Back to results
US 9,002,081 B2Granted
Computer-aided bone scan assessment with automated lesion detection and quantitative assessment of bone disease burden changes
OwnerMatthew Sherman Browninventor
Life of this patent
Estimating the term…
Abstract
A computer aided bone scan assessment system and method provide automated lesion detection and quantitative assessment of bone disease burden changes.
Claims26
An automated method for processing a subject bone scan image and quantifying bone lesion burden, the method comprising the steps of:
- providing a subject bone scan image formed from pixels;
- performing atlas-based anatomic segmentation of the image in order to identify a set of anatomic regions included on the image;
- normalizing the intensity of the image such that normal bone in the image has an intensity that corresponds with that of normal bone in one or more reference bone scan images;
- detecting bone lesions in each region of the image by comparing the intensities of pixels in the region with a region specific intensity threshold derived from the one or more reference bone scan images;
- and,
- quantifying bone lesion burden using characteristics of a set of pixels corresponding to detected bone lesions.
3 more elements
An automated method for processing a subject bone scan image and quantifying bone lesion burden, the method comprising the steps of:
- scanning a bony anatomical structure with a scanner to produce a subject bone scan image formed by pixels having intensities indicating rates of bone metabolism;
- using an anatomical atlas, anatomically segmenting the subject image into regions;
- normalizing the intensities of subject image pixels using a normal bone indicium from a set of reference bone scan images and a normal bone indicium from the subject image;
- detecting a number of lesion pixels, if any, in the subject image using subject image pixel intensity and reference image derived intensity thresholds that are region specific;
- and,
- from characteristics of the lesion pixels, if any, quantifying bone lesion burden.
3 more elements
A device for processing a subject bone scan image and quantifying bone lesion burden, the device comprising:
- a radiotracer type scanner, a processor, and digital data memory;
- an anatomic atlas stored in memory;
- from reference scans
- a normal bone indicium is derived and stored in memory and
- a set of lesion indicating intensity thresholds that are region specific is derived and stored in memory;
- the scanner operable to acquire a subject bone scan image formed from pixels;
- the processor operable to use the anatomic atlas to anatomically segment the subject image;
- the processor operable to use the normal bone indicium to normalize the intensities of pixels in the subject image;
- the processor operable use the intensity thresholds to detect lesion pixels;
- and,
- the processor operable to calculate quantitative lesion burden metrics from lesion pixel characteristics.
8 more elements
A method for processing bone scan images and quantifying bone lesion burden of a particular patient, the method comprising the steps of:
- from a set of reference images, determining a normalizing constant from the intensities of normal bone in the image;
- from the set of reference images, determining for each of multiple atlas based regions a corresponding bone lesion intensity threshold;
- for a patient first image, processing the image using the normalizing constant and the set of bone lesion intensity thresholds;
- for a patient second image made at least weeks after the first image, processing the image using the normalizing constant and the set of bone lesion intensity thresholds;
- and,
- indicating patient response by comparing a bone lesion burden corresponding with the first image and a bone lesion burden corresponding with the second image.
3 more elements
Dependencies are inferred from the claim text.