Is there a tool in Slicer that allows me to achieve a similar effect as in the attached image?
This looks like a surface generated by thresholding a blood pool in a contrast CT, and then that surface is cut with a plane or ROI. The inside surface is colored yellow, and the outside surface is colored red. That is all easily achieved in Slicer. The lighting parameters (very reflective shiny surfaces with pretty dramatic lighting, ambient shadows) differ from Slicer defaults but are probably achievable with the Lights module (in Slicer Sandbox extension). So, I believe the answer is yes. The text labels are probably more easily added after taking a screenshot of the image you want than within Slicer.
@mikebind Completely agree. I was just thinking about the following part:
The inside surface is colored yellow, and the outside surface is colored red
I think the way to achieve this is not straightforward (or there is a feature I didn think of). So in the Models module you can set if you want to visualize the front-facing or back-facing side of the surface (see Visible Sides option), and the only thing I can think of to make them different is to duplicate the surface, choose different colors for each, then set this setting to front-facing in one and back-facing in the other.
The “Models” module allows you to set a “Backface Color Offset” in the Display > 3D Display > Advanced section.
The first entry there is a “Hue” offset (the other two are Saturation and Value offsets from the HSV color model). It’s not the most intuitive control, but it does allow you to independently control the appearance of the front and back sides of a surface model without having to duplicate the model. Since the display updates immediately, you also don’t really have to do anything like calculate the Hue offset you want, you can just click the up or down arrows on the control until you see the color you want on the back side of the model. Here is a clipped model with yellow on the front face and red on the back face:
Good to know!! Thanks for the tip, apparently there are still features in Slicer core I don’t know about ![]()
What type of model should be used for these operations? A segmented model? Or some other processed model (stl, label model)?
How did you “open” the model?
You can segment the blood pool using Segment Editor, export the segmentation to a model (in Data module, right-click on the segmentation and choose “Export visible segments to model”), then go to Models module to set up clipping, material properties (if you want the shiny plastic appearance), backface coloring.
Note that you don’t need to segment and export to model: instead, you can use volume rendering to show vessels and heart chambers with less work and more details. To achieve this you need to make to set the opacity of the contrast agent to 0 in the Scalar Opacity Mapping function:
Thank you for the suggestions. I apologise for the late reply, but I was struggling with post-flu pneumonia and didn’t have the energy to respond earlier.
I’ll try to explain why I’m looking for a tool that can produce 3D segmentations like the one at the beginning of this thread.
I came across an online presentation of a DICOM viewer: https://www.sciement.com/en/viewtify.html.
The way the 3D models were presented really caught my attention, so I’m trying to find a Slicer solution to this problem. The way the models are presented and displayed on the Sony Spatial Reality Display closely resembles how a surgeon actually views the surgical field.
At my hospital, we run a spatial imaging lab and use VR (Oculus) as well as computer monitors. Sciement’s solution is slightly different and I’m curious to know if we could implement something similar. While I understand that I could use Volume Rendering and adjust the parameters to achieve a similar result, this would require the surgeon to have knowledge of Slicer and the ability to manipulate the parameters, which they usually lack the time or inclination for. They want a finished model or a series of DICOM studies prepared for viewing in VR. Would such a solution work without a Sony monitor? I don’t know, but to find out, I first need to create images similar to those produced by Sciement and be able to freely crop, rotate, etc. the resulting images. Thanks for the tips on how to change the model’s display parameters to achieve a similar effect. I haven’t been able to do it yet, but I’m working on it.
That’s very interesting. IMO, radiologists should update their analysis of a study to satisfy surgical expectations. Your post simply means that traditional analysis of a study by radiologists is outdated. The radiologist should provide segmentations, models, centerlines… all objects being interactive in addition to the digital acquisition itself and to the paper report.
We are very, very far from this. Data format should be standardised to be OEM and client software independent. An ISO standard should be defined for this. The teaching of radiology during internship should be rescaled. Digital awareness and fluency as from primary school should be thought of again … …
Things may get better in a few decades: 2, 3, 4 decades… I don’t know.
For the time being, surgeons have no choice but investing thoroughly on software like Slicer or others, or continue feeling having missed the digital train, one that never stops. Because no one will be fulfilling their expectations precisely as they want or need on the spot, each study being unique, not even AI(?) IMO.
May be it’s not the reply you were expecting. It’s just my point of view as a surgeon.
This is the key question, and the answer is: of course you can! The stereo effect may have some utility in some cases, but generally does not make a big difference.
I saw a few demos of stereoscopic displays and how people react to them. People are always very impressed. However, they are impressed for the wrong reason. They think the visualization is awesome because they have not seen a good quality CT displayed using nicely configured volume rendering (not because there is also some stereo effect).




