Medical image classification by incorporating clinical variables and learned features

Medical image classification plays an important role in medical imaging. In this work, we present a novel approach to enhance deep learning models in medical image classification by incorporating clinical variables without overwhelming the information. Unlike most existing deep neural network models...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiahui Liu, Xiaohao Cai, Mahesan Niranjan
Format: Article
Language:English
Published: The Royal Society 2025-03-01
Series:Royal Society Open Science
Subjects:
Online Access:https://royalsocietypublishing.org/doi/10.1098/rsos.241222
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850041503413436416
author Jiahui Liu
Xiaohao Cai
Mahesan Niranjan
author_facet Jiahui Liu
Xiaohao Cai
Mahesan Niranjan
author_sort Jiahui Liu
collection DOAJ
description Medical image classification plays an important role in medical imaging. In this work, we present a novel approach to enhance deep learning models in medical image classification by incorporating clinical variables without overwhelming the information. Unlike most existing deep neural network models that only consider single-pixel information, our method captures a more comprehensive view. Our method contains two main steps and is effective in tackling the extra challenge raised by the scarcity of medical data. Firstly, we employ a pre-trained deep neural network served as a feature extractor to capture meaningful image features. Then, an exquisite discriminant analysis is applied to reduce the dimensionality of these features, ensuring that the low number of features remains optimized for the classification task and striking a balance with the clinical variables information. We also develop a way of obtaining class activation maps for our approach in visualizing models’ focus on specific regions within the low-dimensional feature space. Thorough experimental results demonstrate improvements of our proposed method over state-of-the-art methods for tuberculosis and dermatology issues for example. Furthermore, a comprehensive comparison with a popular dimensionality reduction technique (principal component analysis) is also conducted.
format Article
id doaj-art-8254dd712efd4129947a4a67ef4bf577
institution DOAJ
issn 2054-5703
language English
publishDate 2025-03-01
publisher The Royal Society
record_format Article
series Royal Society Open Science
spelling doaj-art-8254dd712efd4129947a4a67ef4bf5772025-08-20T02:55:45ZengThe Royal SocietyRoyal Society Open Science2054-57032025-03-0112310.1098/rsos.241222Medical image classification by incorporating clinical variables and learned featuresJiahui Liu0Xiaohao Cai1Mahesan Niranjan2School of Electronics and Computer Science, University of Southampton, Southampton, UKSchool of Electronics and Computer Science, University of Southampton, Southampton, UKSchool of Electronics and Computer Science, University of Southampton, Southampton, UKMedical image classification plays an important role in medical imaging. In this work, we present a novel approach to enhance deep learning models in medical image classification by incorporating clinical variables without overwhelming the information. Unlike most existing deep neural network models that only consider single-pixel information, our method captures a more comprehensive view. Our method contains two main steps and is effective in tackling the extra challenge raised by the scarcity of medical data. Firstly, we employ a pre-trained deep neural network served as a feature extractor to capture meaningful image features. Then, an exquisite discriminant analysis is applied to reduce the dimensionality of these features, ensuring that the low number of features remains optimized for the classification task and striking a balance with the clinical variables information. We also develop a way of obtaining class activation maps for our approach in visualizing models’ focus on specific regions within the low-dimensional feature space. Thorough experimental results demonstrate improvements of our proposed method over state-of-the-art methods for tuberculosis and dermatology issues for example. Furthermore, a comprehensive comparison with a popular dimensionality reduction technique (principal component analysis) is also conducted.https://royalsocietypublishing.org/doi/10.1098/rsos.241222medical imagingclassificationdiscriminant analysisclinical variablesclass activation map
spellingShingle Jiahui Liu
Xiaohao Cai
Mahesan Niranjan
Medical image classification by incorporating clinical variables and learned features
Royal Society Open Science
medical imaging
classification
discriminant analysis
clinical variables
class activation map
title Medical image classification by incorporating clinical variables and learned features
title_full Medical image classification by incorporating clinical variables and learned features
title_fullStr Medical image classification by incorporating clinical variables and learned features
title_full_unstemmed Medical image classification by incorporating clinical variables and learned features
title_short Medical image classification by incorporating clinical variables and learned features
title_sort medical image classification by incorporating clinical variables and learned features
topic medical imaging
classification
discriminant analysis
clinical variables
class activation map
url https://royalsocietypublishing.org/doi/10.1098/rsos.241222
work_keys_str_mv AT jiahuiliu medicalimageclassificationbyincorporatingclinicalvariablesandlearnedfeatures
AT xiaohaocai medicalimageclassificationbyincorporatingclinicalvariablesandlearnedfeatures
AT mahesanniranjan medicalimageclassificationbyincorporatingclinicalvariablesandlearnedfeatures