Systems and methods for characterizing spatial distortion in 3d imaging systems

Abstract

The disclosure provides systems and methods for characterizing spatial distortions in location data determined by an imaging system, e.g., as employed in imaged guided therapy. A custom-formed three dimensional phantom for an imaging system includes a plurality of control points fixed in space with well-known positions. The phantom is fixed to a stereotactic frame defining a calibrated reference and imaged. The image data is analyzed to determine spatial locations of the control points. A comparison is made between the known and determined spatial locations for at least a subset of the control points to compensate for spatial distortion in the raw image data. The control points can have fixed locations known to an accuracy of about 100 μm or better. The initial estimate for the detected location of a control point can be accurate to about ±0.5 pixel or better.

1 . (canceled) 
     
     
         2 . A system for characterizing spatial errors in a medical imaging system, the system comprising:
 a substantially rigid three-dimensional phantom having a plurality of control points, wherein the control points are interspersed uniformly within and throughout a three-dimensional volume and are substantially rigidly fixed with respect to each other;   a stereotactic frame rigidly connectable to the phantom such that the control points can be positioned to occupy first known spatial locations within an imaging space of the imaging system; and   analysis hardware in communication with the imaging system so as to obtain data indicative of second measured spatial locations of at least a volumetric subset of the plurality of control points as determined by the imaging system and wherein the analysis hardware is configured to compare the first known spatial locations with the corresponding second measured spatial locations and calculate an indicia of any determined variation between the first known spatial locations and the corresponding second measured spatial locations for at least the volumetric subset of the plurality of control points.   
     
     
         3 . The system of  claim 2 , further comprising a chamber configured to enclose the phantom. 
     
     
         4 . The system of  claim 3 , wherein the chamber is configured to enclose the phantom in a substantially fluid tight manner and wherein the chamber is further provided with contrast material. 
     
     
         5 . The system of  claim 2 , wherein the analysis hardware is further configured to operate on the second measured spatial locations so as to create a corrected set of measured spatial locations having reduced spatial distortion with respect to the first known spatial locations. 
     
     
         6 . The system of  claim 2 , wherein the plurality of control points are distributed in a substantially non-uniform spacing within a sub-volume of the phantom. 
     
     
         7 . The system of  claim 2 , wherein the phantom is custom matched to the imaging system. 
     
     
         8 . The system of  claim 7 , wherein the phantom is custom matched to at least one specific imaging volume of the imaging system. 
     
     
         9 . An apparatus for characterizing volumetric spatial distortion of a medical imaging system, the apparatus comprising:
 a substantially rigid volumetric phantom comprising:
 a three-dimensional volume; 
 an outer perimeter generally enclosing the three-dimensional volume; 
 a plurality of control points that are interspersed uniformly within the outer perimeter and throughout the three-dimensional volume and that are substantially rigidly fixed with respect to each other and with respect to the outer perimeter; 
   a fixture configured to rigidly fix the phantom to a spatial reference point such that the control points occupy known spatial locations within an imaging space of the imaging system; and   analysis hardware configured to:
 communicate with the imaging system to obtain measured spatial locations of at least a volumetric subset of the plurality of control points as determined by the imaging system; 
 compare the known spatial locations with the corresponding measured spatial locations as determined by the imaging system; and 
 calculate any determined variation between the known spatial locations and the corresponding measured spatial locations for at least the volumetric subset of the plurality of control points. 
   
     
     
         10 . The apparatus of  claim 9 , wherein the plurality of control points are distributed in a substantially non-uniform spacing within a sub-volume of the phantom. 
     
     
         11 . The apparatus of  claim 9 , wherein the analysis hardware is further configured to use indicia of the determined variation to apply a correction factor, thereby correcting for volumetric spatial distortion of the imaging system. 
     
     
         12 . The apparatus of  claim 9 , wherein the plurality of control points comprise intersections of elongate rigid material. 
     
     
         13 . The apparatus of  claim 9 , wherein the three-dimensional volume has numerous control points volumetrically spaced and thereby configured to be imaged from any of a plurality of angles, the angles corresponding to treatment angles. 
     
     
         14 . The apparatus of  claim 13 , wherein the control points are spaced and configured to allow for effective calibration of a proton beam therapy system having high resolution treatment capabilities. 
     
     
         15 . The apparatus of  claim 9 , wherein the outer perimeter is generally cylindrical, the control points are interspersed in bands exhibiting cylindrical symmetry, and bands of successively wider radius have control points that are spaced farther from each other than in the previous band. 
     
     
         16 . The apparatus of  claim 9 , wherein the control points have one point to point spacing in a target subvolume and a different point to point spacing outside the target subvolume but still within the three-dimensional volume and the outer perimeter thereof. 
     
     
         17 . A non-transitory computer-readable medium with instructions stored thereon, that when executed by one or more hardware processors, perform the steps comprising:
 applying a three dimensional low pass filter to an original image volume resulting from an imaging operation of a three dimensional structure having a plurality of control points;   determining a selected kernel matched to the three dimensional structure;   convoluting the filtered image volume with the selected kernel to increase the intensity of a plurality of pixels of the filtered image volume that are at the control points;   locating local maxima for the convoluted image;   comparing at least a subset of first known spatial locations of the control points with corresponding located local maxima; and   converting the original image volume to a corrected image volume with reduced spatial distortion with respect to location data.   
     
     
         18 . The non-transitory computer storage medium of  claim 17 , wherein the instructions, when executed by one or more hardware processors, perform the process of calculating indicia of any indicated spatial distortion between the first known spatial locations and the corresponding located local maxima. 
     
     
         19 . The non-transitory computer storage medium of  claim 17 , wherein the instructions, when executed by one or more hardware processors, perform the process of resampling image intensity curves in multiple dimensions. 
     
     
         20 . The non-transitory computer storage medium of  claim 19 , wherein the instructions, when executed by one or more hardware processors, perform the process of making a decision whether or not to iterate previously performed processes and, when it is decided to iterate previously performed processes, iterating the processes of:
 convoluting the filtered image volume with the selected kernel;   locating local maxima for the convoluted image; and   resampling image intensity curves in multiple dimensions and, when it is decided not to iterate, to perform the process of using a center of gravity of a last set of resampled intensity curves as estimates for a spatial coordinate point.   
     
     
         21 . The non-transitory computer storage medium of  claim 20 , wherein the instructions, when executed by one or more hardware processors, perform the process of calculating indicia of any indicated spatial distortion between the first known spatial locations and the corresponding located local maxima. 
     
     
         22 . A computer-implemented method of correcting spatial distortion in location data determined by an imaging system, the method comprising:
 applying a three dimensional low pass filter to an original image volume resulting from an imaging operation of a three dimensional structure having a plurality of control points;   determining a selected kernel matched to the three dimensional structure;   convoluting the filtered image volume with the selected kernel to increase the intensity of a plurality of pixels of the filtered image volume that are at the control points;   locating local maxima for the convoluted image;   comparing at least a subset of first known spatial locations of the control points with corresponding located local maxima; and   converting the original image volume to a corrected image volume with reduced spatial distortion with respect to location data.   
     
     
         23 . The method of  claim 22 , further comprising calculating indicia of any indicated spatial distortion between the first known spatial locations and the corresponding located local maxima. 
     
     
         24 . The method of  claim 22 , further comprising resampling image intensity curves in multiple directions. 
     
     
         25 . The method of  claim 24 , further comprising making a decision whether or not to iterate previously performed processes and, when it is decided to iterate previously performed processes, iterating the processes of:
 convoluting the filtered image volume with the selected kernel;   locating local maxima for the convoluted image; and   resampling image intensity curves in multiple dimensions and, when it is decided not to iterate, to perform the process of using a center of gravity of a last set of resampled intensity curves as estimates for a spatial coordinate point.   
     
     
         26 . The method of  claim 25 , further comprising calculating indicia of any indicated spatial distortion between the first known spatial locations and the corresponding located local maxima. 
     
     
         27 . A system for correcting spatial distortion in location data determined by an imaging system, comprising:
 a processor that
 applies a three dimensional low pass filter to an original image volume resulting from an imaging operation of a three dimensional structure having a plurality of control points; 
 determines a selected kernel matched to the three dimensional structure; and 
 convolutes the filtered image volume with the selected kernel; 
   a memory for storing the convoluted image;   wherein the processor further:
 locates local maxima for the convoluted image to provide automated detection of phantom control points; 
 compares at least a subset of first known spatial locations of the control points with corresponding located local maxima; and 
 converts the original image volume to a corrected image volume with reduced spatial distortion with respect to location data. 
   
     
     
         28 . The system of  claim 27 , wherein the processor further resamples image intensity curves in multiple dimensions. 
     
     
         29 . The system of  claim 28 , wherein the processor further decides whether to or not to iterate previously performed processes and when it is decided to iterate previously performed processes, iterates the process of:
 convoluting the filtered image volume with the selected kernel;   locating local maxima for the convoluted image; and   resampling image intensity curves in multiple dimensions and, when it is decided not to iterate, to perform the process of using a center of gravity of a last set of resampled intensity curves as estimates for a spatial coordinate point.   
     
     
         30 . The system of  claim 27 , wherein the processor further calculates indicia of any indicated spatial distortion between the first known spatial locations and the corresponding located local maxima.