System and method for segmentation and visualization of medical image data

1. A computing device comprising:

a processor;

a display coupled to the processor;

a user interface coupled to the processor for entering data into the computing device; and

a memory coupled to the processor, the memory storing program instructions that when executed by the processor, causes the processor to:

load a plurality of two-dimensional MRI image data, the two-dimensional MRI image data taken along multiple planes to create a stack of two-dimension MRI images;

load a three-dimensional anatomical model associated with an anatomical area of the two-dimensional MRI image data;

segment the three-dimensional model may into multiple components, at least one of the components being modified to accurately represent the stacked two-dimensional images to form a modified three-dimensional model; and

adjust at least one property on the modified three-dimensional model to form a modified anatomical model to match the medical data image.

2. The computing device of claim 1 , wherein the memory storing program instructions executed by the processor, causes the processor to form one of an automated video or animation creation of the modified anatomical model.

3. The computing device of claim 1 , wherein the memory storing program instructions executed by the processor, causes the processor to generate, using machine learning trained by previous segmented three-dimensional models, an MRI report.

4. The computing device of claim 1 , wherein the memory storing program instructions executed by the processor, causes the processor to:

segment each component;

detect and grade abnormalities in each component;

compile a summary of the detected and graded abnormalities.

5. The computing device of claim 4 , wherein the memory storing program instructions executed by the processor, causes the processor to segment each component using a semantic segmentation network trained using previous segmented three-dimensional models.

6. The computing device of claim 4 , wherein the memory storing program instructions executed by the processor, causes the processor to detect and grade abnormalities in each component using two separate VGG convolutional networks 85 trained with semi-supervised methods and radiological findings labeled using Natural Language Processing (NLP).

7. The computing device of claim 1 , wherein the memory storing program instructions executed by the processor, causes the processor to store the modified anatomical model in the database as a new anatomical model.

8. The computing device of claim 1 , wherein the memory storing program instructions executed by the processor, causes the processor to:

divide each component into a plurality of sections, wherein each section having adjustable properties;

adjusting a desired section of a desired component to conform the modified three-dimensional model to match the medical data image.

9. The computing device of claim 8 , wherein the adjustable properties comprises: height, radius, thickness, length, position, and scale.

10. A computing device comprising:

a processor;

a display coupled to the processor;

a user interface coupled to the processor for entering data into the computing device; and

a memory coupled to the processor, the memory storing program instructions that when executed by the processor, causes the processor to:

load a plurality of two-dimensional MRI image data, the two-dimensional MRI image data taken along multiple planes to create a stack of two-dimension MRI images;

load a three-dimensional anatomical model associated with an anatomical area of the two-dimensional MRI image data;

segment the three-dimensional model may into multiple components, at least one of the components being modified to accurately represent the stacked two-dimensional images to form a modified three-dimensional model;

adjust at least one property on the modified three-dimensional model to form a modified anatomical model to match the medical data image;

detect if areas on the modified three-dimensional anatomical model exceeds predefined ranges indicating potential injured areas; and

form one of an automated video or animation creation of the modified anatomical model.

11. The computing device of claim 10 , wherein the memory storing program instructions executed by the processor, causes the processor to generate, using machine learning trained by previous segmented three-dimensional models, an MRI report.

12. The computing device of claim 10 , wherein the memory storing program instructions executed by the processor, causes the processor to:

segment each component;

detect and grade abnormalities in each component;

compile a summary of the detected and graded abnormalities.

13. The computing device of claim 12 , wherein the memory storing program instructions executed by the processor, causes the processor to segment each component using a semantic segmentation network trained using previous segmented three-dimensional models.

14. The computing device of claim 12 , wherein the memory storing program instructions executed by the processor, causes the processor to detect and grade abnormalities in each component using two separate VGG convolutional networks 85 trained with semi-supervised methods and radiological findings labeled using Natural Language Processing (NLP).

15. The computing device of claim 10 , wherein the memory storing program instructions executed by the processor, causes the processor to store the modified anatomical model in the database as a new anatomical model.

16. The computing device of claim 10 , wherein the memory storing program instructions executed by the processor, causes the processor to:

divide each component into a plurality of sections, wherein each section having adjustable properties;

adjusting a desired section of a desired component to conform the modified three-dimensional model to match the medical data image.

17. The computing device of claim 16 , wherein the adjustable properties comprises: height, radius, thickness, length, position, and scale.

18. A computing device comprising:

a processor;

a display coupled to the processor;

a user interface coupled to the processor for entering data into the computing device; and

a memory coupled to the processor, the memory storing program instructions that when executed by the processor, causes the processor to:

load a plurality of two-dimensional MRI image data, the two-dimensional MRI image data taken along multiple planes to create a stack of two-dimension MRI images;

load a three-dimensional anatomical model associated with an anatomical area of the two-dimensional MRI image data;

segment the three-dimensional model may into multiple components, at least one of the components being modified to accurately represent the stacked two-dimensional images to form a modified three-dimensional model;

generate, using machine learning trained by previous segmented three-dimensional models, an MRI report, wherein generate using machine learning trained by previous segmented three-dimensional models, an MRI report comprises:

segment each component;

detect and grade abnormalities in each component; and

compile a summary of the detected and graded abnormalities;

adjust at least one property on the modified three-dimensional model to form a modified anatomical model to match the medical data image;

detect if areas on the modified three-dimensional anatomical model exceeds predefined ranges indicating potential injured areas;

store the modified anatomical model in the database as a new anatomical model; and

form one of an automated video or animation creation of the modified anatomical model.

19. The computing device of claim 18 , wherein the memory storing program instructions executed by the processor, causes the processor to segment each component using a semantic segmentation network trained using previous segmented three-dimensional models.

20. The computing device of claim 18 , wherein the memory storing program instructions executed by the processor, causes the processor to detect and grade abnormalities in each component using two separate VGG convolutional networks 85 trained with semi-supervised methods and radiological findings labeled using Natural Language Processing (NLP).