Using Few-Shot Learning to Classify Primary Lung Cancer and Other Malignancy with Lung Metastasis in Cytological Imaging via Endobronchial Ultrasound Procedures

TL;DR

This paper introduces a few-shot learning approach for classifying lung cancer types in cytological images from endobronchial ultrasound, addressing data scarcity and morphological similarities to improve early diagnosis.

Contribution

The study develops a hybrid pretrained few-shot learning model with contrastive learning and fine-tuning, achieving better accuracy in low-data lung cancer classification tasks.

Findings

Achieved 49.59% accuracy, outperforming existing methods.

With only 20 samples, accuracy increased to 55.48%.

Demonstrated potential for identifying rare cancer types in clinical settings.

Abstract

This study presents a computer-aided diagnosis (CAD) system to assist early detection of lung metastases during endobronchial ultrasound (EBUS) procedures, significantly reducing follow-up time and enabling timely treatment. Due to limited cytology images and morphological similarities among cells, classifying lung metastases is challenging, and existing research rarely targets this issue directly.To overcome data scarcity and improve classification, the authors propose a few-shot learning model using a hybrid pretrained backbone with fine-grained classification and contrastive learning. Parameter-efficient fine-tuning on augmented support sets enhances generalization and transferability. The model achieved 49.59% accuracy, outperforming existing methods. With 20 image samples, accuracy improved to 55.48%, showing strong potential for identifying rare or novel cancer types in low-data clinical environments.