An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes

Diabetes is a chronic disease due to higher blood sugar (or Glucose) levels in the blood. This study proposes a novel attention-based loss function and a lightweight artificial neural network (ANN) called Diabetic Lite (DB-Lite) for diabetes prediction in the Pima Indian Diabetes Dataset (PIDD). We...

Full description

Saved in:
Bibliographic Details
Main Authors: Santanu Roy, Reshma Rachel Cherish, Gifty Roy
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Healthcare Analytics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772442525000188
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850118084498554880
author Santanu Roy
Reshma Rachel Cherish
Gifty Roy
author_facet Santanu Roy
Reshma Rachel Cherish
Gifty Roy
author_sort Santanu Roy
collection DOAJ
description Diabetes is a chronic disease due to higher blood sugar (or Glucose) levels in the blood. This study proposes a novel attention-based loss function and a lightweight artificial neural network (ANN) called Diabetic Lite (DB-Lite) for diabetes prediction in the Pima Indian Diabetes Dataset (PIDD). We show that the Pima dataset has many challenges. It is a small and imbalanced dataset; moreover, many features are non-linearly correlated in this dataset. The novelties of this research work are as follows: (i) A novel loss function of attention-based binary cross entropy (ABCE) is proposed for the first time to alleviate the statistical imbalance present within the Pima dataset. This ABCE loss function is incorporated in the DB-Lite model, which is trained from scratch. (ii) A Swish activation function is deployed in the hidden layer of DB-Lite instead of Rectified Linear Unit (ReLU) to deal with the non-linear dependency of features with the final outcome. (iii) The synthetic minority oversampling technique (SMOTE) is used as a pre-processing technique to mitigate the class imbalance problem from the Pima dataset. (iv) An adaptive learning rate is utilized while training the model to speed up the convergence of the DB-Lite model. Our final proposed framework has achieved 99.7% accuracy, 99.4% precision, 99.8% recall, and 99.6% F1 score in testing, which is the best result on this Pima dataset. The Welch t-testing (as a statistical hypothesis testing) and 10-fold cross-validation are utilized to prove the validity of the proposed loss function.
format Article
id doaj-art-8a868236b0dc416c9fec6992d172cf36
institution OA Journals
issn 2772-4425
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Healthcare Analytics
spelling doaj-art-8a868236b0dc416c9fec6992d172cf362025-08-20T02:35:57ZengElsevierHealthcare Analytics2772-44252025-06-01710039910.1016/j.health.2025.100399An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetesSantanu Roy0Reshma Rachel Cherish1Gifty Roy2Pandit Deendayal Energy University (PDEU), Department of Computer Science and Engineering, Gandhinagar, India; Corresponding author.Ramaiah Institute of Technology, Department of Computer Science and Engineering, Bangalore, IndiaChrist (Deemed to be University), Department of Computer Science and Engineering, Bangalore, IndiaDiabetes is a chronic disease due to higher blood sugar (or Glucose) levels in the blood. This study proposes a novel attention-based loss function and a lightweight artificial neural network (ANN) called Diabetic Lite (DB-Lite) for diabetes prediction in the Pima Indian Diabetes Dataset (PIDD). We show that the Pima dataset has many challenges. It is a small and imbalanced dataset; moreover, many features are non-linearly correlated in this dataset. The novelties of this research work are as follows: (i) A novel loss function of attention-based binary cross entropy (ABCE) is proposed for the first time to alleviate the statistical imbalance present within the Pima dataset. This ABCE loss function is incorporated in the DB-Lite model, which is trained from scratch. (ii) A Swish activation function is deployed in the hidden layer of DB-Lite instead of Rectified Linear Unit (ReLU) to deal with the non-linear dependency of features with the final outcome. (iii) The synthetic minority oversampling technique (SMOTE) is used as a pre-processing technique to mitigate the class imbalance problem from the Pima dataset. (iv) An adaptive learning rate is utilized while training the model to speed up the convergence of the DB-Lite model. Our final proposed framework has achieved 99.7% accuracy, 99.4% precision, 99.8% recall, and 99.6% F1 score in testing, which is the best result on this Pima dataset. The Welch t-testing (as a statistical hypothesis testing) and 10-fold cross-validation are utilized to prove the validity of the proposed loss function.http://www.sciencedirect.com/science/article/pii/S2772442525000188Artificial neural networkDiabetes predictionClass imbalanceAttention-based binary cross entropySynthetic minority oversampling technique (SMOTE)
spellingShingle Santanu Roy
Reshma Rachel Cherish
Gifty Roy
An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
Healthcare Analytics
Artificial neural network
Diabetes prediction
Class imbalance
Attention-based binary cross entropy
Synthetic minority oversampling technique (SMOTE)
title An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
title_full An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
title_fullStr An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
title_full_unstemmed An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
title_short An attention-based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
title_sort attention based loss function and synthetic minority oversampling technique for alleviating class imbalance in predicting diabetes
topic Artificial neural network
Diabetes prediction
Class imbalance
Attention-based binary cross entropy
Synthetic minority oversampling technique (SMOTE)
url http://www.sciencedirect.com/science/article/pii/S2772442525000188
work_keys_str_mv AT santanuroy anattentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes
AT reshmarachelcherish anattentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes
AT giftyroy anattentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes
AT santanuroy attentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes
AT reshmarachelcherish attentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes
AT giftyroy attentionbasedlossfunctionandsyntheticminorityoversamplingtechniqueforalleviatingclassimbalanceinpredictingdiabetes