Display of classifier output and confidence measure in an image
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/00
G06K-009/62
출원번호
US-0253471
(2008-10-17)
등록번호
US-8175362
(2012-05-08)
발명자
/ 주소
Sathyanarayana, Shashidhar
출원인 / 주소
Boston Scientific Scimed, Inc.
대리인 / 주소
Frommer Lawrence & Haug LLP
인용정보
피인용 횟수 :
0인용 특허 :
15
초록▼
Systems and methods are described for displaying classifier output and confidence measure in an image. The confidence measure advantageously provides additional information to the user indicating the accuracy of the classification result. Based on the classification accuracy, the user may accept or
Systems and methods are described for displaying classifier output and confidence measure in an image. The confidence measure advantageously provides additional information to the user indicating the accuracy of the classification result. Based on the classification accuracy, the user may accept or reject the classification result. In an exemplary embodiment, the classifier output is displayed on the image by color coding regions in the image based on their classifications. The confidence measure is displayed by adjusting the transparencies of the color coded regions according to their confidence measures. In one embodiment, only the classifications having confidence measures above a threshold are displayed. In other embodiments, the classifier output and confidence measure may be displayed separately, contour lines may be drawn through image regions having similar confidence measures, and the confidence measure may be displayed for a region under a pointer that the user can move within the image.
대표청구항▼
1. A method for displaying an ultrasound image, comprising: classifying regions of the ultrasound image, wherein each region is classified into one of a plurality of classes;determining a confidence measure for each of the classified regions; anddisplaying, on the ultrasound image, the classificatio
1. A method for displaying an ultrasound image, comprising: classifying regions of the ultrasound image, wherein each region is classified into one of a plurality of classes;determining a confidence measure for each of the classified regions; anddisplaying, on the ultrasound image, the classification and confidence measure for at least one of the regions including displaying contour lines on the image to indicate the confidence measure. 2. The method of claim 1, wherein displaying the classification comprises color coding at least one of the classified regions of the image based on the corresponding class. 3. The method of claim 2, wherein displaying the confidence measure comprises adjusting a transparency of each color coded region based on the corresponding confidence measure. 4. The method of claim 3, wherein adjusting the transparency comprises making the color-coded region less transparent for a higher confidence measure. 5. The method of claim 2, wherein displaying the confidence measure comprises displaying the confidence measure when a pointer is within the region. 6. The method of claim 1, wherein displaying the classification and confidence measure comprises displaying the classification and confidence measure of a region only if the confidence measure of that region is at or above a threshold. 7. The method of claim 6, wherein displaying the classification comprises color coding at least one region of the image with a confidence measure that is at or above the threshold. 8. The method of claim 7, wherein displaying the confidence measure comprises adjusting a transparency of each color coded region based on the corresponding confidence measure. 9. The method of claim 1, wherein displaying the classification and confidence measure comprises displaying the classification as color coding in at least one classified region of a first image based on the corresponding class and displaying the confidence measure of at least one classified region by adjusting an intensity level of the region in a second image based on the confidence measure. 10. A system for displaying an ultrasound image, comprising: a display;at least one processor in communication with the display; anda non-transitory computer readable storage medium having processor-executable instructions, the processor-executable instructions when installed onto the system enables the system to perform actions, comprising: classify regions of the ultrasound image, wherein each region is classified into one of a plurality of classes;determine a confidence measure for each of the classified regions; anddisplay, on the ultrasound image generated on the display, the classification and confidence measure for at least one of the regions including displaying contour lines on the image to indicate the confidence measure. 11. The system of claim 10, further comprising an image source. 12. The system of claim 11, wherein the image source is an ultrasound device configured and arranged to generate ultrasound images. 13. The system of claim 10, wherein displaying the classification comprises color coding at least one classified region of the image based on the corresponding class. 14. The system of claim 13, wherein displaying the confidence measure comprises adjusting a transparency of each color coded region based on the corresponding confidence measure. 15. The system of claim 10, wherein displaying the classification and confidence measure comprises displaying the classification and confidence measure of a region only if the confidence measure of that region is at or above a threshold. 16. A non-transitory computer readable storage medium having processor-executable instructions, the processor-executable instructions when installed onto a system enable the system to perform actions, comprising: classify regions of an ultrasound image, wherein each region is classified into one of a plurality of classes;determine a confidence measure for each of the classified regions; anddisplay, on the ultrasound image, the classification and confidence measure for at least one of the regions on the image including displaying contour lines on the image to indicate the confidence measure. 17. The non-transitory computer readable storage medium of claim 16, wherein displaying the classification comprises color coding at least one classified region of the image based on the corresponding class. 18. The non-transitory computer readable storage medium of claim 17, wherein displaying the confidence measure comprises adjusting a transparency of each color coded region based on the corresponding confidence measure.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (15)
Stupp,Steven Elliot; Carpenter,Chris; Newman,Lawrence; Lanckriet,Gert, Apparatus for determining association variables.
Eshelman, Larry J.; Gutta, Srinivas; Pelletier, Daniel; Strubbe, Hugo J.; Milanski, John, Automatic system for monitoring independent person requiring occasional assistance.
Gevins Alan S. (San Francisco CA) Morgan Nelson H. (San Francisco CA) Greer Douglas S. (San Francisco CA), Brain wave source network location scanning method and system.
Akselrod, Solange; Tepper, Ron; Zimmer, Yair, Method and system for automatic classification and quantitative evaluation of adnexal masses based on a cross-sectional or projectional images of the adnex.
Sheehan Florence H. (Mercer Island WA) Haralick Robert M. (Seattle WA) Lee Chang-Kyu (Seattle WA), Method for determining the contour of an in vivo organ using multiple image frames of the organ.
Nishikawa Robert M. ; Jiang Yulei ; Ashizawa Kazuto ; Doi Kunio, Methods for improving the accuracy in differential diagnosis on radiologic examinations.
Kramer James F. (Stanford CA) George William R. (San Francisco CA) Lindener Peter (Los Altos CA), Strain-sensing goniometers, systems and recognition algorithms.
Merickel Michael B. (Charlottesville VA) Adams Ann H. (Charlottesville VA) Brookeman James R. (Charlottesville VA), System and method for the noninvasive identification and display of breast lesions and the like.
Krishnan, Sriram; Gupta, Alok; Rao, R. Bharat; Comaniciu, Dorin; Zhou, Xiang Sean, Systems and methods for automated diagnosis and decision support for heart related diseases and conditions.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.