Lung Cancer: Performance of Automated Lung Nodule Detection Applied to Cancers Missed in a CT Screening Program
RadiologyPublished 1 December 2002
Samuel G. Armato, Feng Li, Maryellen L. Giger, Heber MacMahon, Shusuke Sone, Kunio Doi
Citations279
SJR quartileQ1
SJR score3.86
SNIP3.55
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Evaluating the performance of a fully automated computerized method for the detection of lung nodules in computed tomographic (CT) scans in the identification of lung cancers that may be missed during visual interpretation found it to be effective.
Abstract
With an automated lung nodule detection method, a large fraction (84%, 32 of 38) of missed cancers in a database of low-dose CT scans were detected correctly.
Keywords
Medicine
BiometricsApplied Multivariate Statistical Analysis.
11,424 Citations1988Andrea Johnson, Dean W. Wichern
CA A Cancer Journal for CliniciansCancer Statistics, 2002
2,931 Citations2002Ahmedin Jemal, Anish Thomas +2 more
From 1992 to 1998, cancer death rates declined in males and females, while cancer incidence rates decreased among males and increased slightly among females, most notably, African‐American men showed the largest decline for both incidence and mortality.
The LancetEarly Lung Cancer Action Project: overall design and findings from baseline screening
2,512 Citations1999Claudia I. Henschke, Dorothy I. McCauley +9 more
Low-dose CT can greatly improve the likelihood of detection of small non-calcified nodules, and thus of lung cancer at an earlier and potentially more curable stage, although false-positive CT results are common.
Investigative RadiologyROC Methodology in Radiologic Imaging
1,635 Citations1986Charles E. Metz
This article develops ROC concepts in an intuitive way by identifying the fundamental issues that motivate ROC analysis and practical techniques for ROC data collection and data analysis.
The LancetMass screening for lung cancer with mobile spiral computed tomography scanner
933 Citations1998Shusuke Sone, Shodayu Takashima +9 more
RadiologyPeripheral lung cancer: screening and detection with low-dose spiral CT versus radiography.
839 Citations1996Masahiro Kaneko, K. Eguchi +5 more
Low-dose spiral CT was superior to chest radiography in the screening and detection of peripheral lung cancer in high-risk individuals, and further large-scale studies are warranted, however, to clarify the efficacy and cost-effectiveness of low- dose spiral CT in a randomized, controlled population.
Statistics in MedicineMaximum likelihood estimation of receiver operating characteristic (ROC) curves from continuously-distributed data
786 Citations1998Charles E. Metz, B. Herman +1 more
Two new algorithms for fitting binormal ROC curves to continuously-distributed data are developed: a true ML algorithm (LABROC4) and a quasi-ML algorithm (LabROC5) that requires substantially less computation with large data sets.
British Journal of CancerResults of three-year mass screening programme for lung cancer using mobile low-dose spiral computed tomography scanner
499 Citations2001S Sone, F Li +9 more
This trial allowed detection of nearly 11 times the expected annual number of early lung cancers, compared to those in the initial screening, and repeat CT allowed the detection of more aggressive, rapidly growing lung cancers.
Investigative RadiologyVisual Scanning, Pattern Recognition and Decision-making in Pulmonary Nodule Detection
447 Citations1978Harold L. Kundel, Calvin F. Nodine +1 more
Using the assumption that prolonged dwell time indicates intensive processing of visual data, a model was developed for nodule detection that includes four steps: orientation, scanning, pattern recognition and decision-making.
RadiographicsComputerized Detection of Pulmonary Nodules on CT Scans
376 Citations1999Samuel G. Armato, Maryellen L. Giger +4 more
A computerized scheme to automatically detect pulmonary nodules on CT images, which includes both two- and three-dimensional analyses, yielded an area under the ROC curve of 0.93 in the task of classifying candidates detected during thresholding as nodules or nonnodules.
American Journal of RoentgenologyGlossary of terms for thoracic radiology: recommendations of the Nomenclature Committee of the Fleischner Society
336 Citations1984W. J. Tuddenham
The Glossary was originally proposed at the first general meeting of the Society of 1971 in the belief that standardization of terms with respect to the description of radiographic findings would facilitate the exchange of information and is now judged to be sufficiently inclusive to be of general interest.
RadiologyPulmonary nodules: detection with thick-section spiral CT versus conventional CT.
251 Citations1993Martine Rémy‐Jardin, Jacques Rémy +2 more
Spiral volumetric computed tomography with single breath-hold technique was compared with conventional sequential CT in 39 patients, finding mean number of nodules per patient was significantly higher with spiral versus conventional CT.
Medical PhysicsAutomated detection of lung nodules in CT scans: Preliminary results
250 Citations2001Samuel G. Armato, Maryellen L. Giger +1 more
A fully automated computerized method for the detection of lung nodules in helical computed tomography (CT) scans of the thorax is developed, based on two-dimensional and three-dimensional analyses of the image data acquired during diagnostic CT scans.
Annals of Internal MedicineScreening for Lung Cancer
236 Citations1989David M. Eddy
Four recent controlled trials and one case-control study failed to show that screening reduces lung cancer mortality even in high-risk persons (smokers), and screening for lung cancer is not recommended.
RadiologySmall Pulmonary Nodules: Evaluation with Repeat CT—Preliminary Experience
233 Citations1999David F. Yankelevitz, Rajiv Gupta +2 more
Preliminary experience with early repeat CT suggests that a single repeat CT scan obtained 30 days after the first scan can depict growth in most malignant tumors as small as 5 mm.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>A Free Response Approach To The Measurement And Characterization Of Radiographic Observer Performance</title>
213 Citations1977Philip C. Bunch, John F. Hamilton +2 more
A scheme for reinterpretation of earlier conventional Receiver Operating Characteristic (ROC) curves in terms of the free response approach is presented, and a description of a much simpler and more efficient experimental procedure is provided.
Investigative RadiologyComputerized Detection of Pulmonary Nodules in Computed Tomography Images
201 Citations1994Maryellen L. Giger, Kyongtae T. Bae +1 more
Developing automated methods to detect nodules on CT images of the thorax using various computer-vision techniques and a priori information of the morphologic characteristics of pulmonary nodules are developed.
RadiologyLung Cancers Missed at Low-Dose Helical CT Screening in a General Population: Comparison of Clinical, Histopathologic, and Imaging Findings
199 Citations2002Feng Li, Shusuke Sone +4 more
The lung cancers missed at low-dose CT screening in this series generally were very subtle and appeared as small faint nodules, overlapping normal structures, or opacities in a complex background of other disease.
RadiologyChest CT: Automated Nodule Detection and Assessment of Change over Time—Preliminary Experience
195 Citations2001Jane P. Ko, Margrit Betke
A computer system that automatically identifies nodules at chest computed tomography, quantifies their diameter, and assesses for change in size at follow-up matched that by the thoracic radiologist (Spearman rank correlation coefficient).
RadiologyPulmonary nodule detection: low-dose versus conventional CT.
177 Citations1998Henry Rusinek, David P. Naidich +6 more
Low-dose CT is acceptable for pulmonary nodule identification, making it suitable for primary screening, and results confirm the strong effect of size, location, and angiocentricity on the sensitivity of nodule detection with conventional CT.
RadiologyNon-small-cell lung cancer: results of the New York screening program.
153 Citations1984R T Heelan, Betty J. Flehinger +5 more
Radiographs of 168 patients with non-small-cell lung cancer were reviewed; following a negative initial examination, 102 tumors were detected during routine annual screening, while 66 were diagnosed during the interval between screenings.
RadiologyPulmonary Nodules: Experimental and Clinical Studies at Low-Dose CT
151 Citations1999Stefan Diederich, Horst Lenzen +8 more
Pulmonary nodules were detected reliably at CT with 50 mA and pitch of 2 or with 25mA and a pitch of 1, however, further reduction of the dose to that used at chest radiography was associated with a significant decrease in the number of nodules 5 mm or smaller that were detected, possibly due to artifacts.
American Journal of RoentgenologyDetection of pulmonary nodules by computed tomography
110 Citations1977Muhm, L R Brown +1 more
RadiologyPrimary carcinoma of the lung overlooked at CT: analysis of findings in 14 patients.
96 Citations1996Charles S. White, Benjamin Romney +4 more
Endobronchial location and lower lobe predominance were the most common characteristics of overlooked lung cancer at CT and may have contributed to failure to diagnose the tumor.
RadiologyCT Screening for Lung Cancer: Not Ready for Routine Practice
68 Citations2001Edward F. Patz, William C. Black +1 more
It remains to be seen if CT will truly reduce mortality, and the appropriate hypothesis-driven studies still must be performed and the results carefully analyzed before CT screening for lung cancer can be accepted as the standard of care.
RadiologyCT Screening for Lung Cancer: Coping with Nihilistic Recommendations
66 Citations2001Olli S. Miettinen, Claudia I. Henschke
This article is an attempt to help the critically thinking practitioner understand the foundation of the official nihilism in evidence and reasoning, as she or he suspects that something may be seriously wrong in this.
CHEST JournalComputed Tomography of the Thorax: A Status Report
40 Citations1981Robert D. Pugatch, L. Jack Faling
Computed tomography of the thorax shows early promise of important diagnostic advances and is ideal for detecting pleural abnormalities and for displaying underlying parenchymal disease in patients with complex pleuroparen chymal shadows on conventional films.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Image processing for computer-aided diagnosis of lung cancer screening system by CT (LSCT)</title>
38 Citations1998Toshiaki Okumura, Tomoko Miwa +6 more
A new method named Variable-New-Quoit filter for the automatic recognition of the pathological shadow candidates is proposed, which can satisfactorily reduce the number of CT cross sections by this method, containing the abnormal shadow candidates.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Computer-aided diagnosis system for lung cancer based on retrospective helical CT image</title>
36 Citations2000Yuji Ukai, Noboru Niki +7 more
An algorithm is developed that can compare automatically the slice images of present and early CT scans for the assistance of comparative reading in retrospect and indicates that the CAD system is useful to increase the efficiency of the mass screening process.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Automatic detection of pulmonary nodules in low-dose screening thoracic CT examinations</title>
36 Citations1999Martin Fiebich, Christian Wietholt +5 more
A technique is developed, which might help the radiologist in the detection of pulmonary nodules in CT exams of the chest, which was applied to a small subset of studies evaluating the feasibility of screening lung cancer.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Object-based deformation technique for 3D CT lung nodule detection</title>
27 Citations1999Shyhliang A. Lou, Chun-Long Chang +2 more
An object-based deformation method to detect lung nodules from CT images in three-dimension (3-D) is reported, which uses a deformable seed object technique to differentiate lung nodule from blood vessels and airways.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Three-dimensional approach to lung nodule detection in helical CT</title>
26 Citations1999Samuel G. Armato, Maryellen L. Giger +3 more
An automated method for the detection of lung nodules in helical computed tomography (CT) images that incorporates 2D and 3D analyses to exploit the volumetric image data acquired during a CT examination is developed.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Lung cancer detection based on helical CT images using curved-surface morphology analysis</title>
24 Citations1999Hiroshi Taguchi, Yoshiki Kawata +7 more
This paper describes lung cancer detection based on helical CT Images using curved surface morphology analysis, and proposes a computer-aided diagnosis (CAD) that will increase diagnostic confidence.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Computer-aided diagnosis system for lung cancer based on retrospective helical CT images</title>
19 Citations1999Hitoshi Satoh, Yuji Ukai +7 more
An algorithm is developed that can compare automatically the slice images of present and early CT scans for the assistance of comparative reading in retrospect and indicates that the CAD system is useful to increase the efficiency of the mass screening process.
Radiologic Clinics of North AmericaAdvances in Computed Tomography of the Thorax
19 Citations1983W. Richard Webb
The author discusses the CT detection of diseases such as lung nodules, mediastinal masses, and hilar lesions; CT's use as a screening technique; and its replacement of angiography in the evaluation of intrathoracic vascular lesions.
Lecture notes in computer scienceComputer assisted lung cancer diagnosis based on helical images
18 Citations1995Keizo Kanazawa, M. Kubo +5 more
A computer assisted automatic diagnosis system of lung cancer that detects tumor candidates in its early stage from the helical CT images reduces the time complexity and increases the diagnosis confidence.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Computerized lung nodule detection: comparison of performance for low-dose and standard-dose helical CT scans</title>
11 Citations2001Samuel G. Armato, Maryellen L. Giger +3 more
An automated method to automatically analyze structures within a CT scan and identify those structures that represent lung nodules and demonstrates promising performance in its ability to accurately detect lung nodule detection in standard-dose and low-dose CT images.
Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Analysis of a three-dimensional lung nodule detection method for thoracic CT scans</title>
11 Citations2000Samuel G. Armato, Maryellen L. Giger +1 more
An automated method to analyze the three- dimensional nature of structures within CT scans and identify those structures that represent lung nodules is developed, demonstrating promising performance in its ability to detect Lung nodules in CT images.