Absolute ground truth-based validation of computer-aided nodule detection and volumetry in low-dose CT imaging

被引:2
作者
D'hondt, Louise [1 ,2 ,5 ]
Kellens, Pieter-Jan [1 ]
Torfs, Kwinten [3 ]
Bosmans, Hilde [3 ]
Bacher, Klaus [1 ]
Snoeckx, Annemiek [2 ,4 ]
机构
[1] Univ Ghent, Fac Med & Hlth Sci, Dept Human Struct & Repair, Proeftuinstr 86, Ghent, Belgium
[2] Univ Antwerp, Fac Med, Univ Plein 1, Antwerp, Belgium
[3] Katholieke Univ Leuven, Univ Hosp Leuven, Dept Gastroenterol & Hepatol, Leuven, Belgium
[4] Antwerp Univ Hosp, Dept Radiol, Drie Eikenstr 655, Edegem, Belgium
[5] Proeftuinstr 86, B-9000 Ghent, Belgium
来源
PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS | 2024年 / 121卷
关键词
Computer-aided detection; Automatic nodule volumetry; Artificial intelligence; Anthropomorphic chest phantom; Pulmonary nodules; Nodule morphology; Low-dose computed tomography; Lung cancer screening; LUNG-CANCER; PULMONARY NODULES; PERFORMANCE; AI;
D O I
10.1016/j.ejmp.2024.103344
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Purpose: To validate the performance of computer-aided detection (CAD) and volumetry software using an anthropomorphic phantom with a ground truth (GT) set of 3D-printed nodules. Methods: The Kyoto Kaguku Lungman phantom, containing 3D-printed solid nodules including six diameters (4 to 9 mm) and three morphologies (smooth, lobulated, spiculated), was scanned at varying CTDIvol levels (6.04, 1.54 and 0.20 mGy). Combinations of reconstruction algorithms (iterative and deep learning image reconstruction) and kernels (soft and hard) were applied. Detection, volumetry and density results recorded by a commercially available AI-based algorithm (AVIEW LCS + ) were compared to the absolute GT, which was determined through mu CT scanning at 50 mu m resolution. The associations between image acquisition parameters or nodule characteristics and accuracy of nodule detection and characterization were analyzed with chi square tests and multiple linear regression. Results: High levels of detection sensitivity and precision (minimal 83 % and 91 % respectively) were observed across all acquisitions. Neither reconstruction algorithm nor radiation dose showed significant associations with detection. Nodule diameter however showed a highly significant association with detection (p < 0.0001). Volumetric measurements for nodules > 6 mm were accurate within 10 % absolute range from volume(GT), regardless of dose and reconstruction. Nodule diameter and morphology are major determinants of volumetric accuracy (p < 0.001). Density assignment was not significantly influenced by any parameters. Conclusions: Our study confirms the software's accurate performance in nodule volumetry, detection and density characterization with robustness for variations in CT imaging protocols. This study suggests the incorporation of similar phantom setups in quality assurance of CAD tools.
引用
收藏
页数:10
相关论文
共 46 条
[11]   Detection of pulmonary nodules with scoutless fixed-dose ultra-low-dose CT: a prospective study [J].
Gheysens, Gerald ;
De Wever, Walter ;
Cockmartin, Lesley ;
Bosmans, Hilde ;
Coudyzer, Walter ;
De Vuysere, Sofie ;
Lefere, Mathieu .
EUROPEAN RADIOLOGY, 2022, 32 (07) :4437-4445
[12]  
Gold G., 2020, Canon White Paper
[13]   Detection of lung cancer through low-dose CT screening (NELSON): a prespecifi ed analysis of screening test performance and interval cancers [J].
Horeweg, Nanda ;
Scholten, Ernst Th ;
de Jong, Pim A. ;
van der Aalst, Carlijn M. ;
Weenink, Carla ;
Lammers, Jan-Willem J. ;
Nackaerts, Kristiaan ;
Vliegenthart, Rozemarijn ;
ten Haaf, Kevin ;
Yousaf-Khan, Uraujh A. ;
Heuvelmans, Marjolein A. ;
Thunnissen, Erik ;
Oudkerk, Matthijs ;
Mali, Willem ;
de Koning, Harry J. .
LANCET ONCOLOGY, 2014, 15 (12) :1342-1350
[14]  
Hsieh J., 2019, White Paper (JB68676XX)
[15]   Implementation of the cloud-based computerized interpretation system in a nationwide lung cancer screening with low-dose CT: comparison with the conventional reading system [J].
Hwang, Eui Jin ;
Goo, Jin Mo ;
Kim, Hyae Young ;
Yi, Jaeyoun ;
Yoon, Soon Ho ;
Kim, Yeol .
EUROPEAN RADIOLOGY, 2021, 31 (01) :475-485
[16]   Development and performance evaluation of a deep learning lung nodule detection system [J].
Katase, Shichiro ;
Ichinose, Akimichi ;
Hayashi, Mahiro ;
Watanabe, Masanaka ;
Chin, Kinka ;
Takeshita, Yuhei ;
Shiga, Hisae ;
Tateishi, Hidekatsu ;
Onozawa, Shiro ;
Shirakawa, Yuya ;
Yamashita, Koji ;
Shudo, Jun ;
Nakamura, Keigo ;
Nakanishi, Akihito ;
Kuroki, Kazunori ;
Yokoyama, Kenichi .
BMC MEDICAL IMAGING, 2022, 22 (01)
[17]   ESR/ERS statement paper on lung cancer screening [J].
Kauczor, Hans-Ulrich ;
Baird, Anne-Marie ;
Blum, Torsten Gerriet ;
Bonomo, Lorenzo ;
Bostantzoglou, Clementine ;
Burghuber, Otto ;
Cepicka, Blanka ;
Comanescu, Alina ;
Couraud, Sebastien ;
Devaraj, Anand ;
Jespersen, Vagn ;
Morozov, Sergey ;
Agmon, Inbar Nardi ;
Peled, Nir ;
Powell, Pippa ;
Prosch, Helmut ;
Ravara, Sofia ;
Rawlinson, Janette ;
Revel, Marie-Pierre ;
Silva, Mario ;
Snoeckx, Annemiek ;
van Ginneken, Bram ;
van Meerbeeck, Jan P. ;
Vardavas, Constantine ;
von Stackelberg, Oyunbileg ;
Gaga, Mina .
EUROPEAN RADIOLOGY, 2020, 30 (06) :3277-3294
[18]   Applications of artificial intelligence in the thorax: a narrative review focusing on thoracic radiology [J].
Kim, Yisak ;
Park, Ji Yoon ;
Hwang, Eui Jin ;
Lee, Sang Min ;
Park, Chang Min .
JOURNAL OF THORACIC DISEASE, 2021, 13 (12) :6943-6962
[19]   Outstanding negative prediction performance of solid pulmonary nodule volume AI for ultra-LDCT baseline lung cancer screening risk stratification [J].
Lancaster, Harriet L. ;
Zheng, Sunyi ;
Aleshina, Olga O. ;
Yu, Donghoon ;
Chernina, Valeria Yu. ;
Heuvelmans, Marjolein A. ;
de Bock, Geertruida H. ;
Dorrius, Monique D. ;
Gratama, Jan Willem ;
Morozov, Sergey P. ;
Gombolevskiy, Victor A. ;
Silva, Mario ;
Yi, Jaeyoun ;
Oudkerk, Matthijs .
LUNG CANCER, 2022, 165 :133-140
[20]   A narrative review of deep learning applications in lung cancer research: from screening to prognostication [J].
Lee, Jong Hyuk ;
Hwang, Eui Jin ;
Kim, Hyungjin ;
Park, Chang Min .
TRANSLATIONAL LUNG CANCER RESEARCH, 2022, 11 (06) :1217-1229