Methods and apparatus for estimating clinical measures
In a magnetic resonance imaging display system, the brain parenchymal fraction, a clinical measure for brain atrophy, is found by selection of white matter, grey matter, and/or cerebrospinal fluid based on quantitative magnetic resonance properties.
1. A computerized method of estimating an intracranial volume (ICV) of a head, the method comprising:
obtaining images of quantitative magnetic resonance (MR) properties of the head,
based on the images, selecting different tissues of the head,
generating a mask including all pixels corresponding to types of the selected tissues,
based on the generated mask, forming a contiguous volume, and
estimating the ICV using the contiguous volume;
wherein the quantitative MR properties correspond to at least one of R1 and R2 relaxation rate or proton density, or relaxation time, where T1=1/R1 and T2=1/R2.
2. The method of claim 1 , wherein the selected different tissues comprises white matter (WM), and grey matter (GM) and cerebrospinal fluid (CSF).
3. The method of claim 2 , wherein the selected different tissue types are only WM, GM and CSF.
4. The method of claim 1 , wherein a region-growing algorithm selects the largest contiguous ICV volume in a dataset to exclude false positives outside the ICV and to include false negatives inside the ICV.
5. The method of claim 4 , wherein an edge of the selected ICV is refined by using a threshold of proton density (PD)=50% for inclusion of pixels at the edge of the ICV.
6. The method of claim 1 , wherein cerebrospinal fluid (CSF) is selected within the ICV based on quantitative MR properties, to obtain a brain parenchymal volume (BPV) as ICV−CSF.
7. The method of claim 6 , wherein a brain parenchymal fraction (BPF) is determined as BPV/ICV.
8. The method of claim 6 , wherein white matter (WM) is selected within the BPV based on the quantitative MR properties, to obtain a white matter brain fraction (WMBF) as WM/BPV.
9. The method of claim 6 , wherein grey matter (GM) is selected within the BPV based on the quantitative MR properties, to obtain a grey matter brain fraction (GMBF) as GM/BPV.
10. The method of claim 1 , wherein white matter (WM) is selected within the ICV based on the quantitative MR properties, to obtain a white matter fraction (WMF) as WM/ICV.
11. The method of claim 1 , wherein grey matter (GM) is selected within the ICV based on the quantitative MR properties, to obtain a grey matter fraction (GMF) as GM/ICV.
12. The method of claim 1 , wherein cerebrospinal fluid (CSF) is selected within the ICV based on quantitative MR properties, to obtain a cerebrospinal fluid fraction (CSFF) as CSF/ICV.
13. The method of claim 1 , wherein white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF) are selected within the ICV based on the quantitative MR properties, to obtain a remaining non-recognize fraction (NF) as (ICV−WM−GM−CSF)/ICV.
14. The method of claim 1 , wherein white matter (WM) and grey matter (GM) are selected within the ICV based on the quantitative MR properties, to obtain a white/grey ratio (WGR) as WM/GM.
15. A device for computerized estimation of an intracranial volume (ICV) of a head, the device comprising:
an electronic processor and memory configured for obtaining images of quantitative magnetic resonance (MR) properties of the head, for selecting different tissues of the head based on the images, for generating a mask including all pixels corresponding to the selected tissue types, for forming a contiguous volume based on the generated mask, and for estimating the ICV using the contiguous volume;
wherein the quantitative MR properties correspond to at least one of R1 and R2 relaxation rate or proton density, or relaxation time, where T1=1/R1 and T2=1/R2.
16. The device of claim 15 , wherein the selected different tissues include white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF).
17. The device of claim 16 , wherein the selected different tissue types are only WM, GM, and CSF.
18. The device of claim 16 , wherein the electronic processor and memory is further configured for selecting WM, GM and CSF within the ICV based on the quantitative MR properties, and for obtaining a remaining non-recognized fraction (NF) as (ICV−WM−GM−CSF)/ICV.
19. The device of claim 15 , wherein the electronic processor and memory is further configured for applying a region-growing algorithm configured to select a largest contiguous ICV volume in a dataset to exclude false positives outside the ICV and to include false negatives inside the ICV.
20. The device of claim 15 , wherein the electronic processor and memory is further configured for refining an edge of the selected ICV by using a threshold of proton density (PD)=50% for inclusion of pixels at the edge of the ICV.
21. The device of claim 15 , wherein the electronic processor and memory is further configured for selecting cerebrospinal fluid (CSF) within the ICV based on quantitative MR properties, and for obtaining a brain parenchymal volume (BPV) as ICV−CSF.
22. The device of claim 21 , wherein the device is configured to determine a brain parenchymal fraction (BPF) as BPV/ICV.
23. The device of claim 22 , wherein the electronic processor and memory is further configured for selecting white matter (WM) within the BPV based on the quantitative MR properties, and for obtaining a white matter brain fraction (WMBF) as WM/BPV.
24. The device of claim 22 , wherein the electronic processor and memory is further configured for selecting grey matter (GM) within the BPV based on the quantitative MR properties, and for obtaining a grey matter brain fraction (GMBF) as GM/BPV.
25. The device of claim 15 , wherein the electronic processor and memory is further configured for selecting white matter (WM) within the ICV based on the quantitative MR properties, and for obtaining a white matter fraction (WMF) as WM/ICV.
26. The device of claim 15 , wherein the electronic processor and memory is further configured for selecting grey matter (GM) within the ICV based on the quantitative MR properties, and for obtaining a grey matter fraction (GMF) as GM/ICV.
27. The device of claim 15 , wherein the electronic processor and memory is further configured for selecting cerebrospinal fluid (CSF) within the ICV based on quantitative MR properties, and for obtaining a cerebrospinal fluid fraction (CSFF) as CSF/ICV.
28. The device of claim 15 , wherein the electronic processor and memory is further configured for selecting white matter (WM) and grey matter (GM) within the ICV based on the quantitative MR properties, and for obtaining a white/grey ratio (WGR) as WM/GM.
29. A non-transitory digital storage medium having stored thereon computer program instructions that when executed by a computer causes the computer to perform the steps of:
obtaining images of quantitative magnetic resonance (MR) properties of a head,
based on the images, selecting different tissues of the head,
generating a mask including all pixels corresponding to types of the selected tissue,
based on the generated mask, forming a contiguous volume, and
estimating an intracranial volume (ICV) of the head using the contiguous volume;
wherein the quantitative MR properties correspond to at least one of R1 and R2 relaxation rate or proton density, or relaxation time, where T1=1/R1 and T2=1/R2.
30. A computerized method of estimating an intracranial volume (ICV) of a head, the method comprising:
obtaining images of quantitative magnetic resonance (MR) properties of the head;
based on the images, selecting different tissues of the head;
generating a mask including all pixels corresponding to types of the selected tissues;
based on the generated mask, forming a contiguous volume; and
estimating the ICV using the contiguous volume;
wherein a region-growing algorithm selects the largest contiguous ICV volume in a dataset to exclude false positives outside the ICV and to include false negatives inside the ICV; and an edge of the selected ICV is refined by using a threshold of proton density (PD)=50% for inclusion of pixels at the edge of the ICV.
31. A device for computerized estimation of an intracranial volume (ICV) of a head, the device comprising:
an electronic processor and memory configured for obtaining images of quantitative magnetic resonance (MR) properties of the head, for selecting different tissues of the head based on the images, for generating a mask including all pixels corresponding to the selected tissue types, for forming a contiguous volume based on the generated mask, and for estimating the ICV using the contiguous volume;
wherein the electronic processor and memory is further configured for refining an edge of the selected ICV by using a threshold of proton density (PD)=50% for inclusion of pixels at the edge of the ICV.