US6320976B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Computer-assisted diagnosis method and system for automatically determining diagnostic saliency of digital images
Assignee: Siemens AG
•
Filing: 1999-04-01
•
Granted: 2001-11-20
•
Expires: 2019-04-01
Abstract
Abstract A computer-assisted diagnosis method and system are provided for automatically determining diagnostic saliency of digital images. The method includes the step of providing filters for evaluating the image. Each filter is designed to identify a specific type of diagnostic finding, and is associated with the following: a virtual window for defining regions in the image at which the filter is applied; a set of training image patches corresponding to typical appearances of the specific type of diagnostic finding; a distance measure between the training image patches and the regions in the image defined by the virtual window; and a feature set corresponding to the distance measure. The filters are applied to the image to compute distances between the regions in the image defined by the virtual window and the training image patches based on the distance measure and the feature set, for each of the plurality of filters. Regions in the image are ranked as corresponding to a particular type of diagnostic finding based the computed distances.
Inventors
- 👤 Sreerama K. V. Murthy
- 👤 Jianzhong Qian
- 👤 Carol L. Novak
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 22:15
CPC Classifications
A; A61; A61B; A61B5/00; A61B5/0002; A61B6/00; A61B6/56; G; G06; G06T; G06T7/00; G06T7/0002; G06T7/0012; G06V; G06V10/00; G06V10/20; G06V10/25; A61B6/50; A61B6/502; A61B8/00; A61B8/08; A61B8/0825; A61B8/56; A61B8/565; G06V2201/00; G06V2201/03; G06V2201/032
Claims Analysis
Total Claims: 39
Claim 1
Independent Claim
180 words
A computer assisted diagnosis system for automatically determining diagnostic saliency of regions in a digital image comprised of a plurality of pixels, the system comprising: a memory unit; a plurality of filters stored in said memory unit, each of the plurality of filters designed to identify a specific type of diagnostic finding; a virtual window for defining regions in the image at which each of the plurality of filters is applied; a set of training image patches corresponding to typical appearances of the specific type of diagnostic finding; a distance measure set corresponding to the training image patches and the regions in the image defined by the virtual window; a feature set corresponding to the distance measure set; and a processor for applying each of the plurality of filters to the image to compute distances between the regions in the image defined by the virtual window and the training image patches bases on the distance measure set and the feature set, and ranking regions in the image corresponding to a particular type of diagnostic finding based on the computed distances.
Claim 2
Dependent Claim
180 words
A computer assisted diagnosis system for automatically determining diagnostic saliency of regions in a digital image comprised of a plurality of pixels, the system comprising: a memory unit; a plurality of filters stored in said memory unit, each of the plurality of filters designed to identify a specific type of diagnostic finding; a virtual window for defining regions in the image at which each of the plurality of filters is applied; a set of training image patches corresponding to typical appearances of the specific type of diagnostic finding; a distance measure set corresponding to the training image patches and the regions in the image defined by the virtual window; a feature set corresponding to the distance measure set; and a processor for applying each of the plurality of filters to the image to compute distances between the regions in the image defined by the virtual window and the training image patches bases on the distance measure set and the feature set, and ranking regions in the image corresponding to a particular type of diagnostic finding based on the computed distances.
Claim 3
Dependent Claim
27 words
The system according to claim 1 , wherein said processor applies each of the plurality of filters so as to evaluate each of the plurality of pixels.
Claim 4
Dependent Claim
43 words
The system according to claim 1 , wherein a response of a filter is inversely proportional to a distance between a region in the image defined by the virtual window and a closest training image patch in the set of training image patches.
Claim 5
Dependent Claim
32 words
The system according to claim 1 , wherein said processor applies some of the plurality of filters to a multi-resolution image pyramid constructed from the image, and aggregates results between pyramid levels.
Claim 6
Dependent Claim
19 words
The system according to claim 1 , wherein said processor normalizes outputs of each of the plurality of filters.
Claim 7
Dependent Claim
28 words
The system according to claim 1 , further comprising an adjusting device for one of uniformly and independently adjusting a sensitivity of each of the plurality of filters.
Claim 8
Dependent Claim
28 words
The system according to claim 1 , wherein said processor identifies and ranks foci of attention belonging to a particular type of diagnostic finding using individual filter responses.
Claim 9
Dependent Claim
26 words
The system according to claim 1 , wherein said processor combines outputs of the plurality of filters into a single diagnostic saliency map for the image.
Claim 10
Dependent Claim
43 words
The system according to claim 8 , wherein said processor combines the outputs of the plurality of filters using winner-take-all operators between the regions in the image defined by the virtual windows, and one of aggregation operators and winner-take-all operators between different filters.
Claim 11
Dependent Claim
53 words
The system according to claim 8 , wherein the diagnostic map comprises a numeric diagnostic saliency score for each of the regions in the image defined by the virtual windows associated with the plurality of filters, the specific type of diagnostic finding at each of the regions and characteristics of the diagnostic finding.
Claim 12
Dependent Claim
31 words
The system according to claim 8 , wherein said processor identifies and ranks diagnostic foci of attention in the image at a given level of sensitivity, using the diagnostic saliency map.
Claim 13
Dependent Claim
33 words
The system according to claim 1 , wherein said processor generates a numeric diagnostic saliency score for the regions in the image defined by the virtual windows associated with the plurality of filters.
Claim 14
Dependent Claim
42 words
The system according to claim 1 , wherein said processor pre-processes each of the training image patches in each set of training image patches to optimize a time for distance computation with a region in the image defined by the virtual window.
Claim 15
Dependent Claim
20 words
The system according to claim 1 , wherein said processor removes salt-and-pepper type of noise from outputs of the filters.
Claim 16
Dependent Claim
26 words
The system according to claim 1 , wherein said processor identifies an individual training image patch based on an output of the individual training image patch.
Claim 17
Dependent Claim
33 words
The system according to claim 1 , wherein said processor integrates domain knowledge rules into one of normalizing outputs of the plurality of filters and generating a diagnostic saliency map for the image.
Claim 18
Dependent Claim
27 words
The system according to claim 1 , wherein said distance measure set is specific to the specific type of diagnostic finding identified by said plurality of filters.
Claim 19
Dependent Claim
26 words
The system according to claim 1 , wherein said feature set is specific to the specific type of diagnostic finding identified by the plurality of filters.
Claim 20
Dependent Claim
28 words
The system according to claim 1 , wherein at least some of said plurality of filters are specific to at least one predetermined resolution of the digital image.
Claim 21
Dependent Claim
17 words
The system according to claim 1 , wherein each of said plurality of filters is disease specific.
Claim 22
Dependent Claim
26 words
The system according to claim 1 , wherein the digital image is a digital mammogram and the specific type of diagnostic finding corresponds to breast cancer.
Claim 23
Dependent Claim
187 words
A computer assisted diagnosis method for automatically determining diagnostic saliency of locations in a digital image comprised of a plurality of pixels, the method comprising the steps of: providing a plurality of filters for evaluating the image, wherein each of the plurality of filters is designed to identify a specific type of diagnostic finding; providing a virtual window for defining regions in the image at which each of the plurality of filters is applied; providing a set of training image patches corresponding to typical appearances of the specific type of diagnostic finding; providing a distance measure set corresponding to the training image patches and the regions in the image defined by the virtual window; providing a feature set corresponding to the distance measure set; applying the plurality of filters to the image to compute distances between the regions in the image defined by the virtual window and the training image patches based on the distance measure set and the feature set, for each of the plurality of filters; and ranking regions in the image as corresponding to a particular type of diagnostic finding based the computed distances.
Claim 24
Dependent Claim
187 words
A computer assisted diagnosis method for automatically determining diagnostic saliency of locations in a digital image comprised of a plurality of pixels, the method comprising the steps of: providing a plurality of filters for evaluating the image, wherein each of the plurality of filters is designed to identify a specific type of diagnostic finding; providing a virtual window for defining regions in the image at which each of the plurality of filters is applied; providing a set of training image patches corresponding to typical appearances of the specific type of diagnostic finding; providing a distance measure set corresponding to the training image patches and the regions in the image defined by the virtual window; providing a feature set corresponding to the distance measure set; applying the plurality of filters to the image to compute distances between the regions in the image defined by the virtual window and the training image patches based on the distance measure set and the feature set, for each of the plurality of filters; and ranking regions in the image as corresponding to a particular type of diagnostic finding based the computed distances.
Claim 25
Dependent Claim
28 words
The method according to claim 22 , wherein said applying step applies each of the plurality of filters so as to evaluate each of the plurality of pixels.
Claim 26
Dependent Claim
43 words
The method according to claim 22 , wherein a response of a filter is inversely proportional to a distance between a region in the image defined by the virtual window and a closest training image patch in the set of training image patches.
Claim 27
Dependent Claim
44 words
The method according to claim 22 , further comprising the step of pre-processing each of the training image patches in each set of training image patches to optimize a time for distance computation with a region in the image defined by the virtual window.
Claim 28
Dependent Claim
41 words
The method according to claim 22 , further comprising the steps of: constructing a multi-resolution image pyramid from the image; applying some of the plurality of filters to the multi-resolution image pyramid; and aggregating results between pyramid levels for each filter.
Claim 29
Dependent Claim
21 words
The method according to claim 22 , further comprising the step of normalizing outputs of each of the plurality of filters.
Claim 30
Dependent Claim
26 words
The method according to claim 22 , further comprising the step of adjusting sensitivities of the plurality of filters to be one of uniform and independent.
Claim 31
Dependent Claim
30 words
The method according to claim 22 , further comprising the steps of identifying and ranking foci of attention belonging to a particular type of diagnostic finding using individual filter responses.
Claim 32
Dependent Claim
28 words
The method according to claim 22 , further comprising the step of combining outputs of the plurality of filters into a single diagnostic saliency map for the image.
Claim 33
Dependent Claim
33 words
The method according to claim 22 , further comprising the steps of identifying and ranking diagnostic foci of attention in the image at a given level of sensitivity, using the diagnostic saliency map.
Claim 34
Dependent Claim
45 words
The method according to claim 22 , further comprising the step of combining the outputs of the plurality of filters using winner-take-all operators between the regions in the image defined by the virtual windows, and one of aggregation operators and winner-take-all operators between different filters.
Claim 35
Dependent Claim
22 words
The method according to claim 22 , further comprising the step of removing salt-and-pepper type of noise from outputs of the filters.
Claim 36
Dependent Claim
40 words
The method according to claim 22 , further comprising the step of generating a numeric diagnostic saliency score for the regions in the image defined by the virtual windows associated with the plurality of filters based on the computed distances.
Claim 37
Dependent Claim
72 words
The method according to claim 22 , further comprising the step of combining outputs of the plurality of filters into a single diagnostic saliency map for the image, the map comprising a numeric diagnostic saliency score for each of the regions in the image defined by the virtual windows associated with the plurality of filters, the specific type of diagnostic finding at each of the regions and characteristics of the diagnostic finding.
Claim 38
Dependent Claim
28 words
The method according to claim 22 , further comprising the step of identifying an individual training image patch based on an output of the individual training image patch.
Claim 39
Dependent Claim
35 words
The method according to claim 22 , further comprising the step of integrating domain knowledge rules into one of normalizing outputs of the plurality of filters and generating a diagnostic saliency map for the image.
References Cited (Backward Citations)
US5287275A
Cited
US5822742A
Cited
US5870493A
Cited
US5680485A
Cited
US5710833A
Cited
US5857030A
Cited
US5923779A
Cited
US5828769A
Cited
US5983218A
Cited
US6058206A
Cited
US6163629A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2004-11-19 | 2005-05-21 | 2005-11-20 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2008-11-19 | 2009-05-21 | 2009-11-20 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2012-11-19 | 2013-05-21 | 2013-11-20 | $7700.00 | Window Closed / Lapsed |