Time series classification based correlational neural network with bidirectional LSTM for automated detection of kidney disease
- Submitting institution
-
Kingston University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-47-1385
- Type
- D - Journal article
- DOI
-
10.1109/JSEN.2020.3028738
- Title of journal
- IEEE Sensors Journal
- Article number
- -
- First page
- 4811
- Volume
- 21
- Issue
- -
- ISSN
- 1530-437X
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
-
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper proposes a novel non-invasive analytical solution to detect Chronic Kidney Disease (CKD) through analysis of human saliva samples. The analytical solution includes an improved hybrid deep learning model that combines both a one-dimensional Correlational Neural Network (1-D CorrNN) and a bidirectional Long Short-Term Memory network. The proposed model was trained and validated with a time-series dataset of 104 individuals collected via a CKD sensing module developed in this project. The results show that the proposed detection module and classification algorithm provides more accurate predictions (98%) than conventional methods.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -