Skip to main content

Hybrid Architecture Based Intelligent Diagnosis Assistant for GP.

Wang, R., Jayathunge, K., Page, R., Li, H., Zhang, J. J. and Yang, X., 2024. Hybrid Architecture Based Intelligent Diagnosis Assistant for GP. BMC Medical Informatics and Decision Making, 24, 15.

Full text available as:

[img]
Preview
PDF (OPEN ACCESS ARTICLE)
s12911-023-02398-8.pdf - Published Version
Available under License Creative Commons Attribution.

3MB
[img] PDF
Copy of the manuscript.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial.

1MB

DOI: 10.1186/s12911-023-02398-8

Abstract

As the first point of contact for patients, General Practitioners (GPs) play a crucial role in the National Health Service (NHS). Anaccurate primary diagnosis from the GP can alleviate the burden on specialists and reduce the time needed to re-confirm the patient’scondition, allowing for more efficient further examinations. However, GPs have broad but less specialized knowledge, which limits theaccuracy of their diagnosis. Therefore, it is imperative to introduce an intelligent system to assist GPs in making decisions. This paperintroduces two data augmentation methods, the Complaint Symptoms Integration Method and Symptom Dot Separating Method, tointegrate essential information into the Integration dataset. Additionally, it proposes a hybrid architecture that fuses the features ofwords from different representation spaces. Experiments demonstrate that, compared to commonly used pre-trained attention-basedmodels, our hybrid architecture delivers the best classification performance for four common neurological diseases on the enhancedIntegration dataset. For example, the classification accuracy of the BERT+CNN hybrid architecture is 0.897, which is a 5.1%improvement over both BERT and CNN with 0.846. Finally, this paper develops an AI diagnosis assistant web application thatleverages the superior performance of this architecture to help GPs complete primary diagnosis efficiently and accurately.

Item Type:Article
ISSN:1472-6947
Uncontrolled Keywords:GP; Referral letter; Primary diagnosis; Hybrid architecture; Text classification; AI diagnosis assistant
Group:Faculty of Media & Communication
ID Code:39326
Deposited By: Symplectic RT2
Deposited On:05 Jan 2024 14:06
Last Modified:12 Jan 2024 11:06

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -