Improved bio-inspired with machine learning computing approach for thyroid prediction

Thyroid illness is widely recognised as a prevalent health condition that can result in a range of health disorders. Thyroid illnesses, namely hypothyroidism and hyperthyroidism, are widespread worldwide and present considerable health consequences. These conditions have a particularly noticeable ef...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Scientific reports Jg. 15; H. 1; S. 22524 - 26
Hauptverfasser: Kesavulu, Divya, R, Kannadasan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London Nature Publishing Group UK 02.07.2025
Nature Portfolio
Schlagworte:
639/166
692/700
Decision Trees
Deep Learning
Feature selection
Female
Humanities and Social Sciences
Humans
Machine Learning
multidisciplinary
Neural Networks, Computer
Optimisation techniques
Particle snake swarm optimization
Science
Science (multidisciplinary)
Support Vector Machine
Thyroid
Thyroid Diseases - diagnosis
Thyroid Gland - pathology
Deep learning
DT
Feature selection
RF
Optimisation techniques
CNN-LSTM
Machine learning
KNN
SVM
Particle snake swarm optimization
Thyroid
ISSN:2045-2322, 2045-2322
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Thyroid illness is widely recognised as a prevalent health condition that can result in a range of health disorders. Thyroid illnesses, namely hypothyroidism and hyperthyroidism, are widespread worldwide and present considerable health consequences. These conditions have a particularly noticeable effect on women in places such as Asia, Latin America, and Africa. Precise and prompt identification of thyroid diseases is essential because of their significant impact on metabolism and general well-being. This study investigates the use of several machine learning (ML) and deep learning (DL) methods, such as random forest (RF), decision tree, SVM, and KNN, to improve the precision of predicting thyroid illnesses. The objective is to enhance the performance of these models by using more sophisticated optimization methods like particle snake swarm optimization (PSSO). The results of the analysis, assessed using parameters such as accuracy, recall, precision, F1-score, and Specificity, clearly show substantial improvements in predictive capability. The random forest with PSSO model attained an accuracy of 98.7%, an F1-score of 98.47%, a Precision of 98.51%, a Recall of 98.7%, and a Specificity of 98%. Notably, PSSO-RF outperformed a CNN-LSTM deep learning baseline by 2.98% (95.72%) in accuracy, highlighting the effectiveness of bio-inspired optimisation in improving conventional machine learning models. This research adds to the increasing data of research that uses computational innovations to improve healthcare outcomes. The results indicate that by using RF with PSSO optimiser, an accuracy score of 98.7% can be achieved. This performance surpasses that of current state-of-the-art models.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-025-03299-8