Computer-aided Diagnosis Tools for Early, Non-invasive Detection of Lung Cancer (06048, 09018, 10062, 11056 & 12062)
- Detailed Technology Description
- None
- *Abstract
-
Features and Benefits
- Early detection and diagnosis of lung cancer
- Fully automatic detection of lung nodules in 3D
- Non-bias quantitative data analysis for accurate determination of cancerous nodules
- Registration of images with compensation for lung expansion and contraction, and patient movement
*This technology portfolio is available for licensing, further development, or industrial partnering
Technology
University of Louisville researchers are developing a number of novel technologies aimed at early and accurate detection of lung cancer. The CAD system technology for automatic detection of lung nodules in low dose CT (LDCT) scans requires segmentation of raw scanning information to isolate the lung tissues from the rest of the chest cavity and extraction of the 3D anatomic structures from the segmented lung tissue to reduce the searching space before isolating the true nodules from the other extracted structures. This novel model increases nodule topology flexibility, allowing for various nodules to be detected simultaneously by the same technique; includes automatic parameter estimation of the nodule models using the gray level information of the segmented data; and provides an enhanced accuracy of the CAD system without increasing overall diagnosis time.
A fully automated, image-based diagnostic system technology for early diagnosis of malignant lung nodules achieves early diagnosis by locating the nodule, by radiologist or lung CAD system; segments the lung in the LDCT images; performs a novel data registration (global and local alignment) of two successive LDCT scans to correct for motion and establish correspondence between the nodules; segments the nodules; and measures the volumetric changes between the nodules. Since all computations following the detection of lung nodules by LDCT images are performed in a fully automatic mode, this approach is much simpler for clinical use and could lead to precise diagnosis and identification of the development of detected lung nodules. This technology is also applicable for diagnosing brain tumors, colon cancer, prostate cancer, and liver cancer.
The 3D shape analysis technology for early assessment of detected lung nodules provides a method for diagnosing malignant lung nodules by their shape, rather than conventional growth rate. This technology performs 3D nodule segmentations with a deformable 3D boundary controlled by two probabilistic visual appearance models, and a 3D Delaunay triangulation to construct a 3D mesh model of the segmented lung nodule surface. The 3D mesh model is mapped to a unit sphere, the spherical harmonics (SHs) for the surface are computed, and the number of SHs to delineate the lung nodule are determined. This novel technology demonstrates the probability of malignancy from the first lung nodule detection, can quantify the shape changes of any medical organ, and is applicable for early diagnosis of autism, dyslexia, Alzheimer's, and prostate cancer.
The 3D volumetric analysis software uses an input of binary images to create a 3D volume in memory, refines that volume, and converts it to a Delaunay mesh where a unit sphere is created and used as the base harmonic to construct a linear combination of SHs. The differences between the original mesh and the reconstructed mesh allows for the creation of an error array that is statistically analyzed to classify objects as cancerous or noncancerous with a high degree of accuracy. This technology is applicable to any medical organ. Finally, the technology provides for the use of SOX9, as a companion diagnostic for XB05, a small molecule with anticancer activities. SOX9 is a predictive biomarker that potentially allows for the pre-selection of patients likely to respond to a XB05 and due to this capability, is highly likely to expedite the clinical development of such compounds.
Markets Addressed
The global 3D imaging market is expected to grow at a CAGR of 26.7% from $3.01 billion in 2013 to $9.82 billion in 2018. More specifically, the 3D medical imaging market is expected to reach $5.9 billion by 2017 due to increases in 3D medical imaging procedures performed and 3D modalities sold; thus, leading to increased acceptance and use of 3D visualization techniques. Additionally, the U.S. CAD/CAM system market is expected to grow to over $300 million by 2020.
Technology Status
IP Status:
- U.S. Patent No. 8,073,226, "Automatic detection and monitoring of nodules and shaped targets in image data," ULRF Ref. 06048
- U.S. Patent No. 8,731,255, "Computer aided diagnostic system incorporating lung segmentation and registration," ULRF Ref. 09018
- U.S. Patent No. 9,014,456, "Computer aided diagnostic system incorporating appearance analysis for diagnosing malignant lung nodules," ULRF Ref. 10062
- U.S. Patent No. 9,230,320, "Computer aided diagnostic system incorporating shape analysis for diagnosing malignant lung nodules," ULRF Ref. 11056
- U.S. Provisional Patent App. No. 62/394,315, "Accurate detection and assessment of radiation induced lung injury based on computational model and computed tomography," (pending) ULRF Ref. 16078
- Registered Copyright for "Mesh Diagnostic Software," U.S. Copyright Office Case No. 1-748867631, ULRF Ref. 12062
Publications:
- A. El-Baz, G. Beache, G. Gimel'farb, K. Suzuki, K. Okada, A. Elnakib, A. Soliman, and B. Abdollahi, "Computer Aided Diagnosis Systems for Lung Cancer: Challenges and Methodologies," International Journal of Biomedical Imaging, vol. 2013, pp. 1-46, 2013.
- A. El-Baz, G. Gimel'farb, R. Falk, and M. Abo El-Ghar, "Automatic Analysis of 3D Low Dose CT Images for Early Diagnosis of Lung Cancer," Pattern Recognition Journal, vol. 42, no. 6, pp. 1041-1051, June 2009.
- A. El-Baz, G. Gimel'farb, R. Falk, and M. Abo El-Ghar, "Automatic Analysis of 3D Low Dose CT Images for Early Diagnosis of Lung Cancer," Pattern Recognition Journal, vol. 42, no. 6, pp. 1041-1051, June 2009.
- A. El-Baz, G. Gimel'farb, R. Falk, M. Abou El-Ghar, and H. Refaie, "Promising Results for Early Diagnosis of Lung Cancer," Proc. of IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI'08), Paris, France, May 14–17, 2008, pp. 1151-1154.
- A. El-Baz, G. Gimel'farb, R. Falk, and M. Abou El-Ghar, "A Novel Image-Based Diagnostic System for Early Diagnosis of Lung Cancer," Presented at AACC Oak Ridge Conference, San Jose, California, April 17-18, 2008. (Abstract)
- Country/Region
- USA
