TU Wien:Visualisierung medizinischer Daten 1 VU (Sramek, Dimitrov)/Prüfung 2018-06-29

10 pages with about 13 questions overall.

I already forgot some of them, here are the ones I've remembered:

1 Acquisition

1.1 MRI

What is it, how does it work?

Four different brain/skull MIR images were given (T1, T2, T1 + Contrast, Diffusion) - Describe what you see

1.2 DICOM

What is it, it's structure.

Describe it's relation with medical imaging

1.3 3d view of head with view into brain

A 3d rendering of human head was given

Explain how you would create such a rendering (which modalities to use,..)

2 Reconstruction

2.1 Explain aliasing from a signal theoretic point of view

2.2 What is pre-aliasing, post-aliasing. How can post-aliasing be avoided?

2.3 Two images with artifacts

Identify the modalities and the kind of artifacts (streak-noise, inhomogeneity, ...) you see in each image

3 Volume Rendering

3.1 Identify Rendering Methods

Two images were given.

Describe which rendering method was used (should have been Direct Volume Rendering and Surface Rendering ) and explain your reasoning.

Describe each method and use cases.

3.2 Compositing methods graphic was given (first, additive, maximum, cut, accumulate)

Three skull images (one in color, two in grayscale)

Describe which compositing method was used for each image and explain your reasoning

4 Marching Cubes

4.1 A 2d grid was given - Draw the contour that encloses an isovalue of 5 using the marching squares algorithm

4.2 Show the ambiguous case for marching cubes - what is the problem?

5 Transfer functions

5.1 What are transfer functions. Why are multi-dimensional transfer functions needed

5.2 Graphic with different colored circles was given (F0, F1, ... - like in the slides)

Separate circle F4.

Which transfer function to use and why?

Draw the filter graph (f'- f graph)