We present a novel method for a femoral head segmentation and a density measurement. Our approach is based on a semi-automatical femoral head segmentation from a CT dataset based on finding an optimal path through the axial slices transformed to polar coordinates. The cost function itself is based on a combination of corticallis properties, mostly the directional behavior of the 3D gradients and their size in 2D slices, where they form typical "channels." The final volume is computed using filling and morphological algorithms and its properties are further measured. The final implementation was experimentally validated on the radiology department of the Bulovka hospital and allows a radiologist to intuitively and accurately estimate the femoral head density in approximately 1 to 3 minutes.