Computer-aided detection of regions of interest in tomographic breast imagery

Abstract

Disclosed are methods, and associated systems comprising processors, input devices and output devices, of detecting regions of interest in a tomographic breast image. The methods may comprise: acquiring tomographic breast image data; deriving a plurality of synthetic sub-volumes from the tomographic breast image data; wherein each subvolume is defined by parallel planar top and bottom surfaces; wherein planar top and bottom surfaces of successive subvolumes are parallel to each other; and wherein a top planar surface of a sub-volume is offset from a top planar surface of a prior sub-volume, such that successive sub-volumes overlap; for each sub-volume, deriving a two-dimensional image; for each two-dimensional image, identifying regions of interest therein; deriving at least one region of interest of potential clinical interest from a plurality of identified regions of interest; and outputting information associated with at least one derived region of interest of potential clinical interest.

1 . In a system comprising a processor, an input device and an output device, a method of detecting regions of interest in a tomographic breast image, comprising:
 a. by means of an input device, acquiring tomographic breast image data;   b. by means of a processor, deriving a plurality of synthetic sub-volumes from the tomographic breast image data;   wherein each subvolume is defined by parallel planar top and bottom surfaces;   wherein planar top and bottom surfaces of successive subvolumes are parallel to each other; and   wherein a top planar surface of a sub-volume is offset from a top planar surface of a prior sub-volume, such that successive sub-volumes overlap;   c. by means of a processor, for each sub-volume, deriving a two-dimensional image;   d. by means of a processor, for each two-dimensional image, identifying regions of interest therein:   e. by means of a processor, deriving at least one region of interest of potential clinical interest from a plurality of identified regions of interest; and   f. by means of an output device, outputting information associated with at least one derived region of interest of potential clinical interest.   
     
     
         2 . The method of  claim 1 , wherein the tomographic breast image data is acquired by means of an image acquisition unit. 
     
     
         3 . The method of  claim 1 , wherein the tomographic breast image data comprises a tomographic breast volume. 
     
     
         4 . The method of  claim 3 , wherein the tomographic breast volume is acquired by means of an image acquisition unit obtaining a plurality of two-dimensional breast images of an anatomical breast at differing angles, and a processor computing the tomographic breast volume from the plurality of two-dimensional breast images. 
     
     
         5 . The method of claim I , wherein the tomographic breast image data is acquired from at least one of a computer network and a storage device. 
     
     
         6 . The method of  claim 5 , wherein the tomographic breast image data comprises a tomographic breast volume. 
     
     
         7 . The method of  claim 6 , wherein the tomographic breast volume is acquired by obtaining a plurality of two-dimensional breast images of an anatomical breast at differing angles by means of at least one of a computer network and a storage device, computing the tomographic breast volume from the plurality of two-dimensional breast images by means of a processor. 
     
     
         8 . The method of claim I, wherein deriving at least one region of interest of potential clinical interest from a plurality of identified regions of interest comprises:
 e1. for each region of interest identified in a plurality of two-dimensional images, determining a location of the region of interest;   e2. for each located region of interest, deriving a further sub-volume enclosing said located region of interest; and   e3. for each located region of interest, further evaluating said located region of interest to determine if it is of potential clinical interest.   
     
     
         9 . The method of  claim 1 , wherein each subvolume has a same thickness as all other subvolumes. 
     
     
         10 . The method of  claim 9 , wherein the same thickness is a predetermined thickness. 
     
     
         11 . The method of claim wherein each subvolume is offset from the prior subvolume by a same offset amount. 
     
     
         12 . The method of  claim 11 , wherein the same offset amount is a predetermined offset amount. 
     
     
         13 . The method of  claim 1 , wherein outputting information associated with at least one derived region of interest of potential clinical interest comprises displaying said at least one derived region of interest in conjunction with at least a portion of a breast image. 
     
     
         14 . The method of  claim 13 , wherein information associated with at least one derived region of interest of potential clinical interest comprises at least one CAD mark. 
     
     
         15 . The method of  claim 13 , wherein information associated with at least one derived region of interest of potential clinical interest comprises a location of the said at least one region of interest. 
     
     
         16 . The method of  claim 1 , wherein each synthetic sub-volume is derived from a plurality of consecutive thin slices of the tomographic breast volume. 
     
     
         17 . The method of  claim 16 , wherein each thin slice is in the range of about 1 mm, about 3 mm. thick. 
     
     
         18 . The method of  claim 1 , wherein a thickness of each subvolume is in the range of about 3 mm. to about 30 mm. 
     
     
         19 . The method of  claim 18 , wherein an offset between top planar surfaces of successive sub-volumes is in the range of about 1 mm. to about 3 mm. 
     
     
         20 . The method of  claim 18 , wherein an offset between top planar surfaces of successive sub-volumes is equal to a thickness of a thin slice of a tomographic breast volume. 
     
     
         21 . The method of  claim 1 , wherein each two-dimensional image is derived performing an intensity projection algorithm on a subvolume. 
     
     
         22 . The method of  claim 1 , wherein each region of interest is identified by executing a suspicious lesion detection algorithm on a two-dimensional image. 
     
     
         23 . The method of  claim 22 , wherein the suspicious lesion detection algorithm identifies regions of interest comprising microcalcifications, density masses, and/or spiculated masses. 
     
     
         24 . A system for detecting regions of interest in a tomographic breast image, comprising:
 at least one input device, configured to acquire tomographic breast image data;   at least one processor, configured to:   a. derive a plurality of synthetic sub-volumes from the tomographic breast image data;   wherein each subvolume is defined by parallel planar top and bottom surfaces;   wherein planar top and bottom surfaces of successive subvolumes are parallel to each other; and   wherein a top planar surface of a sub-volume is offset from a top planar surface of a prior sub-volume, such that successive sub-volumes overlap;   b. for each sub-volume, derive a two-dimensional image;   c. for each two-dimensional image, identify regions of interest therein;   d. derive at least one region of interest of potential clinical interest from a plurality of identified regions of interest; and   at least one output device, configured to output information associated with at least one derived region of interest of potential clinical interest.   
     
     
         25 . The system of  claim 24 , wherein the tomographic breast image data is acquired by means of an image acquisition unit. 
     
     
         26 . The system of  claim 24 , wherein the tomographic breast image data comprises a tomographic breast volume. 
     
     
         27 . The system of  claim 26 , wherein the tomographic breast volume is acquired by means of an image acquisition unit obtaining a plurality of two-dimensional breast images of an anatomical breast at differing angles, and a processor computing the tomographic breast volume from the plurality of two-dimensional breast images. 
     
     
         28 . The system of  claim 24 , wherein the tomographic breast image data is acquired from at least one of a computer network and a storage device. 
     
     
         29 . The system of  claim 28 , wherein the tomographic breast image data comprises a tomographic breast volume. 
     
     
         30 . The system of  claim 29 , wherein the tomographic breast volume is acquired by obtaining a plurality of two-dimensional breast images of an anatomical breast at differing angles by means of at least one of a computer network and a storage device, computing the tomographic breast volume from the plurality of two-dimensional breast images by means of a processor. 
     
     
         31 . The system of  claim 24 , wherein deriving at least one region of interest of potential clinical interest from a plurality of identified regions of interest comprises:
 d1. for each region of interest identified in a plurality of two-dimensional images, determining a location of the region of interest;   d2. for each located region of interest, deriving a further sub-volume enclosing said located region of interest; and   d3. for each located region of interest, further evaluating said located region of interest to determine if it is of potential clinical interest.   
     
     
         32 . The system of  claim 24 , wherein each subvolume has a same thickness as all other subvolumes. 
     
     
         33 . The system of  claim 32 , wherein the same thickness is a predetermined thickness. 
     
     
         34 . The system of  claim 24 , wherein each subvolume is offset from the prior subvolume by a same offset amount. 
     
     
         35 . The system of  claim 34 , wherein the same offset amount is a predetermined offset amount. 
     
     
         36 . The system of  claim 24 , wherein outputting information associated with at least one derived region of interest of potential clinical interest comprises displaying said at least one derived region of interest in conjunction with at least a portion of a breast image, 
     
     
         37 . The system of  claim 36 , wherein information associated with at least one derived region of interest of potential clinical interest comprises at least one CAD mark. 
     
     
         38 . The system of  claim 36 , wherein information associated with at least one derived region of interest of potential clinical interest comprises a location of the said at least one region of interest. 
     
     
         39 . The system of  claim 24 , wherein each synthetic sub-volume is derived from a plurality of consecutive thin slices of the tomographic breast volume. 
     
     
         40 . The system of  claim 39 , wherein each thin slice is in the range of about 1 mm. to about 3 mm. thick. 
     
     
         41 . The system of  claim 24 , wherein a thickness of each subvolume is in the range of about 3 mm. to about 30 mm. 
     
     
         42 . The system of  claim 41 , wherein an offset between top planar surfaces of successive sub-volumes is in the range of about 1 mm. to about 3 mm. 
     
     
         43 . The system of  claim 41 , wherein an offset between top planar surfaces of successive sub-volumes is equal to a thickness of a thin slice of a tomographic breast volume. 
     
     
         44 . The system of  claim 24 , wherein each two-dimensional image is derived by performing an intensity projection algorithm on a subvolume. 
     
     
         45 . The system of  claim 24 , wherein each region of interest is identified by executing a suspicious lesion detection algorithm on a two-dimensional image. 
     
     
         46 . The system of  claim 45 , wherein the suspicious lesion detection algorithm identifies regions of interest comprising microcalcifications, density masses, and/or spiculated masses.