Lung Nodule-SSM: Self-Supervised Lung Nodule Detection and Classification in Thoracic CT Images

TL;DR

This paper introduces LungNodule-SSM, a self-supervised learning approach using transformer architectures to detect and classify lung nodules in CT images without requiring annotated data, achieving high accuracy on the LUNA 16 dataset.

Contribution

It presents a novel self-supervised framework leveraging DINOv2 and transformers for lung nodule detection and classification, reducing dependence on annotated datasets.

Findings

Achieved 98.37% accuracy on LUNA 16 dataset.

Outperformed state-of-the-art methods in lung nodule detection.

Demonstrated effectiveness of self-supervised learning in medical imaging.

Abstract

Lung cancer remains among the deadliest types of cancer in recent decades, and early lung nodule detection is crucial for improving patient outcomes. The limited availability of annotated medical imaging data remains a bottleneck in developing accurate computer-aided diagnosis (CAD) systems. Self-supervised learning can help leverage large amounts of unlabeled data to develop more robust CAD systems. With the recent advent of transformer-based architecture and their ability to generalize to unseen tasks, there has been an effort within the healthcare community to adapt them to various medical downstream tasks. Thus, we propose a novel "LungNodule-SSM" method, which utilizes selfsupervised learning with DINOv2 as a backbone to enhance lung nodule detection and classification without annotated data. Our methodology has two stages: firstly, the DINOv2 model is pre-trained on unlabeled CT scans to learn robust feature representations, then secondly, these features are fine-tuned using transformer-based architectures for lesionlevel detection and accurate lung nodule diagnosis. The proposed method has been evaluated on the challenging LUNA 16 dataset, consisting of 888 CT scans, and compared with SOTA methods. Our experimental results show the superiority of our proposed method with an accuracy of 98.37%, explaining its effectiveness in lung nodule detection. The source code, datasets, and pre-processed data can be accessed using the link:https://github.com/EMeRALDsNRPU/Lung-Nodule-SSM-Self-Supervised-Lung-Nodule-Detection-and-Classification/tree/main