X-ray or gamma ray systems or devices – Specific application – Computerized tomography
Reexamination Certificate
2002-08-08
2004-09-21
Bruce, David (Department: 2882)
X-ray or gamma ray systems or devices
Specific application
Computerized tomography
C378S008000, C378S901000
Reexamination Certificate
active
06795521
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to methods and systems for automated and interactive processing of medical computer tomographic (CT) images, and is more specifically related to computerized methods and systems for multi-structure enhancement, volume matching, object analysis, and object detection in thoracic CT images using digital image processing techniques.
2. Background Art
Lung cancer has the highest cancer mortality for both men and women worldwide. Early detection and treatment of localized lung cancer at a potentially curable stage can significantly increase the patient survival rates. Studies have shown a survival rate of approximately 60% when lung cancer is detected in the early stages. However, only approximately 15% of lung cancers are diagnosed at an early stage when the disease is still localized.
Diagnosis of lung cancer can be accomplished using either projectional (i.e., chest radiography) or cross-sectional (i.e., computer tomography) techniques. Chest X-ray films have been used for lung cancer diagnosis as a conventional method for mass screening, due to their ready availability and reasonable sensitivity at showing localized lung abnormalities. However, there are obvious disadvantages inherent in the use of a projection image, the most notable being the masking of nearly half of the lungs by overlaying structures such as the heart and diaphragm.
Computer tomography (CT) provides a cross-sectional image of the lung, as opposed to the projection image provided by a chest X-ray. Since the early 1990s, the volumetric CT technique has been available to provide virtually contiguous spiral scans that cover the chest in a few seconds. This technique has greatly reduced CT image artifacts caused by unequal respiratory cycles, partial volume, and cardiac motion. Newer models of the helical CT system are capable of performing the scan and image reconstruction simultaneously. Detectability of pulmonary nodules has been greatly improved with this modality (Zerhouni et al., “Factors Influencing Quantitative CT Measurements of Solitary Pulmonary Nodules,”
J. Comput. Assisted Tomography
, 6:1075-87, 1982; Siegelman et al., “Solitary Pulmonary Nodules: CT Assessment,”
Radiology
, 160:307-312, 1986; Zerhouni et al., “CT of Pulmonary Nodule: A Cooperative Study,”
Radiology
, 160:319-327, 1986; Webb, W. R., “Radiologic Evaluation of the Solitary Pulmonary Nodule,”
Am. J. Roentgenology
, 154:701-708, 1990). High-resolution CT has also proved to be effective in characterizing the edges of pulmonary nodules (Zwirewich et al., “Solitary Pulmonary Nodule: High-Resolution CT and Radiologic-Pathologic Correlation,”
Radiology
, 79:469-476, 1991).
Recent studies have demonstrated that spiral CT can detect small lung nodules that are barely visible on chest X-ray films (Henschke et al., “Early Lung Cancer Action Project: Overall Design and Findings from Baseline Screening,”
The Lancet
, Vol. 354, pp. 99-105, 1999; Sone et al., “Mass Screening for Lung Cancer with Mobil Spiral Computed Tomography Scanner,”
The Lancet
, Vol. 351, pp. 1242-45, 1998). Zwirewich and his colleagues reported that shadows of nodule spiculation correlate pathologically with irregular fibrosis, localized lymphatic spread of a tumor, or an infiltrative tumor growth; pleural tags represent fibrotic bands that usually are associated with juxtacicatrical pleural retraction; and low attenuation bubble-like patterns are correlated with bronchioloalveolar carcinomas. These are common CT image patterns associated with malignant processes of lung masses.
Although skilled pulmonary radiologists have a high degree of accuracy in diagnosis of lung cancer using advanced CT imaging technology, there remain challenges that can not be overcome using current methods of training or by attaining high levels of clinical skill and experience. These include the miss rate for detection of small pulmonary nodules, the detection of minimal interstitial lung disease, and the detection of changes in preexisting interstitial lung disease.
Because a majority of solitary pulmonary nodules (SPN) are benign, Siegelman et al. (1986) determined three main criteria for benignancy: (a) high attenuation values distributed diffusely throughout the nodule, (b) a representative CT number of at least 164 HU, and (c) hamartomas, which are lesions 2.5 cm or less in diameter with sharp and smooth edges and a central focus of fat with CT numbers of −40 to −120 HU. These reports suggest that these are features that a computer-aided analytical tool could use to differentiate benign from malignant lesions. As to detection of lung cancers by various modes of CT, Remy-Jardin et al. (Remy-Jardin et al., “Pulmonary Nodules: Detection with Thick-Section Spiral CT versus Conventional CT,”
Radiology
, 187:513-520, 1993) showed that the thick-section (10 mm) helical CT markedly reduces cardiac motion artifacts and misses fewer lung nodules than the conventional CT. In Japan, CT-based lung cancer screening programs have been developed (Tateno et al., “Development of X-ray CT for Lung Cancer Detection,”
Shih-Iryo
, 17(10):28-32, 1990; Iinuma et al., “Basic Idea of Lung Cancer Screening CT (LSCT) and Its Preliminary Evaluation,”
Jap. J. Radiol. Med
., 52(2):182-190, 1992). In the U.S., however, only a limited demonstration project, funded by the NIH/NCI, using helical CT has been reported (Yankelevitz et al., “Repeat CT Scanning for Evaluation of Small Pulmonary Nodules,”
Radiology
, 1999). The trend towards using helical CT as one of the clinical tools for screening lung cancer has four motivating factors: (1) an alternative to the low sensitivity of chest radiography in the detection of small cancers (smaller than 10 mm); (2) the development of higher throughput, low-dose helical CT; (3) the potential reduction of health care costs using helical CT; and (4) the development of a computer diagnostic system as an aid for pulmonary radiologists. One can anticipate that the cost of each CT examination will still be higher than that of a conventional chest X-ray. However, for the high-risk population, the greater potential of this imaging modality for detecting early lung cancer may outweigh its increased cost.
Several clinical trials are now underway in the U.S. Studies at the Mayo Clinic and the University of South Florida were funded in 1999 and have recently begun. The method is being shown to detect small volume lung cancers, but with false positive rates for nodule detection in the range of 23-50%. In Japan, direct questioning of Drs. Sone and Kaneko (Oct. 1-3, 1999, Conference on Screening for Lung Cancer, Cornell Medical College, New York) indicated that the false negative rate in the work they were reporting was 10 to 15% and sometimes as high as 90% when the lesion is very small. In addition, Dr. Sone reported at that meeting that 16% of those patients undergoing thoracotomy had non-cancerous lesions and an additional 6% had atypical adenomatous hyperplasia, a benign lesion of uncertain malignant potential.
Given the frequency of false positive detections in both the Japanese and U.S. studies, the frequency of false negative exams in Japanese studies, and the frequency of thoracotomy uncovering only benign disease, there is a pressing need for development of improved diagnostic methods, as the use of screening CT is rapidly increasing and high false positive rates will result in many unnecessary procedures. Computer algorithms have been shown to increase sensitivity and/or specificity in the detection of lung cancer on chest radiographs and breast cancer on mammograms. Thus, application of these methods to screening CT is appropriate.
U.S. Pat. No. 6,125,194, filed Feb. 4, 1998, having the same assignee as the present application and incorporated herein by reference, is aimed specifically at identification of the smallest of lung nodules, 3 to 20 mm, well within the size limits for T
1
cancer (<30 mm). If there is to be a measurable benefit in the use of the helical CT, it is likely to
Hsu Li-Yueh
Li Ruiping
Lin Jyh-Shyan
Lure Fleming Y.-M.
Martello Edward A.
Bruce David
Deus Technologies LLC
Gluck Jeffrey W.
Kinberg Robert
Venable LLP
LandOfFree
Computer-aided diagnosis system for thoracic computer... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Computer-aided diagnosis system for thoracic computer..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Computer-aided diagnosis system for thoracic computer... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3203745