Development and Evaluation of Novel Ophthalmology Domain-specific Neural Word Embeddings to Predict Visual Prognosis

Overview

Journal Int J Med Inform

Specialty Medical Informatics

Date 2021 Apr 23

PMID 33892445

Citations 8

Authors

Sophia Wang

Benjamin Tseng

Tina Hernandez-Boussard

Affiliations

Soon will be listed here.

Abstract

Objective: To develop and evaluate novel word embeddings (WEs) specific to ophthalmology, using text corpora from published literature and electronic health records (EHR).

Materials And Methods: We trained ophthalmology-specific WEs using 121,740 PubMed abstracts and 89,282 EHR notes using word2vec continuous bag-of-words architecture. PubMed and EHR WEs were compared to general domain GloVe WEs and general biomedical domain BioWordVec embeddings using a novel ophthalmology-domain-specific 200-question analogy test and prediction of prognosis in 5547 low vision patients using EHR notes as inputs to a deep learning model.

Results: We found that many words representing important ophthalmic concepts in the EHR were missing from the general domain GloVe vocabulary, but covered in the ophthalmology abstract corpus. On ophthalmology analogy testing, PubMed WEs scored 95.0 %, outperforming EHR (86.0 %) and GloVe (91.0 %) but less than BioWordVec (99.5 %). On predicting low vision prognosis, PubMed and EHR WEs resulted in similar AUROC (0.830; 0.826), outperforming GloVe (0.778) and BioWordVec (0.784).

Conclusion: We found that using ophthalmology domain-specific WEs improved performance in ophthalmology-related clinical prediction compared to general WEs. Deep learning models using clinical notes as inputs can predict the prognosis of visually impaired patients. This work provides a framework to improve predictive models using domain-specific WEs.

Citing Articles

Visual acuity prediction on real-life patient data using a machine learning based multistage system.

Schlosser T, Beuth F, Meyer T, Kumar A, Stolze G, Furashova O Sci Rep. 2024; 14(1):5532.

PMID: 38448469 PMC: 10917755. DOI: 10.1038/s41598-024-54482-2.

Use of artificial intelligence in forecasting glaucoma progression.

Thakur S, Dinh L, Lavanya R, Quek T, Liu Y, Cheng C Taiwan J Ophthalmol. 2023; 13(2):168-183.

PMID: 37484617 PMC: 10361424. DOI: 10.4103/tjo.TJO-D-23-00022.

Predicting near-term glaucoma progression: An artificial intelligence approach using clinical free-text notes and data from electronic health records.

Jalamangala Shivananjaiah S, Kumari S, Majid I, Wang S Front Med (Lausanne). 2023; 10:1157016.

PMID: 37122330 PMC: 10133544. DOI: 10.3389/fmed.2023.1157016.

Impact of word embedding models on text analytics in deep learning environment: a review.

Asudani D, Nagwani N, Singh P Artif Intell Rev. 2023; :1-81.

PMID: 36844886 PMC: 9944441. DOI: 10.1007/s10462-023-10419-1.

Deep Learning Approaches for Predicting Glaucoma Progression Using Electronic Health Records and Natural Language Processing.

Wang S, Tseng B, Hernandez-Boussard T Ophthalmol Sci. 2022; 2(2):100127.

PMID: 36249690 PMC: 9559076. DOI: 10.1016/j.xops.2022.100127.

References

Ali N, Khan A, Akunjee M, Ahfat F . Using common ophthalmologic jargon in correspondence can lead to miscommunication. Br J Gen Pract. 2006; 56(533):968-9. PMC: 1934059. View

Zhang Y, Chen Q, Yang Z, Lin H, Lu Z . BioWordVec, improving biomedical word embeddings with subword information and MeSH. Sci Data. 2019; 6(1):52. PMC: 6510737. DOI: 10.1038/s41597-019-0055-0. View

Wang S, Pershing S, Tran E, Hernandez-Boussard T . Automated extraction of ophthalmic surgery outcomes from the electronic health record. Int J Med Inform. 2019; 133:104007. DOI: 10.1016/j.ijmedinf.2019.104007. View

Wen J, Lee C, Keane P, Xiao S, Rokem A, Chen P . Forecasting future Humphrey Visual Fields using deep learning. PLoS One. 2019; 14(4):e0214875. PMC: 6450620. DOI: 10.1371/journal.pone.0214875. View

Lee J, Yoon W, Kim S, Kim D, Kim S, So C . BioBERT: a pre-trained biomedical language representation model for biomedical text mining. Bioinformatics. 2019; 36(4):1234-1240. PMC: 7703786. DOI: 10.1093/bioinformatics/btz682. View

Hamiel U, Hecht I, Nemet A, Peer L, Man V, Hilely A . Frequency, comprehension and attitudes of physicians towards abbreviations in the medical record. Postgrad Med J. 2018; 94(1111):254-258. DOI: 10.1136/postgradmedj-2017-135515. View

Lowe H, Ferris T, Hernandez P, Weber S . STRIDE--An integrated standards-based translational research informatics platform. AMIA Annu Symp Proc. 2010; 2009:391-5. PMC: 2815452. View

Zhang Y, Marshall I, Wallace B . Rationale-Augmented Convolutional Neural Networks for Text Classification. Proc Conf Empir Methods Nat Lang Process. 2017; 2016:795-804. PMC: 5300751. DOI: 10.18653/v1/d16-1076. View

Berchuck S, Mukherjee S, Medeiros F . Estimating Rates of Progression and Predicting Future Visual Fields in Glaucoma Using a Deep Variational Autoencoder. Sci Rep. 2019; 9(1):18113. PMC: 6888896. DOI: 10.1038/s41598-019-54653-6. View

10.

Banerjee I, Gensheimer M, Wood D, Henry S, Aggarwal S, Chang D . Probabilistic Prognostic Estimates of Survival in Metastatic Cancer Patients (PPES-Met) Utilizing Free-Text Clinical Narratives. Sci Rep. 2018; 8(1):10037. PMC: 6030075. DOI: 10.1038/s41598-018-27946-5. View

11.

Coker M, Huisingh C, McGwin Jr G, Read R, Swanson M, Dreer L . Rehabilitation Referral for Patients With Irreversible Vision Impairment Seen in a Public Safety-Net Eye Clinic. JAMA Ophthalmol. 2018; 136(4):400-408. PMC: 5876806. DOI: 10.1001/jamaophthalmol.2018.0241. View

12.

Wang M, Shen L, Pasquale L, Petrakos P, Formica S, Boland M . An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis. Invest Ophthalmol Vis Sci. 2019; 60(1):365-375. PMC: 6348996. DOI: 10.1167/iovs.18-25568. View

13.

Banerjee I, Bozkurt S, Caswell-Jin J, Kurian A, Rubin D . Natural Language Processing Approaches to Detect the Timeline of Metastatic Recurrence of Breast Cancer. JCO Clin Cancer Inform. 2019; 3:1-12. DOI: 10.1200/CCI.19.00034. View

14.

Wang Y, Liu S, Afzal N, Rastegar-Mojarad M, Wang L, Shen F . A comparison of word embeddings for the biomedical natural language processing. J Biomed Inform. 2018; 87:12-20. PMC: 6585427. DOI: 10.1016/j.jbi.2018.09.008. View

15.

Gu Y, Leroy G, Pettygrove S, Galindo M, Kurzius-Spencer M . Optimizing Corpus Creation for Training Word Embedding in Low Resource Domains: A Case Study in Autism Spectrum Disorder (ASD). AMIA Annu Symp Proc. 2019; 2018:508-517. PMC: 6371367. View

16.

Fontenot J, Bona M, Kaleem M, McLaughlin Jr W, Morse A, Schwartz T . Vision Rehabilitation Preferred Practice Pattern®. Ophthalmology. 2017; 125(1):P228-P278. DOI: 10.1016/j.ophtha.2017.09.030. View

17.

Garcia G, Nitta K, Lavieri M, Andrews C, Liu X, Lobaza E . Using Kalman Filtering to Forecast Disease Trajectory for Patients With Normal Tension Glaucoma. Am J Ophthalmol. 2018; 199:111-119. PMC: 6662653. DOI: 10.1016/j.ajo.2018.10.012. View

18.

Baxter S, Marks C, Kuo T, Ohno-Machado L, Weinreb R . Machine Learning-Based Predictive Modeling of Surgical Intervention in Glaucoma Using Systemic Data From Electronic Health Records. Am J Ophthalmol. 2019; 208:30-40. PMC: 6888922. DOI: 10.1016/j.ajo.2019.07.005. View