A discussion of anemia evaluation method using eyelid conjunctival images

Katsuto Motoyama; Yusuke Takagi; Kazuhito Murakami; Mikiko Shimizu

doi:10.1117/12.3018238

2 May 2024 A discussion of anemia evaluation method using eyelid conjunctival images

Katsuto Motoyama, Yusuke Takagi, Kazuhito Murakami, Mikiko Shimizu

Proceedings Volume 13164, International Workshop on Advanced Imaging Technology (IWAIT) 2024; 131640W (2024) https://doi.org/10.1117/12.3018238
Event: International Workshop on Advanced Imaging Technology (IWAIT) 2024, 2024, Langkawi, Malaysia

Abstract

Anemia is one of the most common social problem, so much so that one in ten people is said to be anemic. Anemia can be diagnosed by observing the coloration of the eyelid conjunctiva or by drawing blood. However, both cases require specialized knowledge and are not easy methods in terms of time and cost. This paper proposes a new method for estimating anemia from facial images. First, the eyelid conjunctiva region is extracted using a cascade classifier. Next, the obtained images were input into a model using a convolutional neural network (CNN), which had already learned the characteristics of anemia, to create a system that automatically estimates the state of anemia. Compared to a model using the standard multilayer perceptron (MLP), the standard MLP-based model had anemia estimation accuracy of 66.7%, while the CNN-based model had an accuracy of 87.5%. The results confirmed the effectiveness of the proposed method.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Katsuto Motoyama, Yusuke Takagi, Kazuhito Murakami, and Mikiko Shimizu "A discussion of anemia evaluation method using eyelid conjunctival images", Proc. SPIE 13164, International Workshop on Advanced Imaging Technology (IWAIT) 2024, 131640W (2 May 2024); https://doi.org/10.1117/12.3018238

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available