Recent changes to Sync MRI data

Sync MRI data modified by Ilan Tal

Ilan Tal — Thu, 14 Nov 2024 08:23:08 -0000

--- v35
+++ v36
@@ -1,4 +1,4 @@
-Most data has the PET and CT acquired on the same machine in the same study. The patient has not moved between scans, with the result that the data is synchronized. When MRI data is used in place of the CT data, the MRI data comes from a different machine at a different time. Thus the patient position is not likely to be synchronized between the 2 data-sets. (See [demo](https://youtu.be/gqtaQgzGw8M))
+Most data has the PET and CT acquired on the same machine in the same study. The patient has not moved between scans, with the result that the data is synchronized. When MRI data is used in place of the CT data, the MRI data comes from a different machine at a different time. Thus the patient position is not likely to be synchronized between the 2 data-sets. (See [demo](https://youtu.be/Majs-P9_MVU)

 [[img src=mismatchData.png]]

@@ -12,7 +12,7 @@

 The sync starts with the axial view with the Z direction matching the appropriate slices. The **Reset** button zeros out X,Y,Z in all (axial, sagittal. coronal) views. The log report gives you a reasonable starting point. In this case you would type in the Z value as 131 (followed by Enter). In the end I chose a value of 129 which was quite close to the suggested value.

-You probably want to align the MRI data in all orientations (axial first, then coronal and sagittal). There are 3 different sets of  **z**, **x** and **y** values, one for each orientation.  (Just to remind you, there is a [demo](https://youtu.be/gqtaQgzGw8M))
+You probably want to align the MRI data in all orientations (axial first, then coronal and sagittal). There are 3 different sets of  **z**, **x** and **y** values, one for each orientation.  (Just to remind you, there is a [demo](https://youtu.be/Majs-P9_MVU))

 The best resolution data in MRI scans are usually in the sagittal view. You may use whatever you have available.
 Return to [Help](Pet Ct Viewer Help)

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 15:54:46 -0000

--- v34
+++ v35
@@ -10,11 +10,9 @@

 [[img src=log.png]]

-The above study is clearly poorly aligned. Pressing the **Reset** button will make a reasonable starting point for the **z** value. In the above case the required z shift was 74. After that there is the question if the center of rotation values are correct or not. For CT the difference in the centers of rotation is likely zero, so checking [[img src=checkbox.png]]** ignore XY** would make no difference. In the above case the large Y shift is primarily due to an erroneous difference in the centers of rotation. Choosing [[img src=checkbox.png]]** ignore XY** will make an obvious improvement. In short, if it helps, use it.
+The sync starts with the axial view with the Z direction matching the appropriate slices. The **Reset** button zeros out X,Y,Z in all (axial, sagittal. coronal) views. The log report gives you a reasonable starting point. In this case you would type in the Z value as 131 (followed by Enter). In the end I chose a value of 129 which was quite close to the suggested value.

-The **Reset** button has been extended to have 2 different states. The reason is that lung CT has been added, where the lung CT (which is read as an MRI study) is indeed correlated to the original CT. It will probably come up correctly with the proper shifts, but you are free to try to improve it. With the first press of the Reset it will shift guessing that it might be uncorrelated data, and the second press of Reset guesses that it is correlated data. You can toggle back and forth to whatever look better, and make any changes from the closest approximation.
-
-You probably want to align the MRI data in all orientations (usually axial first, then coronal and sagittal). There are 3 different sets of  **z**, **x** and **y** values, one for each orientation.  (Just to remind you, there is a [demo](https://youtu.be/gqtaQgzGw8M))
+You probably want to align the MRI data in all orientations (axial first, then coronal and sagittal). There are 3 different sets of  **z**, **x** and **y** values, one for each orientation.  (Just to remind you, there is a [demo](https://youtu.be/gqtaQgzGw8M))

 The best resolution data in MRI scans are usually in the sagittal view. You may use whatever you have available.
 Return to [Help](Pet Ct Viewer Help)

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 15:31:22 -0000

--- v33
+++ v34
@@ -8,7 +8,7 @@

 Many times CT high resolution lung scans are used in place of MRI studies. In almost all cases they will be aligned and you won't need to do anything. With MRI scans the situation is different. MRI and PET/CT are different machines and there is no demand to cross calibrate them. Likewise, it is common that there can be months between the scans, so the patient is not placed in the same relative position. Thus is it common for MRI scans to need a manual alignment. If the registration error is beyond a reasonable limit you will see a log report

-[[img_src=log.png]]
+[[img src=log.png]]

 The above study is clearly poorly aligned. Pressing the **Reset** button will make a reasonable starting point for the **z** value. In the above case the required z shift was 74. After that there is the question if the center of rotation values are correct or not. For CT the difference in the centers of rotation is likely zero, so checking [[img src=checkbox.png]]** ignore XY** would make no difference. In the above case the large Y shift is primarily due to an erroneous difference in the centers of rotation. Choosing [[img src=checkbox.png]]** ignore XY** will make an obvious improvement. In short, if it helps, use it.

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 15:28:37 -0000

--- v32
+++ v33
@@ -8,6 +8,8 @@

 Many times CT high resolution lung scans are used in place of MRI studies. In almost all cases they will be aligned and you won't need to do anything. With MRI scans the situation is different. MRI and PET/CT are different machines and there is no demand to cross calibrate them. Likewise, it is common that there can be months between the scans, so the patient is not placed in the same relative position. Thus is it common for MRI scans to need a manual alignment. If the registration error is beyond a reasonable limit you will see a log report

+[[img_src=log.png]]
+
 The above study is clearly poorly aligned. Pressing the **Reset** button will make a reasonable starting point for the **z** value. In the above case the required z shift was 74. After that there is the question if the center of rotation values are correct or not. For CT the difference in the centers of rotation is likely zero, so checking [[img src=checkbox.png]]** ignore XY** would make no difference. In the above case the large Y shift is primarily due to an erroneous difference in the centers of rotation. Choosing [[img src=checkbox.png]]** ignore XY** will make an obvious improvement. In short, if it helps, use it.

 The **Reset** button has been extended to have 2 different states. The reason is that lung CT has been added, where the lung CT (which is read as an MRI study) is indeed correlated to the original CT. It will probably come up correctly with the proper shifts, but you are free to try to improve it. With the first press of the Reset it will shift guessing that it might be uncorrelated data, and the second press of Reset guesses that it is correlated data. You can toggle back and forth to whatever look better, and make any changes from the closest approximation.

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 13:39:26 -0000

--- v31
+++ v32
@@ -5,6 +5,8 @@
 [[img src=syncMRI.png]]

 This is a manual process where you display the MRI and PET images and the fusion between them. Since this is a manual process you may wish to save the results. There are up to 10 locations where the data can be saved, so the spinner is set at 1 to 10 and is usually labeled by the patient name or something else which is easy to connect to the particular study.
+
+Many times CT high resolution lung scans are used in place of MRI studies. In almost all cases they will be aligned and you won't need to do anything. With MRI scans the situation is different. MRI and PET/CT are different machines and there is no demand to cross calibrate them. Likewise, it is common that there can be months between the scans, so the patient is not placed in the same relative position. Thus is it common for MRI scans to need a manual alignment. If the registration error is beyond a reasonable limit you will see a log report

 The above study is clearly poorly aligned. Pressing the **Reset** button will make a reasonable starting point for the **z** value. In the above case the required z shift was 74. After that there is the question if the center of rotation values are correct or not. For CT the difference in the centers of rotation is likely zero, so checking [[img src=checkbox.png]]** ignore XY** would make no difference. In the above case the large Y shift is primarily due to an erroneous difference in the centers of rotation. Choosing [[img src=checkbox.png]]** ignore XY** will make an obvious improvement. In short, if it helps, use it.

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 13:23:40 -0000

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 13:22:47 -0000

Sync MRI data modified by Ilan Tal

Ilan Tal — Wed, 13 Nov 2024 13:21:46 -0000

Sync MRI data modified by Ilan Tal

Ilan Tal — Thu, 21 Dec 2023 16:17:12 -0000

--- v27
+++ v28
@@ -10,8 +10,7 @@

 The **Reset** button has been extended to have 2 different states. The reason is that lung CT has been added, where the lung CT (which is read as an MRI study) is indeed correlated to the original CT. It will probably come up correctly with the proper shifts, but you are free to try to improve it. With the first press of the Reset it will shift guessing that it might be uncorrelated data, and the second press of Reset guesses that it is correlated data. You can toggle back and forth to whatever look better, and make any changes from the closest approximation.

-Now you may want to switch to the coronal slice and adjust the **z** and **x** values for best overlap in the fusion view. The **Reset** set the **z** value to 74 but it may need some additional adjustment. After the **z** value has been correctly set, return to the axial view and set **x** and **y** for the best overlap. Finally we go to sagittal view. With a bit of luck the adjustments in the coronal and axial views will set the sagittal correctly, but there can be a residual shift. The **sag** control will shift the sagittal image left and right for the best overlap.
+You probably want to align the MRI data in all orientations (usually axial first, then coronal and sagittal). There are 3 different sets of  **z**, **x** and **y** values, one for each orientation.  (Just to remind you, there is a [demo](https://youtu.be/gqtaQgzGw8M))

-The best resolution data in MRI scans are usually in the sagittal view. The Pet Ct Viewer uses axial data, which while it exists in MRI data, is usually of lower resolution. There is a free program which requires registration called [ORS lite](http://www.theobjects.com) which can convert the high resolution sagittal files to axial files.
-
+The best resolution data in MRI scans are usually in the sagittal view. You may use whatever you have available.
 Return to [Help](Pet Ct Viewer Help)

Sync MRI data modified by Ilan Tal

Ilan Tal — Thu, 21 Dec 2023 15:45:58 -0000

--- v26
+++ v27
@@ -1,4 +1,4 @@
-Most data has the PET and CT acquired on the same machine in the same study. The patient has not moved between scans, with the result that the data is synchronized. When MRI data is used in place of the CT data, the MRI data comes from a different machine at a different time. Thus the patient position is not likely to be synchronized between the 2 data-sets. (See [demo](https://youtu.be/3H7bLDivE88))
+Most data has the PET and CT acquired on the same machine in the same study. The patient has not moved between scans, with the result that the data is synchronized. When MRI data is used in place of the CT data, the MRI data comes from a different machine at a different time. Thus the patient position is not likely to be synchronized between the 2 data-sets. (See [demo](https://youtu.be/gqtaQgzGw8M))

 [[img src=mismatchData.png]]