SSDM:A Self Supervised Diffusion Model For Lung Anomaly Detection Using Chest X-rays

Shemy Syed; R Elakkiya; N. E. Pears

SSDM:A Self Supervised Diffusion Model For Lung Anomaly Detection Using Chest X-rays

Shemy Syed, R Elakkiya, N. E. Pears

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Self supervised learning is emerging very quickly in computer vision tasks, which addresses the scarcity of annotated medical images . We introduce a self-supervised approach for
anomaly detection using a Diffusion Probabilistic Model, where the model is trained exclusively on normal chest X-rays and serves as a baseline for identifying anomalies. We implemented a U-net-based Gaussian diffusion model(SSDM) that adds noise in an iterative manner to images and learns to generate them in reverse. Early detection of disease is critical for medical diagnosis.
The empirical results show that the model we put forward has good accuracy in detecting the lung anomaly and thereby helps to diagnose disease at an early stage.

Original language	English
Title of host publication	International Conference on Communication, Computing, Networking, and Control in Cyber-Physical Systems
Subtitle of host publication	CCNCPS 2025
Publisher	IEEE United Arab Emirates Section
Pages	1-6
Number of pages	6
Publication status	Published - 10 Jun 2025

Bibliographical note

This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.

Keywords

Diffusion model; self-supervised learning; mediacal image diagnosis

Access to Document

diffusion_conference _ieee
This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.
Accepted author manuscript, 1.63 MBLicence: CC BY

Cite this

@inproceedings{d837a8eae48b4bbb8176666f568e46f6,

title = "SSDM:A Self Supervised Diffusion Model For Lung Anomaly Detection Using Chest X-rays",

abstract = "Self supervised learning is emerging very quickly in computer vision tasks, which addresses the scarcity of annotated medical images . We introduce a self-supervised approach foranomaly detection using a Diffusion Probabilistic Model, where the model is trained exclusively on normal chest X-rays and serves as a baseline for identifying anomalies. We implemented a U-net-based Gaussian diffusion model(SSDM) that adds noise in an iterative manner to images and learns to generate them in reverse. Early detection of disease is critical for medical diagnosis.The empirical results show that the model we put forward has good accuracy in detecting the lung anomaly and thereby helps to diagnose disease at an early stage.",

keywords = "Diffusion model; self-supervised learning; mediacal image diagnosis",

author = "Shemy Syed and R Elakkiya and Pears, {N. E.}",

note = "This is an author-produced version of the published paper. Uploaded in accordance with the University{\textquoteright}s Research Publications and Open Access policy. ",

year = "2025",

month = jun,

day = "10",

language = "English",

pages = "1--6",

booktitle = "International Conference on Communication, Computing, Networking, and Control in Cyber-Physical Systems",

publisher = "IEEE United Arab Emirates Section",

}

SSDM:A Self Supervised Diffusion Model For Lung Anomaly Detection Using Chest X-rays. / Syed, Shemy; Elakkiya, R; Pears, N. E.
International Conference on Communication, Computing, Networking, and Control in Cyber-Physical Systems: CCNCPS 2025. IEEE United Arab Emirates Section, 2025. p. 1-6.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - SSDM:A Self Supervised Diffusion Model For Lung Anomaly Detection Using Chest X-rays

AU - Syed, Shemy

AU - Elakkiya, R

AU - Pears, N. E.

N1 - This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.

PY - 2025/6/10

Y1 - 2025/6/10

N2 - Self supervised learning is emerging very quickly in computer vision tasks, which addresses the scarcity of annotated medical images . We introduce a self-supervised approach foranomaly detection using a Diffusion Probabilistic Model, where the model is trained exclusively on normal chest X-rays and serves as a baseline for identifying anomalies. We implemented a U-net-based Gaussian diffusion model(SSDM) that adds noise in an iterative manner to images and learns to generate them in reverse. Early detection of disease is critical for medical diagnosis.The empirical results show that the model we put forward has good accuracy in detecting the lung anomaly and thereby helps to diagnose disease at an early stage.

AB - Self supervised learning is emerging very quickly in computer vision tasks, which addresses the scarcity of annotated medical images . We introduce a self-supervised approach foranomaly detection using a Diffusion Probabilistic Model, where the model is trained exclusively on normal chest X-rays and serves as a baseline for identifying anomalies. We implemented a U-net-based Gaussian diffusion model(SSDM) that adds noise in an iterative manner to images and learns to generate them in reverse. Early detection of disease is critical for medical diagnosis.The empirical results show that the model we put forward has good accuracy in detecting the lung anomaly and thereby helps to diagnose disease at an early stage.

KW - Diffusion model; self-supervised learning; mediacal image diagnosis

M3 - Conference contribution

SP - 1

EP - 6

BT - International Conference on Communication, Computing, Networking, and Control in Cyber-Physical Systems

PB - IEEE United Arab Emirates Section

ER -