Data processing: artificial intelligence – Fuzzy logic hardware – Fuzzy neural network
Reexamination Certificate
2000-07-25
2003-11-25
Davis, George B. (Department: 2121)
Data processing: artificial intelligence
Fuzzy logic hardware
Fuzzy neural network
Reexamination Certificate
active
06654728
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to methods and systems for the digital processing of radiological images, and it more specifically relates to an automated method and system for the re-screening and detection of abnormalities, such as lung nodules in radiological chest images, using multi-resolution processing, digital image processing, fuzzy logic and artificial neural networks.
2. Background Art
Lung cancer is the leading type of cancer in both men and women worldwide. Early detection and treatment of localized lung cancer at a potentially curable stage can significantly increase the patients' survival rate. Studies have shown that approximately 68% of retrospectively detected lung cancers were detected by one reader and approximately 82% were detected with an additional reader as a “second-reader”. A long-term lung cancer screening program conducted at the Mayo Clinic found that 90% of peripheral lung cancers were visible in small sizes in retrospect, in earlier radiographs.
Among the common detection techniques, such as chest X-ray, analysis of the types of cells in sputum specimens, and fiber optic examination of bronchial passages, chest radiography remains the most effective and widely used method. Although skilled pulmonary radiologists can achieve a high degree of accuracy in diagnosis, problems remain in the detection of the lung nodules in chest radiography due to errors that cannot be corrected by current methods of training even with a high level of clinical skill and experience.
An analysis of the human error in diagnosis of lung cancer revealed that about 30% of the misses were due to search errors, about 25% of the misses were due to recognition errors, and about 45% of the misses were due to decision-making errors. Reference is made to Kundel, H. L., et al., “Visual Scanning, Pattern Recognition and Decision-Making in Pulmonary Nodule Detection”, Investigative Radiology, May-June 1978, pages 175-181, and Kundel, H. L., et al., “Visual Dwell Indicates Locations of False-Positive and False-Negative Decisions”, Investigative Radiology, June 1989, Vol. 24, pages. 472-478, which are incorporated herein by reference. The analysis suggested that the miss rates for the detection of small lung nodules could be reduced by about 55% with a computerized method. According to the article by Stitik, F. P., “Radiographic Screening in the Early Detection of Lung Cancer”, Radiologic Clinics of North America, Vol. XVI, No. 3, December 1978, pages 347-366, which is incorporated herein by reference, many of the missed lesions would be classified as T1M0 lesions, the stage of non-small cell lung cancer that Mountain, C. F. “Value of the New TNM Staging System for Lung Cancer”, 5
th
World Conference in Lung Cancer Chest, 1989 Vol. 96/1, pages 47-49, which is incorporated herein by reference, indicates has the best prognosis (42%, 5 year survival). It is this stage of lung cancer, with lesions less than 1.5 cm in diameter, and located outside the hilum region that need to be detected by a radiologist in order to improve survival rates.
Computerized techniques, such as computer aided diagnosis (CAD), have been introduced to assist in the diagnosis of lung nodules during the stage of non-small cell lung cancer. The CAD technique requires the computer system to function as a second physician to double check all the films that a primary or first physician has examined. Reduction of false positive detection is the primary objective of the CAD technique in order to improve detection accuracy.
Several CAD techniques using digital image processing and artificial neural networks have been described in numerous publications, exemplary of which are the following, which are incorporated herein by reference:
U.S. Pat. No. 4,907,156 to Doi et al. describes a method for detecting and displaying abnormal anatomic regions existing in a digital X-ray image. A single projection digital X-ray image is processed to obtain signal-enhanced image data with a maximum signal-to-noise ratio (SNR) and is also processed to obtain signal-suppressed image data with a suppressed SNR. Then, difference image data are formed by subtraction of the signal-suppressed image data from the signal-enhanced image data to remove low-frequency structured anatomic background, which is basically the same in both the signal-suppressed and signal-enhanced image data. Once the structured background is removed, feature extraction is performed. For the detection of lung nodules, pixel thresholding is performed, followed by circularity and/or size testing of contiguous pixels surviving thresholding. Threshold levels are varied, and the effect of varying the threshold on circularity and size is used to detect nodules. For the detection of mammographic microcalcifications, pixel thresholding and contiguous pixel area thresholding are performed. Clusters of suspected abnormalities are then detected. However, the algorithm described in the Doi et al. patent seems to reduce false positive rates at the expense of missing several true nodules. This prior art is limited in detection of nodules with size larger than its pre-selected size −1.5 cm. This prior art will also reduce the sensitivity by selecting fixed CDF thresholds (e.g., 97%, 94%, 91%, etc.) since some true nodules will be eliminated during this thresholding process. The algorithm described in the Doi et al. patent utilizes a single classifier (a decision tree classifier) which possesses inherent performance. A decision tree classifier performs classification by eliminating true positives in a sequential way; hence, it is easy to eliminate potential nodules in the first decision node even if the rest of the decision criteria are satisfied. Another important drawback to this prior art is that the physician has to examine every film with both true and false positives identified by the CAD system, so the time spent on the diagnosis increases dramatically.
U.S. Pat. No. 5,463,548 to Asada et al. describes a system for computer-aided differential diagnosis of diseases, and in particular, computer-aided differential diagnosis using neural networks. A first design of the neural network distinguishes between a plurality of interstitial lung diseases on the basis of inputted clinical parameters and radiographic information. A second design distinguishes between malignant and benign mammographic cases based upon similar inputted clinical and radiographic information. The neural networks were first trained using a hypothetical database made up of hypothetical cases for each of the interstitial lung diseases and for malignant and benign cases. The performance of the neural network was evaluated using receiver operating characteristics (ROC) analysis. The decision performance of the neural network was compared to experienced radiologists and achieved a high performance comparable to that of the experienced radiologists. However, Asada's method seems limited to the detection of lung diseases but not lung cancer, which presents different symptoms.
Y. S. P. Chiou, Y. M. F. Lure, and P. A. Ligomenides, “Neural Network Image Analysis and Classification in Hybrid Lung Nodule Detection (HLND) System”, Neural Networks for Processing III Proceedings of the 1993 IEEE-SP Workshop, pp. 517-526. The Chiou et al. article describes a Hybrid Lung Nodule Detection (HLND) system based on artificial neural network architectures, which is developed for improving diagnostic accuracy and speed of lung cancerous pulmonary radiology. The configuration of the HLND system includes the following processing phases: (1) pre-processing to enhance the figure-background contrast; (2) quick selection of nodule suspects based upon the most pertinent feature of nodules; and (3) complete feature space determination and neural classification of nodules. The Chiou et al. article seems to be based on U.S. Pat. No. 4,907,156 to Doi et al., but adds a neural network approach. The Chiou et al. system includes similar shortcomings to those in the Doi et al. system desc
Li Ruiping
Lin Jyh-Shyan
Lure Yuan-Ming Fleming
Xu Xin-Wei
Yeh Hwa-Young Michael
Davis George B.
Deus Technologies LLC
Gluck Jeffrey W.
Kinberg Robert
Venable LLP
LandOfFree
Fuzzy logic based classification (FLBC) method for automated... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Fuzzy logic based classification (FLBC) method for automated..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Fuzzy logic based classification (FLBC) method for automated... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3180851