An automated determination of Blumensaat line using fuzzy system based on physician experience from femur CT image

Blumensaat line is one of the most commonly used direct methods for the assessment of femoral diagnosis and therapy. Currently, the Blumensaat line is determined manually. Therefore, diversity of the determination happens due to the subjective judgment error. To reduce the diversity, we propose an automated determination of Blumensaat Une by using fuzzy logic based on physician knowledge from femur multi-detector row computerized tomography (MDCT) image. The experiment employed six different knees. The six femurs were evaluated by the manual and proposed method. In the results, the length of Blumensaat line was 24.12 ± 3.23 mm (manual) and 23.90 ± 2.41 mm (automated). The angle between Blumensaat line and bone axis was 27.80 ± 6.08 degrees (manual) and 30.68 ± 5.76 degrees (automated). There was no significant difference between the manual and proposed method. We concluded that the proposed method has enough accuracy as same as expert.