Image analysis – Applications – Biomedical applications
Reexamination Certificate
1997-09-05
2001-03-06
Au, Amelia (Department: 2623)
Image analysis
Applications
Biomedical applications
C382S224000, C128S922000
Reexamination Certificate
active
06198839
ABSTRACT:
METHOD AND APPARATUS
The invention relates to a dynamic control and decision making method and apparatus for an image analyzer and more particularly to a method and apparatus that improves system performance of an automated, image based biological specimen analysis system by increasing processing speed and prescreening accuracy.
BACKGROUND OF THE INVENTION
Systems that process image data at rates acceptable for automated diagnostic prescreening, automated diagnostic screening and automated diagnostic screening quality control include the system disclosed in U.S. Pat. No. 5,315,700, entitled “Method And Apparatus For Rapidly Processing Data Sequences”, by Richard S. Johnston et al. issued May 24, 1994 which is incorporated by reference hereto. These systems process images of biological specimen slides such as Pap smear slides. The biological specimen is taken from a patent that is part of a patient population. These systems automatically review the slide and provide an analysis score. The performance of these systems and similar image analysis systems depend on many factors including: the cellular materials composition of the slides to be screened; the speed of the system to scan and process images; the regular patient population composition; the operational mode of the system including system specificity and system sensitivity; and the variations between specimens. The slide processing speed of such systems determines their capacity and thus their operational cost. These systems also have associated signal to noise characteristics. The signal comprises the abnormal cellular material on the processed slides. The noise comprises the artifact or normal cellular material misclassified as abnormal material by these systems. Poor signal to noise characteristics adversely effect the classification effectiveness of these systems.
For example, the task of examining biological specimen slides for the prescreening of cervical cancer demonstrates the need for increasing processing speed and accuracy of automated specimen screeners. Each biological specimen slide exhibits large variability in abnormal cell prevalence. To routinely achieve the high sensitivity required on low-prevalence abnormal specimens, specimens that have a low number of abnormal material, these systems must process a significant number, if not all, of the images taken of the biological specimen slide. Consequently, the number of images these systems must process determines system throughput. In the prior art, the number of images to process is predetermined based on predefined criteria. As a result, the prior art treated each biological specimen slide identically, disregarding data collected from the slide during processing, using a simple sequential test methodology. In some instances this may degrade the signal to noise characteristics of the analysis by including noisy information. The invention recognizes for the first time that an automated analysis achieves optimum signal to noise characteristics during processing. The invention further determines when to stop processing to prevent counterproductive analysis.
Therefore, the invention dynamically processes the biological specimen slide based on data collected from the slide to achieve higher accuracy as well as increased system throughput.
SUMMARY OF THE INVENTION
The invention provides a dynamic decision making method for processing a biological specimen. A computer acquires an image of the biological specimen. The computer processes the image to extract a feature from the image. Using the extracted feature, the computer dynamically chooses additional image processing modules to operate on the image from a predetermined set of image processing modules. The computer may further dynamically adjust the image processing steps performed on the image. The computer then determines whether to acquire a second image and selects the image modules to operate on the second image. The invention thereby provides for enhanced throughput by avoiding unnecessary additional computation.
The invention further provides a dynamic slide classifier for improving system performance of an automated system by increasing processing speed and prescreening accuracy. The dynamic slide classifier includes a means for slide scoring having a process control input and a slide score processing result output; and a means for process control connected to the slide score processing result output, wherein the means for process control is connected to the process control input. In one embodiment, the means for slide scoring further comprises: means for image focusing and acquisition having an image output; means for image processing and feature extraction connected to the image output having a processed image output; and means for processing the processed image output having a control input connected to the process control input. In an alternate embodiment, the means for process control further comprises: means for score calculating connected to the slide score processing result output, wherein the means for score calculating has a review specimen output and a normal specimen output; means for making an automatic inference connected to the slide score processing result output; and means for controlling the means for slide scoring having a control output connected to the process control input.
The invention also provides a method of image scanning and processing comprising the steps of: scanning and processing images based on a current prioritized image scan sequence; terminating slide processing when a classification decision is made; changing the current prioritized image scan sequence to a new prioritized scan sequence; disabling a subset of image processing to reduce unnecessary computation; enabling a new set of image processing to collect new information from the slide; and rescanning and processing certain areas using different image processing.
Other objects, features and advantages of the present invention will become apparent to those skilled in the art through the description of the preferred embodiment, claims and drawings herein wherein like numerals refer to like elements.
REFERENCES:
patent: 4741043 (1998-04-01), Bacus
patent: 4965725 (1990-10-01), Rutenberg
patent: 5023785 (1991-06-01), Adrion et al.
patent: 5235522 (1993-08-01), Bacus
patent: 5249274 (1993-09-01), Sztipanovits et al.
patent: 5287272 (1994-02-01), Rutenberg et al.
patent: 5315700 (1994-05-01), Johnston et al.
patent: 5341507 (1994-08-01), Terada et al.
patent: 5361140 (1994-11-01), Hayenga et al.
patent: 5499097 (1996-03-01), Ortyn et al.
patent: 5528703 (1996-06-01), Lee
patent: 5557097 (1996-09-01), Ortyn et al.
patent: 5757954 (1998-05-01), Kuan et al.
patent: 5797130 (1998-08-01), Nelson et al.
patent: 5799101 (1998-08-01), Lee et al.
patent: 5828776 (1998-10-01), Lee et al.
Davis “Knowledge-Based Cephalometric Analysis: A comparison with Clinicians Using Interactive Computer Methods” Computer and Biomedical Research, vol. 27, pp. 210-218, Sep. 1994.
Hayes-Roth, Barbara, “A Blackboard Architecture for Control”,Artificial Intelligence,1985, pp. 251-321.
Hull, Jonathan et al., “A Blackboard-based Approach to Handwritten ZIP Code Recognition”, State University of New York at Buffalo, Department of Computer Science, Buffalo, NY 14260, pp. 1018-1032.
Nii, H. Penny, “Part One: Blackboard Systems: The Blackboard Model of Problem Solving and the Evolution of Blackboard Architectures”,The AI Magazine,Summer 1986, pp. 38-53.
Reexamination Certificate, U.S. Patent B1 4, 741, 043 to Bacus. Certificate issued Aug. 9, 1994.
Bannister Wendy R.
Kuan Chih-Chau L.
Lee Shih-Jong J.
Meyer Michael G.
Oh Seho
Au Amelia
Leone George
Moffa Emil
Sun Hans
TriPath Imaging, Inc.
LandOfFree
Dynamic control and decision making method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dynamic control and decision making method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic control and decision making method and apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2484426