Literature DB >> 35316697

A hybrid model to identify fall occurrence from electronic health records.

Sunyang Fu1, Bjoerg Thorsteinsdottir2, Xin Zhang2, Guilherme S Lopes3, Sandeep R Pagali2, Nathan K LeBrasseur4, Andrew Wen5, Hongfang Liu5, Walter A Rocca6, Janet E Olson3, Jennifer St Sauver3, Sunghwan Sohn7.   

Abstract

INTRODUCTION: Falls are a leading cause of unintentional injury in the elderly. Electronic health records (EHRs) offer the unique opportunity to develop models that can identify fall events. However, identifying fall events in clinical notes requires advanced natural language processing (NLP) to simultaneously address multiple issues because the word "fall" is a typical homonym.
METHODS: We implemented a context-aware language model, Bidirectional Encoder Representations from Transformers (BERT) to identify falls from the EHR text and further fused the BERT model into a hybrid architecture coupled with post-hoc heuristic rules to enhance the performance. The models were evaluated on real world EHR data and were compared to conventional rule-based and deep learning models (CNN and Bi-LSTM). To better understand the ability of each approach to identify falls, we further categorize fall-related concepts (i.e., risk of fall, prevention of fall, homonym) and performed a detailed error analysis.
RESULTS: The hybrid model achieved the highest f1-score on sentence (0.971), document (0.985), and patient (0.954) level. At the sentence level (basic data unit in the model), the hybrid model had 0.954, 1.000, 0.988, and 0.999 in sensitivity, specificity, positive predictive value, and negative predictive value, respectively. The error analysis showed that that machine learning-based approaches demonstrated higher performance than a rule-based approach in challenging cases that required contextual understanding. The context-aware language model (BERT) slightly outperformed the word embedding approach trained on Bi-LSTM. No single model yielded the best performance for all fall-related semantic categories.
CONCLUSION: A context-aware language model (BERT) was able to identify challenging fall events that requires context understanding in EHR free text. The hybrid model combined with post-hoc rules allowed a custom fix on the BERT outcomes and further improved the performance of fall detection.
Copyright © 2022 Elsevier B.V. All rights reserved.

Entities:  

Keywords:  BERT; EHR; Fall; NLP

Year:  2022        PMID: 35316697      PMCID: PMC9448825          DOI: 10.1016/j.ijmedinf.2022.104736

Source DB:  PubMed          Journal:  Int J Med Inform        ISSN: 1386-5056            Impact factor:   4.730


  18 in total

Review 1.  Defining a fall and reasons for falling: comparisons among the views of seniors, health care providers, and the research literature.

Authors:  Aleksandra A Zecevic; Alan W Salmoni; Mark Speechley; Anthony A Vandervoort
Journal:  Gerontologist       Date:  2006-06

2.  The Mayo Clinic Biobank: a building block for individualized medicine.

Authors:  Janet E Olson; Euijung Ryu; Kiley J Johnson; Barbara A Koenig; Karen J Maschke; Jody A Morrisette; Mark Liebow; Paul Y Takahashi; Zachary S Fredericksen; Ruchi G Sharma; Kari S Anderson; Matthew A Hathcock; Jason A Carnahan; Jyotishman Pathak; Noralane M Lindor; Timothy J Beebe; Stephen N Thibodeau; James R Cerhan
Journal:  Mayo Clin Proc       Date:  2013-09       Impact factor: 7.616

3.  Gait variability and fall risk in community-living older adults: a 1-year prospective study.

Authors:  J M Hausdorff; D A Rios; H K Edelberg
Journal:  Arch Phys Med Rehabil       Date:  2001-08       Impact factor: 3.966

Review 4.  Implementation of multifactorial interventions for fall and fracture prevention.

Authors:  A John Campbell; M Clare Robertson
Journal:  Age Ageing       Date:  2006-09       Impact factor: 10.668

5.  Prevention of falls in the elderly trial (PROFET): a randomised controlled trial.

Authors:  J Close; M Ellis; R Hooper; E Glucksman; S Jackson; C Swift
Journal:  Lancet       Date:  1999-01-09       Impact factor: 79.321

6.  Finding falls in ambulatory care clinical documents using statistical text mining.

Authors:  James A McCart; Donald J Berndt; Jay Jarman; Dezon K Finch; Stephen L Luther
Journal:  J Am Med Inform Assoc       Date:  2012-12-15       Impact factor: 4.497

Review 7.  Clinical concept extraction: A methodology review.

Authors:  Sunyang Fu; David Chen; Huan He; Sijia Liu; Sungrim Moon; Kevin J Peterson; Feichen Shen; Liwei Wang; Yanshan Wang; Andrew Wen; Yiqing Zhao; Sunghwan Sohn; Hongfang Liu
Journal:  J Biomed Inform       Date:  2020-08-06       Impact factor: 6.317

8.  Development and validation of a pragmatic natural language processing approach to identifying falls in older adults in the emergency department.

Authors:  Brian W Patterson; Gwen C Jacobsohn; Manish N Shah; Yiqiang Song; Apoorva Maru; Arjun K Venkatesh; Monica Zhong; Katherine Taylor; Azita G Hamedani; Eneida A Mendonça
Journal:  BMC Med Inform Decis Mak       Date:  2019-07-22       Impact factor: 2.796

9.  Characteristics and utilisation of the Mayo Clinic Biobank, a clinic-based prospective collection in the USA: cohort profile.

Authors:  Janet E Olson; Euijung Ryu; Matthew A Hathcock; Ruchi Gupta; Joshua T Bublitz; Paul Y Takahashi; Suzette J Bielinski; Jennifer L St Sauver; Karen Meagher; Richard R Sharp; Stephen N Thibodeau; Mine Cicek; James R Cerhan
Journal:  BMJ Open       Date:  2019-11-06       Impact factor: 2.692

10.  Assessment of the impact of EHR heterogeneity for clinical research through a case study of silent brain infarction.

Authors:  Sunyang Fu; Lester Y Leung; Anne-Olivia Raulli; David F Kallmes; Kristin A Kinsman; Kristoff B Nelson; Michael S Clark; Patrick H Luetmer; Paul R Kingsbury; David M Kent; Hongfang Liu
Journal:  BMC Med Inform Decis Mak       Date:  2020-03-30       Impact factor: 2.796

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.