A computer-aided diagnosis system using deep-learning analysis to detect lung lesion locations and quantitatively characterize the lesions on computed tomography (CT) images offered a fast and convenient approach for assisting radiologists in the diagnosis of lung nodule pathologies. These findings were presented by Xiaohua Liu, PhD, during ASCO Breakthrough: A Global Summit for Oncology Innovators held recently in Bangkok, Thailand.1 Dr. Liu is a Research Scientist at LinkDoc Technology in Beijing, a company providing big data integration and other services to medical institutions and pharmaceutical companies in China.
The proposed computer-aided diagnosis system offers a fast and convenient approach for assisting the diagnosis of lung nodule pathologies….— Xiaohua Liu, PhD
Tweet this quote
Worldwide, among men and women combined, lung cancer is the most commonly diagnosed and deadliest cancer, with approximately 2,093,876 new cases diagnosed in 2018 and 1,761,007 deaths reported. Although the early detection and treatment of lung cancer improve survival outcome, the accurate identification and characterization of small lung nodules are challenging.
In this study, Dr. Liu and investigators employed deep-learning analysis for lesion detection in patients and performed image-processing techniques to generate quantitative morphology features for assisting lesion diagnosis. The data collected included 3,956 lung CT series with multiple lung nodules from 15 class A hospitals in China, 1,155 lung CT scans from the publicly available Lung Nodule Analysis 2016 data sets, and CT scans from the data sets of the Kaggle Data Science Bowl 2017. Lung nodule annotation was then performed by two experienced radiologists and further assessed by four senior associate chief physicians.
The obtained CT images were randomly selected and split to construct training, validation, and test data sets. After preprocessing, a pretrained residual neural network framework was transferred to develop a robust detection system that could detect the possible lung lesion locations with corresponding probabilities.
The researchers found that the resulting detection system yielded a free-response receiver operating characteristic of 0.4663, recall of 82.46%, and precision of 36.06% for 5-mm to 30-mm nodules. Each detected lesion was labeled by its bounding box and then analyzed through an image-processing algorithm to generate diagnostic assisting features, including longest and shortest diameter, volume, and largest cross-section area, as well as its density type (calcified, solid, partial solid, and ground-glass opacity).
“The proposed computer-aided diagnosis system offers a fast and convenient approach for assisting the diagnosis of lung nodule pathologies, and it is beneficial to relate our research to the current framework of lung cancer diagnosis,” Dr. Liu and his colleagues concluded. ■
DISCLOSURE: Funding for this study was provided by LinkDoc Technology. Dr. Liu is employed in a leadership position by LinkDoc Technology, where he holds stock and ownership interests, and has received research funding and travel and accommodation expenses.
1. Liu X: Deep learning analysis for automatic lung nodule detection. ASCO Breakthrough. Abstract 27. Presented October 12, 2019.