sanda M.},\nbooktitle = {Proceedings of the American Medical Informatics Association Annual Symposium},\nshortbooktitle={AMIA '17},\nlocation= {Washington, DC, USA},\nyear = {2017},\nmonth = nov,\nurl_Paper = {papers/amia_2017a.pdf},\nurl_Slides = {papers/amia_2017a_slides.pdf},\n abstract={\"Successful diagnosis and management of neurological dysfunction relies on proper communication between the neurologist and\nthe primary physician (or other specialists). Because this communication is documented within medical records, the ability to\nautomatically infer the clinical correlations for a patient from his or her medical records would provide an important step towards\nenabling health care systems to automatically identify patients requiring additional follow-up as well as flagging any unexpected\nclinical correlations for review. In this paper, we present a Deep Section Recovery Model (DSRM) which applies deep neural\nlearning on a large body of EEG reports in order to infer the expected clinical correlations for a patient from the information in a\ngiven EEG report by (1) automatically extracting word- and report- level features from the report and (2) inferring the most likely\nclinical correlations and expressing those clinical correlations in natural language. We evaluated the performance of the DSRM\nby removing the clinical correlation sections from EEG reports and measuring how well the model could recover that information\nfrom the remainder of the report. The DSRM obtained a 17\\% improvement over the top-performing baseline, highlighting not\nonly the power of the DSRM but also the promise of automatically recognizing unexpected clinical correlations in the future.\"},\nkeywords = {deep learning, natural language processing, natural language generation, medical informatics, electroencephalogram reports}\n}\n","author_short":["Goodwin, T. R.","Harabagiu, S. M."],"key":"goodwin2017inferring","id":"goodwin2017inferring","bibbaseid":"goodwin-harabagiu-inferringclinicalcorrelationsfromeegreportswithdeepneurallearning-2017","role":"author","urls":{" paper":"http://www.hlt.utdallas.edu/~travis/papers/amia_2017a.pdf"," slides":"http://www.hlt.utdallas.edu/~travis/papers/amia_2017a_slides.pdf"},"keyword":["deep learning","natural language processing","natural language generation","medical informatics","electroencephalogram reports"],"downloads":0,"html":""},"search_terms":["inferring","clinical","correlations","eeg","reports","deep","neural","learning","goodwin","harabagiu"],"keywords":["deep learning","natural language processing","natural language generation","medical informatics","electroencephalogram reports"],"authorIDs":["5a2f21809abaa8bf3500003f"],"dataSources":["ea3q84ZXjZvjBmaC4"]}