Tuberculosis detection using few shot learning

Abstract Tuberculosis (TB), a contagious disease, significantly affects lungs functioning. Amongst multiple detection methodologies, Chest X-ray analysis is considered the most effective methodology. Traditional Deep Learning methodologies have shown good results for TB detection; however, model’s h...

Full description

Saved in:
Bibliographic Details
Main Authors: Kamran Riasat, Akhtar Jamil, Shaha Al-Otaibi, Sania Zeb, Saima Riasat, Shamsa Kanwal
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-97803-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849698510016872448
author Kamran Riasat
Akhtar Jamil
Shaha Al-Otaibi
Sania Zeb
Saima Riasat
Shamsa Kanwal
author_facet Kamran Riasat
Akhtar Jamil
Shaha Al-Otaibi
Sania Zeb
Saima Riasat
Shamsa Kanwal
author_sort Kamran Riasat
collection DOAJ
description Abstract Tuberculosis (TB), a contagious disease, significantly affects lungs functioning. Amongst multiple detection methodologies, Chest X-ray analysis is considered the most effective methodology. Traditional Deep Learning methodologies have shown good results for TB detection; however, model’s huge number of parameters, size, and compute requirements making it unsuitable for its practical deployment. Owing to scarce annotated datasets in medical domain augmented datasets are generated which is not a recommended technique in medical domain. This study presents TB-FSNet consisting of Few Shot Learning - Prototypical Network (FSL-PT) with a modified MobileNet-V2 backbone, incorporating a Self-Attention layer. The significant contribution of this study is to effectively train TB-FSNet in FSL-PT paradigm with six different backbones. The dataset utilised for this study consists of Montgomery County, and Shenzhen Chest X-ray Dataset combined. The proposed method attains highest accuracy of 93.6% with mere 2.21M parameters and 8.67 MB size, while maintaining high performance metrics such as precision, specificity, and sensitivity. Moreover, TB-FSNet is designed for seamless integration into embedded devices, making it suitable for deployment on edge devises. The model processes Chest X-ray images in real-time, providing immediate confidence scores for disease detection. This capability ensures that users can receive accurate diagnostic insights without needing to wait for medical professionals, enhancing the accessibility and efficiency of TB detection.
format Article
id doaj-art-01fdc1cd62a444e6a8172cc48e0d1b47
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-01fdc1cd62a444e6a8172cc48e0d1b472025-08-20T03:18:53ZengNature PortfolioScientific Reports2045-23222025-04-0115111810.1038/s41598-025-97803-9Tuberculosis detection using few shot learningKamran Riasat0Akhtar Jamil1Shaha Al-Otaibi2Sania Zeb3Saima Riasat4Shamsa Kanwal5Directorate of ICT, Allama Iqbal Open UniversityDepartment of AI & DS, FAST-NUCESDepartment of Information Systems, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman UniversityDepartment of AI & DS, FAST-NUCESDepartment of Mathematical Sciences, Fatima Jinnah Women UniversityDepartment of Mathematical Sciences, Fatima Jinnah Women UniversityAbstract Tuberculosis (TB), a contagious disease, significantly affects lungs functioning. Amongst multiple detection methodologies, Chest X-ray analysis is considered the most effective methodology. Traditional Deep Learning methodologies have shown good results for TB detection; however, model’s huge number of parameters, size, and compute requirements making it unsuitable for its practical deployment. Owing to scarce annotated datasets in medical domain augmented datasets are generated which is not a recommended technique in medical domain. This study presents TB-FSNet consisting of Few Shot Learning - Prototypical Network (FSL-PT) with a modified MobileNet-V2 backbone, incorporating a Self-Attention layer. The significant contribution of this study is to effectively train TB-FSNet in FSL-PT paradigm with six different backbones. The dataset utilised for this study consists of Montgomery County, and Shenzhen Chest X-ray Dataset combined. The proposed method attains highest accuracy of 93.6% with mere 2.21M parameters and 8.67 MB size, while maintaining high performance metrics such as precision, specificity, and sensitivity. Moreover, TB-FSNet is designed for seamless integration into embedded devices, making it suitable for deployment on edge devises. The model processes Chest X-ray images in real-time, providing immediate confidence scores for disease detection. This capability ensures that users can receive accurate diagnostic insights without needing to wait for medical professionals, enhancing the accessibility and efficiency of TB detection.https://doi.org/10.1038/s41598-025-97803-9Few Shot LearningTuberculosisSelf Attention
spellingShingle Kamran Riasat
Akhtar Jamil
Shaha Al-Otaibi
Sania Zeb
Saima Riasat
Shamsa Kanwal
Tuberculosis detection using few shot learning
Scientific Reports
Few Shot Learning
Tuberculosis
Self Attention
title Tuberculosis detection using few shot learning
title_full Tuberculosis detection using few shot learning
title_fullStr Tuberculosis detection using few shot learning
title_full_unstemmed Tuberculosis detection using few shot learning
title_short Tuberculosis detection using few shot learning
title_sort tuberculosis detection using few shot learning
topic Few Shot Learning
Tuberculosis
Self Attention
url https://doi.org/10.1038/s41598-025-97803-9
work_keys_str_mv AT kamranriasat tuberculosisdetectionusingfewshotlearning
AT akhtarjamil tuberculosisdetectionusingfewshotlearning
AT shahaalotaibi tuberculosisdetectionusingfewshotlearning
AT saniazeb tuberculosisdetectionusingfewshotlearning
AT saimariasat tuberculosisdetectionusingfewshotlearning
AT shamsakanwal tuberculosisdetectionusingfewshotlearning