System and method for analyzing brain tissue components based on magnetic resonance image
A system for analyzing brain tissue components based on magnetic resonance image. The system includes a memory and a processor. The memory stores instructions. The processor accesses and executes the instructions to perform the following: extracting maps of tissue from a brain magnetic resonance imaging (MRI) corresponding to normal subjects; averaging the maps of tissue according to a number of the normal subjects to generate reference maps that correspond to different tissues; receiving a brain MRI sample having a targeted region; and analyzing the brain MRI sample based on the reference maps and the targeted region to generate an analysis result, in which the analysis result indicates a ratio of tissues in the targeted region of the brain MRI sample.
1. A system for analyzing brain tissue components based on magnetic resonance image, comprising:
a memory, configured to store at least one instruction; and
a processor, communicatively coupled to the memory, wherein the processor is configured to access and execute the at least one instruction to:
extract a plurality of tissue maps from a plurality of brain magnetic resonance imaging (MRI) corresponding to a plurality of normal subjects, wherein the plurality of tissue maps are a plurality of probability maps indicating a plurality of probability distributions of a plurality of tissues in a brain, wherein the plurality of normal subjects are people different from a patient;
sum up the plurality of probability distributions to divide by a number of the plurality of normal subjects, in order to generate a plurality of reference tissue maps, wherein the plurality of reference tissue maps are corresponding to different tissues;
receive a brain MRI sample having a target area marked thereon; and
analyze the brain MRI sample based on the plurality of reference tissue maps and the target area, in order to generate an analysis result, wherein the analysis result indicates a ratio of tissues corresponding to the target area of the brain MRI sample.
2. The system for analyzing brain tissue components based on magnetic resonance image of claim 1 , wherein the plurality of tissue maps comprise a gray matter map, a white matter map and a cerebrospinal fluid map.
3. The system for analyzing brain tissue components based on magnetic resonance image of claim 2 , wherein the ratio of tissues is a gray matter/white matter/cerebrospinal fluid ratio corresponding to the target area of the brain MRI sample.
4. The system for analyzing brain tissue components based on magnetic resonance image of claim 1 , wherein the processor accesses and executes the at least one instruction to average the plurality of tissue maps comprising:
normalize the plurality of tissue maps based on a diffeomorphic anatomical registration through exponential lie algebra, in order to map the plurality of tissue maps to a standard brain space; and
average a plurality of first maps, which are corresponding to a first tissue from the plurality of tissues, from the plurality of normalized tissue maps, in order to generate a first reference tissue map of the plurality of reference tissue maps, wherein the first reference tissue map corresponds to the first tissue.
5. The system for analyzing brain tissue components based on magnetic resonance image of claim 1 , wherein the target area is corresponding to a tumor position.
6. A method for analyzing brain tissue components based on magnetic resonance image, wherein the method is executed by a processor and comprises:
extracting a plurality of tissue maps from a plurality of brain magnetic resonance imaging (MRI) corresponding to a plurality of normal subjects, wherein the plurality of tissue maps are a plurality of probability maps indicating a plurality of probability distributions of a plurality of tissues in a brain, wherein the plurality of normal subjects are people different from a patient;
summing up the plurality of probability distributions to divide by a number of the plurality of normal subjects, in order to generate a plurality of reference tissue maps, wherein the plurality of reference tissue maps are corresponding to different tissues;
receiving a brain MRI sample having a target area marked thereon; and
analyzing the brain MRI sample based on the plurality of reference tissue maps and the target area, in order to generate an analysis result, wherein the analysis result indicates a ratio of tissues corresponding to the target area of the brain MRI sample.
7. The method for analyzing brain tissue components based on magnetic resonance image of claim 6 , wherein the plurality of tissue maps comprise a gray matter map, a white matter map and a cerebrospinal fluid map.
8. The method for analyzing brain tissue components based on magnetic resonance image of claim 7 , wherein the ratio of tissues is a gray matter/white matter/cerebrospinal fluid ratio corresponding to the target area of the brain MRI sample.
9. The method for analyzing brain tissue components based on magnetic resonance image of claim 6 , wherein averaging the plurality of tissue maps comprising:
normalizing the plurality of tissue maps based on a diffeomorphic anatomical registration through exponential lie algebra, in order to map the plurality of tissue maps to a standard brain space; and
averaging a plurality of first maps, which are corresponding to a first tissue from the plurality of tissues, from the plurality of normalized tissue maps, in order to generate a first reference tissue map of the plurality of reference tissue maps, wherein the first reference tissue map corresponds to the first tissue.
10. The method for analyzing brain tissue components based on magnetic resonance image of claim 6 , wherein the target area is corresponding to a tumor position.