[Freesurfer] WM surface errors

Hi, I am trying to reconstruct the surfaces of some 1.5T images that I have. For a subset of images, portions of the white matter surface in the occipital cortex are missed, even though the wm.mgz volume seems to have segmented the area correctly. The aseg also seems to have labelled this region correctly (e.g., it is not labelled erroneously as ventricle).

I have tried using control points, but this had little impact on the final solution. I've attached pics of the surfaces on the anatomical and wm volumes, along with the locations of the control points so you can see for yourself (note that I placed control points on consecutive slices through the areas). Any advice on how I can fix this problem? Thanks, Alex