Lung Cancer Segmentation with Three-Parameter Logistic Type Distribution Model  

作　　者：Debnath Bhattacharyya EaliStephen Neal Joshua N.Thirupathi Rao Yung-cheol Byun 

机构地区：[1]Department of Computer Science and Engineering,Koneru Lakshmaiah Education Foundation,Vaddeswaram,Guntur,522302,Andhra Pradesh,India [2]Department of Computer Science&Engineering,Vignan’s Institute of Information Technology(A),Visakhapatnam,530049,Andhra Pradesh,India [3]Department of Computer Engineering,Jeju National University,102 Jejudaehak-ro,Jeju-si,Jeju-do,690-756,Korea

出　　处：《Computers, Materials & Continua》2023年第4期1447-1465,共19页计算机、材料和连续体（英文）

基　　金：supported by the Ministry of SMEs and Startups (MSS),Korea,under the“Startup growth technology development program (R&D,S3125114)”;by the Ministry of Small and Medium-sized Enterprises (SMEs)and Startups (MSS),Korea,under the“Regional Specialized Industry Development Plus Program (R&D,S3246057)”supervised by the Korea Institute for Advancement of Technology (KIAT).

摘　　要：Lung cancer is the leading cause of mortality in the world affectingboth men and women equally.When a radiologist just focuses on the patient’sbody, it increases the amount of strain on the radiologist and the likelihoodof missing pathological information such as abnormalities are increased.One of the primary objectives of this research work is to develop computerassisteddiagnosis and detection of lung cancer. It also intends to make iteasier for radiologists to identify and diagnose lung cancer accurately. Theproposed strategy which was based on a unique image feature, took intoconsideration the spatial interaction of voxels that were next to one another.Using the U-NET+Three parameter logistic distribution-based technique, wewere able to replicate the situation. The proposed technique had an averageDice co-efficient (DSC) of 97.3%, a sensitivity of 96.5% and a specificity of94.1% when tested on the Luna-16 dataset. This research investigates howdiverse lung segmentation, juxta pleural nodule inclusion, and pulmonarynodule segmentation approaches may be applied to create Computer AidedDiagnosis (CAD) systems. When we compared our approach to four otherlung segmentation methods, we discovered that ours was the most successful.We employed 40 patients from Luna-16 datasets to evaluate this. In termsof DSC performance, the findings demonstrate that the suggested techniqueoutperforms the other strategies by a significant margin.

关 键 词：Magnetic resonance imaging(MRI) lung cancer Luna-16 logistic distribution SEGMENTATION deep learning juxta plural pulmonary nodules 

分 类 号：R445.2[医药卫生—影像医学与核医学] TP391.41[医药卫生—诊断学] TP18[医药卫生—临床医学]