Spline alignment of amygdala and hippocampus

1 Aligning medial temporal lobe anatomy with splines

The alignment will be performed in two stages. First, a 3×3 linear transformation matrix will be estimated and applied to the data. Then a spline based transformation will be estimated and applied to the linearly transformed data.

1.1 The data

There are 8 pieces of data, landmarks and triangulated surfaces for atlas and target hippocampus and amygdala.

1. Atlas landmarks: amygdala_01_landmarks.txt, hippocampus_01_landmarks.txt
2. Atlas surfaces: amygdala_01_surface.byu, hippocampus_01_surface.byu
3. Target landmarks: amygdala_05_landmarks.txt,hippocampus_05_landmarks.txt
4. Target surfaces: amygdala_05_surface.byu,hippocampus_05_surface.byu

1.2 Loading landmarks

Use read_txt_landmarks.m with the filename as an input to read landmarks into a matlab array. The size of the array will be N by 3, for N = 20 for amygdala, and N = 38 for hippocampus. There are a total of 58 landmarks for each the pair of structures.

1.3 Loading surfaces

Use read_byu_surface.m with the filename as an input to read vertices and faces matlab arrays. This will output a vertex array, containing a one 3D vertex per row. In addition, it will contain a face array, containing 3 integers per row. Each of these integers corresponds to a row of the vertex array, and describes how they should be connected to form triangles.

1.4 Spline and linearly transformed landmarks and surfaces

Results are in Generated Images folder with code in hip_amyg.m