DCPV: A taxonomy for deep learning model in computer aided system for human age detection

Date

2023

Authors

Maskey, N.
Hameedi, S.
Dawoud, A.
Jacksi, K.
Al Sadoon, O.H.R.
Salahuddin, A.B.E.

Editors

Daimi, K.
Sadoon, A.

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Book chapter

Citation

Event/exhibition information: 2nd International Conference on Innovations in Computing Research, ICR 2023, Madrid, 04/09/2023-06/09/2023 Source details - Title: Proceedings of the Second International Conference on Innovations in Computing Research (ICR’23), 2023 / Daimi, K., Sadoon, A. (ed./s), pp.64-79

Statement of Responsibility

Conference Name

Abstract

Deep Learning prediction techniques are widely studied and researched for their implementation in Human Age Prediction (HAP) to prevent, treat and extend life expectancy. So far most of the algorithms are based on facial images, MRI scans, and DNA methylation which is used for training and testing in the domain but rarely practiced. The lack of real-world-age HAP application is caused by several factors: no significant validation and devaluation of the system in the real-world scenario, low performance, and technical complications. This paper presents the Data, Classification technique, Prediction, and View (DCPV) taxonomy which specifies the major components of the system required for the implementation of a deep learning model to predict human age. These components are to be considered and used as validation and evaluation criteria for the introduction of the deep learning HAP model. A taxonomy of the HAP system is a step towards the development of a common baseline that will help the end users and researchers to have a clear view of the constituents of deep learning prediction approaches, providing better scope for future development of similar systems in the health domain. We assess the DCPV taxonomy by considering the performance, accuracy, robustness, and model comparisons. We demonstrate the value of the DCPV taxonomy by exploring state-of-the-art research within the domain of the HAP system.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright 2023 The Author(s), under exclusive license to Springer Nature Switzerland

License

Grant ID

Call number

Persistent link to this record