Microcode linker/loader that generates microcode sequences for MRI sequencer by modifying previously generated microcode sequences

• US 5,465,361 A
• Filed: 12/13/1993
• Issued: 11/07/1995
• Est. Priority Date: 09/10/1990
• Status: Expired due to Term

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Hoenninger, John C. III
• Primary Examiner(s): KIM, KENNETH S
• Application Number:
  US08/165,534
• Time in Patent Office: 694 Days
• Field of Search: 324/309, 324/312, 395/500, 395/700
• US Class Current: 717/168
• CPC Class Codes: G05B 2219/25389 Macro's, subroutines
• CPC Class Codes: G05B 2219/2652 Medical scanner
• CPC Class Codes: G06F 8/54 Link editing before load time

Abstract

An extremely fast and efficient Linker for a Magnetic Resonance Imaging (MRI) system Nuclear Magnetic Resonance (NMR) pulse control sequencer efficiently derives subsequent blocks of microcode to be loaded by using the contents of a memory buffer containing previously loaded microcode as a template, Most of the template is reused "as is". Only the relatively few field values in the microinstructions which change from one signal generation process, or cycle, to the next are replaced with new values. Offsets are tabulated of instructions which have associated multi-entry cycle indexed program change table (PCT) values. When further code is to be linked and loaded, the linker accesses the PCTs based on the table and to inserts new values into the appropriate instruction fields. The microcode memory image may be continuously maintained in a host memory buffer and re-edited successive times. The Fast Linker provided by the present invention is capable of continually loading microcode into a sequencer writable control store, and is fast enough to run under a time shared operating system at the same time a higher priority data acquisition and display process is executing.

Claims

1. In a magnetic resonance imaging system of the type including a magnetic field generator, an RF field generator, an RF receiver, a data acquisition subsystem, and a control sequencer, said magnetic field generator and said RF field generator cooperating to stimulate nuclear magnetic resonance phenomena within an object to be imaged, said RF receiver receiving NMR signals produced by said nuclear magnetic resonance phenomena, said data acquisition subsystem acquiring digitized signals, said control sequencer controlling at least one of said magnetic field generator, said RF field generator and said data acquisition subsystem, said control sequencer having a control store for storing and accessing instructions for execution,

a method of efficiently and rapidly generating instructions for execution by said control sequencer,

said method comprising:

(a) generating a sequencer control store memory image including sequencer instructions, said instructions including first parameters that define a first projection;

(b) loading said sequencer control store memory image, including said first projection parameters, into said sequencer control store;

(c) executing, with said sequencer, said sequencer control store memory image loaded into said sequencer control store by said loading step (b);

(d) controlling, with said sequencer, said magnetic field generator and/or said RF field generator in response to execution of said sequencer control store memory image by said executing step (c) so as to stimulate nuclear magnetic resonance phenomena within said object;

(e) acquiring, with said RF receiver and said data acquisition subsystem, first NMR signals emitted by said nuclear magnetic resonance phenomena stimulated by said controlling step (d);

(f) reusing said sequencer control store memory image by rapidly replacing said first projection parameters within said sequencer control store memory image with second projection parameters defining a second projection to create an altered sequence control store memory image, and loading said altered sequencer control store memory image, including said second projection parameters, into said sequencer control store;

(g) executing, with said control sequencer, said altered sequencer control store memory image loaded by said reusing step (f);

(h) controlling, with said sequencer, said magnetic field generator and/or said RF field generator in response to execution of said altered memory image by said executing step (g) so as to stimulate nuclear magnetic resonance phenomena within said object;

(i) acquiring, with said RF receiver and said data acquisition subsystem, second NMR signals emitted by said nuclear magnetic resonance phenomena stimulated by said controlling step (h); and

(j) generating an image of said object based at least in part on said first and second NMR signals acquired by said steps (e) and (i).

2. A method as in claim 1 wherein said executing step (c) includes executing microcode loaded by said loading step (b), and said reusing step (f) is performed subsequently to said executing step (c).

3. A method as in claim 1 wherein:

said loading step (b) includes the step of loading said memory image into an addressable memory space provided by said sequencer control store; and

said reusing step (f) includes the step of rewriting said control store with said altered memory image.

4. A method as in claim 1 wherein

said generating step (a) includes storing and maintaining said memory image in a host memory buffer;

and

said reusing step (f) comprises the step of editing said memory image in place in said host memory buffer.

5. A method as in claim 1 wherein said generating step (a) comprises storing a linked microcode memory image in a buffer, and said loading step (b) comprises loading said stored linked microcode memory image from said buffer into said control sequencer control store.

6. A method as in claim 1 wherein:

said generating step (a) includes;

defining an outer loop index value, and

specifying, within said memory image, plural physical addresses in dependence on said outer loop index value, and

said reusing step (f) comprises keeping constant the physical addresses within said memory image that are dependent on said outer loop index value.

7. A method as in claim 1 wherein:

said generating step (a) includes providing, within said memory image, at least one jump address to another location within said memory image, and

said reusing step (f) includes offsetting said jump address.

8. A method as in claim 1 wherein:

said generating step (a) includes providing, within said memory image, at least one jump address to a subroutine, and

said reusing step (f) includes maintaining said jump address constant.

9. A method as in claim 1 wherein said generating step (a) includes:

providing a Linker command defining an outer loop index, and

resolving Linker command nested loop dependent PCT parameter references except for PCT and other command references dependent on the outer loop index of the Linker command.

10. A method as in claim 1 wherein:

said generating step (a) comprises defining a Linker outer loop, defining a microcode image for at least a first page of the sequencer control store memory corresponding to a first part of the Linker outer loop, and identifying and resolving outer loop dependent microcode fields, and

said reusing step (f) includes reusing said microcode image for at least a second page of the control sequencer control store corresponding to at least one successive part of the Linker outer loop, and altering at least some of said outer loop dependent microcode fields.

11. A method as in claim 1 wherein said generating step (a) includes creating an instruction block, creating at least one data structure locating instructions within said instruction block that contain references dependent on an outer loop index, and repetitively copying said instruction block to form said memory image, said copying step including using said data structure to resolve at least those references dependent on said outer loop index.

12. A method as in claim 11 wherein said generating step (a) further includes storing, within said data structure, at least one address offset locating said instructions from the beginning of the block.

13. A method as in claim 11 wherein said generating step (a) further includes storing, within said data structure, at least one microcode field and at least one corresponding reference to be resolved.

14. A method as in claim 1 wherein said generating step (a) includes:

(1) linking a first main outer loop dependent block of microcode, and

(2) subsequently to said step (1), creating a template of the memory image for a first page of said sequencer control store by copying said microcode block repetitively to provide contiguous blocks of higher memory while changing addresses for jump instructions which refer to other instructions within said contiguous blocks.

15. A method as in claim 14 wherein said generating step (a) further includes placing within said first page a jump instruction to a second page of said sequencer control store if more than one page of microcode is required.

16. A method as in claim 14 wherein said generating step (a) further includes placing a stop instruction within said first page if more than one page of microcode is not required.

17. A method as in claim 14 wherein said generating step (a) further includes:

(x) defining at least one subroutine as part of said memory image; and

(y) creating a further template of a further memory image for a second page of said sequencer control store by copying said subroutine and the first main loop dependent block from said template of the memory image for said first page to a second memory image template while changing addresses of said jump instructions to reflect at least one physical address of the second page of said sequencer control store, to provide a second main loop dependent block.

18. A method as in claim 17 wherein said creating step (y) further includes repetitively copying said second main loop dependent block to higher contiguous addresses in the second memory image template while changing addresses therein to reflect said higher contiguous addresses.

19. A method as in claim 17 wherein said creating step (y) further includes placing, within said second memory image template, a jump instruction to said first sequencer control store page if more than two pages of microcode are needed.

20. A method as in claim 17 wherein said creating step (y) further includes placing a stop instruction within said second memory image template if more than two pages are not needed.

21. A method as in claim 1 wherein said generating step (a) comprises:

(1) creating a data structure containing entries referencing instructions that contain projection dependent values, and

(2) reading said data structure entries and successively resolving said projection dependent values by supplying said first projection parameters based at least in part on said data structure entries using at least one outer loop index value.

22. A method as in claim 1 wherein said rapidly replacing step is performed, at least in part, concurrently with said executing step (c).

23. A method as in claim 1 wherein said generating step (a) comprises storing said memory image in a computer memory, and said loading step (b) comprises copying said image from said computer memory to said control store.

61. A method as in claim 21 further including:

(1) creating a data structure containing entries referencing instructions that contain projection dependent values, and

(2) reading said data structure entries and successively resolving said projection dependent values by supplying said first projection parameters based at least in part on said data structure entries using at least one outer loop index value.

24. A magnetic resonance imaging system for generating an image of an object, said system including:

a magnetic field generator;

an RF field generator cooperating with said magnetic field generator to stimulate nuclear magnetic resonance phenomena within said object;

a data acquisition subsystem for acquiring digitized NMR signals;

a control sequencer for controlling at least one of said magnetic field generator, said RF field generator and said data acquisition subsystem, said control sequencer having a control store for storing and accessing instructions for execution by said sequencer;

a host computer including a computer memory, said host computer performing the following functions;

(a) generating a first block of sequencer instructions including first parameters that define a first projection,

(b) storing said first block of sequencer instructions within said computer memory,

(c) loading said first block of sequencer instructions, including said first projection parameters, into said sequencer control store,

(d) rapidly altering said first block of instructions stored within said computer memory by maintaining said first block but replacing said stored first projection parameters with second projection parameters defining a second projection, and

(e) loading said altered first block of sequencer instructions, including said second projection parameters, into said control sequencer control store;

said sequencer including a processor that executes said first block of sequencer instructions and said altered first block of sequencer instructions out of said sequencer control store, and controls said magnetic field generator and/or said RF field generator in response to said execution so as to stimulate nuclear magnetic resonance phenomena within said object, said stimulated phenomena generating NMR signals;

a receiver and said data acquisition subsystem cooperating to receive and acquire at least first NMR signals corresponding to said first projection and second NMR signals corresponding to said second projection; and

a display coupled to said data acquisition subsystem that generates an image of said object based on said acquired first and second NMR signals.

25. Apparatus as in claim 24 wherein said block of instructions specifies at least one sequencer control store physical address.

26. Apparatus as in claim 24 wherein said first parameters comprise MRI cycle and/or slice dependent values.

27. Apparatus as in claim 24 wherein said host computer includes means for (a) loading said block of instructions into said control sequencer control store, and (b) subsequently loading said altered block of instructions into said control sequencer control store.

28. Apparatus as in claim 24 wherein said host computer includes means for overwriting, with said altered block of instructions, said block of instructions loaded into said control sequencer control store.

29. A system as in claim 24 wherein said host computer includes:

means for defining an outer loop index value,

means for specifying, within said first block, plural physical addresses in dependence on said outer loop index value, and

means for keeping constant, during said rapidly altering operation, the physical addresses within said first block that are dependent on said outer loop index value.

30. A system as in claim 24 wherein said host computer includes:

means for providing, within said first block, at least one jump address to another location within said first block, and

means for offsetting said jump address during said rapidly altering operation.

31. A system as in claim 24 wherein said host computer includes:

means for providing, within said first block, at least one jump address to a subroutine, and

means for maintaining said jump address constant during said rapidly altering operation.

32. A system as in claim 24 wherein said host computer comprises:

means for providing a Linker command defining an outer loop index, and

means for resolving Linker command nested loop dependent PCT parameter references except for PCT and other command references dependent on the outer loop index of the Linker command.

33. A system as in claim 24 wherein said host computer comprises:

means for defining a Linker outer loop, defining with said first block a microcode image for at least a first page of the sequencer control store memory corresponding to a first part of the Linker outer loop, and identifying and resolving outer loop dependent microcode fields, and

means for reusing said microcode image for at least a second page of the control sequencer control store corresponding to at least one successive part of the Linker outer loop, and altering at least some of said outer loop dependent microcode fields.

34. A system as in claim 24 wherein said host computer includes means for creating at least one data structure locating instructions within said first instruction block that contain references dependent on an outer loop index, and means for repetitively copying said first instruction block to form a memory image, said copying means including means for using said data structure to resolve at least those references dependent on said outer loop index.

35. A system as in claim 34 wherein said host computer further includes means for storing, within said data structure, at least one address offset locating said instructions from the beginning of the first instruction block.

36. A system as in claim 34 wherein said host computer further includes means for storing, within said data structure, a microcode field and at least one associated reference to be resolved.

37. A system as in claim 24 wherein said host computer includes:

means for linking a first main outer loop dependent block of microcode, and

means for creating a template of the memory image for a first page of said sequencer control store by copying said microcode block repetitively to provide contiguous blocks of higher memory while changing addresses for jump instructions which refer to other instructions within said contiguous blocks.

38. A system as in claim 37 wherein said template creating means includes means for placing within said first page a jump instruction to a second page of said sequencer control store if more than one page of microcode is required.

39. A system as in claim 37 wherein said template creating means includes means for placing a stop instruction within said first page if more than one page of microcode is not required.

40. A system as in claim 37 wherein said host computer further includes:

means for defining at least one subroutine as part of said memory image; and

means for creating a further template of a further memory image for a second page of sequencer control store by copying said subroutine and the first main loop dependent block from a first memory image template to a second memory image template while changing addresses of all jump instructions to reflect at least one physical address of the second page of said sequencer control store, to provide a second main loop dependent block.

41. A system as in claim 40 wherein said creating means includes means for repetitively copying said second main loop dependent block to higher contiguous addresses in the second memory image template while changing addresses to reflect said higher contiguous addresses.

42. A system as in claim 40 wherein said creating means includes means for placing, within said second memory image template, a jump instruction to said first sequencer control store page if more than two pages are needed.

43. A system as in claim 40 wherein said creating means includes means for placing a stop instruction to said first sequencer control store page within said second memory image template if more than two pages are not needed.

44. A system as in claim 24 wherein said host computer comprises:

means for creating a data structure containing entries referencing instructions that contain projection dependent values, and

means for reading said data structure entries and successively resolving said projection dependent values by supplying said first projection parameters based at least in part on said data structure entries using at least one outer loop index value.

45. In a magnetic resonance imaging system for imaging an object, said system including:

(i) a magnetic field generator,

(ii) an RF field generator,

(iii) a data acquisition subsystem,

(iv) a digital sequencer coupled to said magnetic field generator, said RF field generator, and said data acquisition subsystem, said sequencer including a sequencer memory for storing instructions,

(v) an RF receiver,

(vi) a display device, and

(vii) a digital computing arrangement coupled to said sequencer, said digital computing arrangement including a buffer,

a method of efficiently generating and supplying instructions to said sequencer memory comprising the following steps;

(A) storing a set of sequencer instructions in said buffer;

(B) generating a first set of projection-specific parameters corresponding to a first plurality of slices of said object;

(C) writing said first set of projection-specific parameters into said buffer as parameters for at least some of said set of stored sequencer instructions;

(D) copying said stored sequencer instructions including said first set of slice-specific parameters from said buffer to said sequencer memory;

(E) executing said instructions including said first set of slice-specific parameters with said sequencer from said sequencer memory to control at least one of said magnetic field generator and said RF field generator;

(F) receiving NMR signals corresponding to said first plurality of slices of said object with said RF receiver and digitizing them with said data acquisition subsystem;

(G) generating a further set of slice-specific parameters corresponding to a further plurality of slices of said object;

(H) reusing said set of said stored sequencer instructions by;

(h1) replacing said first set of stored projection specific values with said further set of projection specific values, and (h2) copying said set of sequencer instructions including said further set of projection specific parameters from said buffer to said sequencer memory;

(I) executing said instructions copied by said reusing step (H) including said further set of projection specific parameters from said sequencer memory with said sequencer to control at least one of said magnetic field generator and said RF field generator;

(J) receiving NMR signals corresponding to said further plurality of slices with said RF receiver and said data acquisition subsystem; and

(K) generating, on said display device, an image based on said received NMR signals.

46. A method as in claim 45 wherein:

said method further includes defining an outer loop index value, and specifying, within said stored sequencer instructions, plural physical addresses in dependence on said outer loop index value, and

wherein said reusing step (H) comprises keeping constant the physical addresses within said stored sequencer instructions that are dependent on said outer loop index value.

47. A method as in claim 45 wherein:

said storing step (A) includes storing, within said buffer, an instruction having at least one jump address to another instruction within said buffer and

said reusing step (H) includes offsetting said jump address.

48. A method as in claim 45 wherein:

said storing step (A) includes storing, within said buffer, an instruction having at least one jump address to a subroutine, and

said reusing step (H) includes maintaining said jump address constant.

49. A method as in claim 45 further including:

providing a Linker command defining an outer loop index, and

resolving Linker command nested loop dependent PCT parameter references except for PCT and other command references dependent on the outer loop index of the Linker command.

50. A method as in claim 45 wherein:

said method further comprises defining a Linker outer loop, defining a microcode image for at least a first page of the sequencer control store memory corresponding to a first part of the Linker outer loop, said step (A) storing said image in said buffer, and identifying and resolving outer loop dependent microcode fields, and

said reusing step (H) includes reusing said microcode image stored in said buffer for at least a second page of the control sequencer control store corresponding to at least one successive part of the Linker outer loop, and altering at least some of said outer loop dependent microcode fields.

51. A method as in claim 45 wherein said method further includes creating an instruction block, creating at least one data structure locating instructions within said instruction block that contain references dependent on an outer loop index, and wherein said step (A) comprises repetitively copying said instruction block into said buffer, said copying step including using said data structure to resolve at least those references dependent on said outer loop index.

52. A method as in claim 51 further including storing, within said data structure, at least one address offset locating said instructions from the beginning of the block.

53. A method as in claim 51 further including storing within said data structure the microcode field and at least one reference to be resolved.

54. A method as in claim 45 further including:

(1) linking a first main outer loop dependent block of microcode, and

(2) subsequently to said step (1), creating a template of the memory image for a first page of said sequencer control store by copying said microcode block repetitively to provide contiguous blocks of higher memory while changing addresses for jump instructions which refer to other instructions within said contiguous blocks, wherein said storing step comprises storing said template in said buffer.

55. A method as in claim 54 further including placing within said first page a jump instruction to a second page of said sequencer control store if more than one page of microcode is required.

56. A method as in claim 54 further including placing a stop instruction within said first page if more than one page of microcode is not required.

57. A method as in claim 54 further including:

(x) defining at least one subroutine as part of a memory image; and

(y) creating a further template of a further memory image for a second page of sequencer control store by copying said subroutine and the first main loop dependent block from a first memory image template to a second memory image template while changing addresses of all jump instructions to reflect at least one physical address of the second page of said sequencer control store, to provide a second main loop dependent block.

58. A method as in claim 57 wherein said creating step (y) further includes repetitively copying said second main loop dependent block to higher contiguous addresses in the second memory image template while changing addresses to reflect said higher contiguous addresses.

59. A method as in claim 57 wherein said creating step (y) further includes placing, within said second memory image template, a jump instruction to said first sequencer control store page if more than two pages are needed.

60. A method as in claim 57 wherein said creating step (y) further includes placing a stop instruction to said first sequencer control store page within said second memory image template if more than two pages are not needed.

62. In a magnetic resonance imaging system including:

a magnetic field generator;

an RF field generator cooperating with said magnetic field generator to stimulate nuclear magnetic resonance phenomena within an object to be imaged;

a control sequencer for controlling in real time at least one of said magnetic field generator and said RF field generator, said control sequencer including a control store for storing and accessing instructions for execution by said sequencer;

an RF receiver for receiving NMR signals;

a data acquisition subsystem; and

a display that displays an image based at least in part on said received NMR signals,

a method of linking microcode for execution by said control sequencer, said method comprising;

(1) creating a data structure having elements locating, within a first sequencer instruction block, at least one value dependent on an outer loop index;

(2) creating a first sequencer control store memory image template by successively copying said first instruction block;

(3) accessing said data structure elements;

(4) rapidly locating and resolving outer loop index dependent values within said first template based at least in part on said accessed data structure elements;

(5) creating a second sequencer control store memory image template by successively copying said first instruction block;

(6) rapidly locating and resolving outer loop index dependent values within said second template based at least in part on said accessed data structure elements;

(7) loading said first and second control store memory images into said sequencer control store;

(8) executing said loaded first and second control store memory images with said sequencer;

(9) controlling said magnetic and RF field generators based on said executing step to generate fields stimulating NMR phenomena within said object to be imaged;

(10) receiving NMR signals with said RF receiver;

(11) digitizing said received NMR signals with said data acquisition subsystems; and

(12) generating at least one image on said display based on said digitized received NMR signals.

63. A magnetic resonance imaging method comprising:

(1) providing a data structure having elements locating, within a first instruction block, at least one value dependent on an outer loop index;

(2) creating a first sequencer control store memory image template by replicating said first instruction block;

(3) rapidly locating within said first template and resolving outer loop index dependent values based at least in part on said data structure elements;

(4) creating a second sequencer control store memory image template by replicating said first instruction block;

(5) rapidly locating within said second template and resolving outer loop index dependent values based at least in part on said data structure elements;

(6) loading said first and second templates into a sequencer control store;

(7) executing instructions within said loaded templates with a sequencer coupled to said control store;

(8) controlling, in real time, generation of a magnetic field based on said sequencer execution;

(9) controlling, in real time, generation of an RF field based on said sequencer execution;

(10) receiving NMR signals generated in response to said magnetic and RF fields;

(11) acquiring and digitizing said received NMR signals, and

(12) displaying an image based at least in part on said digitized acquired NMR signals.

64. Magnetic resonance imaging apparatus comprising:

a main memory;

a sequencer having a control store;

means coupled to said main memory for generating a first block of sequencer instructions and for storing said first block within said main memory;

data structure generating means coupled to said main memory for creating a data structure having elements locating, within said first instruction block, at least one value dependent on an outer loop index;

replicating means coupled to said main memory for creating first and second sequencer control store memory image templates by replicating and storing, in said main memory, said first instruction block;

linking means, coupled to said main memory and coupled to receive said data structure, for rapidly locating within said first and second templates and resolving outer loop index dependent values based at least in part on said data structure elements to produce first and second sequencer memory images;

loading means coupled to said linking means for loading said first and second sequencer memory images into said sequencer control store,

wherein said sequencer further includes;

executing means coupled to said control store for executing instructions within said first and second sequencer memory images loaded into said sequencer control store,

means coupled to said executing means for controlling generation of a magnetic field based on said sequencer execution, and

means coupled to said executing means for controlling generation of an RF field based on said sequencer execution,

wherein said apparatus further comprises;

an RF receiver that receives NMR signals generated in response to said magnetic and RF fields;

a data acquisition subsystem that digitizes and acquires said received NMR signals; and

a display coupled to said data acquisition subsystem for displaying an image based at least in part on said received digitized and acquired NMR signals.