Parallelization of Preprocessing for Automatic
Detection of Pulmonary Nodules in 3D CT
Images
Junpeng Weng, Yu Tang
(&)
, and Di Lin
University of Electronic Science and Technology of China, No. 4, North Jianshe
Road, Chengdu, P.R. China
wengjunpeng@qq.com, {yutang,lindi}@uestc.edu.cn
Abstract. Computer-aided diagnosis (CAD) systems have been introduced for
therapeutic institutions to help radiologists to diagnose lung cancer at the early
stage. This paper works on the parallelization of preprocessing steps in the CAD
systems. We first propose a parallelization scheme based on static data grouping.
The 3D lung CT images are grouped in the form of 2D image slices, and the
grouped data is parallelized. Then, a pipeline-based parallelization scheme is
proposed. Mutually independent steps are distributed to the respective execution
units, and the preprocessing operations are performed in a parallel pipeline
manner. Finally, a bus-based parallelization scheme is proposed. Based on the
pipeline, a control unit is added to dynamically assign computing tasks to the
various execution units. We test our schemes in an existing CAD system with
data from Lung Nodule Analysis 2016 (LUNA16) dataset, and analyze the
performance of data testing.
Keywords: Lung nodule detection Á Parallelization Á Image processing
1 Introduction
Lung cancer is one of the most deadly cancers in the world. According to the Cancer
Research Report released by the American Cancer Research Center in 2018, there are
more than 2 million new cases of lung cancer in the world in 2018, accounting for
about 11.6% of all new cancer cases; more than 1.7 million deaths, accounting for the
total about 18.4% of cancer cases, which far exceed other cancer types [1]. Clinical
studies show that the five-year survival rate of early lung cancer patients can reach
70%, but the five-year survival rate of patients with advanced lung cancer significantly
drops to less than 15% [2]. These studies indicate that the survival rate of lung cancer is
closely related to whether they can receive early treatment.
Pulmonary nodule is one of the common lesions in the lungs, and is often seen as
early manifestations of lung cancer. The early detection of pulmonary nodules plays a
crucial role in the early treatment of lung cancer, but it is hard to identify potential lung
cancer patients who have developed pulmonary nodules under the condition of limited
medical conditions. To solve the problem, computer-aided detection systems are
introduced into the therapeutic facility. By collaborating with the CAD systems, it is
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 584–590, 2020.
https://doi.org/10.1007/978-981-15-0187-6_69
Detection of Pulmonary Nodules in 3D CT
Images
Junpeng Weng, Yu Tang
(&)
, and Di Lin
University of Electronic Science and Technology of China, No. 4, North Jianshe
Road, Chengdu, P.R. China
wengjunpeng@qq.com, {yutang,lindi}@uestc.edu.cn
Abstract. Computer-aided diagnosis (CAD) systems have been introduced for
therapeutic institutions to help radiologists to diagnose lung cancer at the early
stage. This paper works on the parallelization of preprocessing steps in the CAD
systems. We first propose a parallelization scheme based on static data grouping.
The 3D lung CT images are grouped in the form of 2D image slices, and the
grouped data is parallelized. Then, a pipeline-based parallelization scheme is
proposed. Mutually independent steps are distributed to the respective execution
units, and the preprocessing operations are performed in a parallel pipeline
manner. Finally, a bus-based parallelization scheme is proposed. Based on the
pipeline, a control unit is added to dynamically assign computing tasks to the
various execution units. We test our schemes in an existing CAD system with
data from Lung Nodule Analysis 2016 (LUNA16) dataset, and analyze the
performance of data testing.
Keywords: Lung nodule detection Á Parallelization Á Image processing
1 Introduction
Lung cancer is one of the most deadly cancers in the world. According to the Cancer
Research Report released by the American Cancer Research Center in 2018, there are
more than 2 million new cases of lung cancer in the world in 2018, accounting for
about 11.6% of all new cancer cases; more than 1.7 million deaths, accounting for the
total about 18.4% of cancer cases, which far exceed other cancer types [1]. Clinical
studies show that the five-year survival rate of early lung cancer patients can reach
70%, but the five-year survival rate of patients with advanced lung cancer significantly
drops to less than 15% [2]. These studies indicate that the survival rate of lung cancer is
closely related to whether they can receive early treatment.
Pulmonary nodule is one of the common lesions in the lungs, and is often seen as
early manifestations of lung cancer. The early detection of pulmonary nodules plays a
crucial role in the early treatment of lung cancer, but it is hard to identify potential lung
cancer patients who have developed pulmonary nodules under the condition of limited
medical conditions. To solve the problem, computer-aided detection systems are
introduced into the therapeutic facility. By collaborating with the CAD systems, it is
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 584–590, 2020.
https://doi.org/10.1007/978-981-15-0187-6_69