A Lung Nodule Dataset with Histopathology-based Cancer Type Annotation

TL;DR

This paper introduces a new publicly available dataset of lung CT images with detailed histopathology-based cancer type annotations, aiming to improve CAD systems' ability to classify lung cancer types.

Contribution

The authors curated a diverse, annotated lung CT dataset with 330 nodules from 95 patients, enabling more precise cancer classification and supporting the development of better diagnostic tools.

Findings

Classical models achieved promising classification accuracy.

The dataset demonstrated feasibility for training detection and classification models.

Enhanced data availability can improve lung cancer diagnosis accuracy.

Abstract

Recently, Computer-Aided Diagnosis (CAD) systems have emerged as indispensable tools in clinical diagnostic workflows, significantly alleviating the burden on radiologists. Nevertheless, despite their integration into clinical settings, CAD systems encounter limitations. Specifically, while CAD systems can achieve high performance in the detection of lung nodules, they face challenges in accurately predicting multiple cancer types. This limitation can be attributed to the scarcity of publicly available datasets annotated with expert-level cancer type information. This research aims to bridge this gap by providing publicly accessible datasets and reliable tools for medical diagnosis, facilitating a finer categorization of different types of lung diseases so as to offer precise treatment recommendations. To achieve this objective, we curated a diverse dataset of lung Computed Tomography (CT) images, comprising 330 annotated nodules (nodules are labeled as bounding boxes) from 95 distinct patients. The quality of the dataset was evaluated using a variety of classical classification and detection models, and these promising results demonstrate that the dataset has a feasible application and further facilitate intelligent auxiliary diagnosis.