Abstract
Introduction
Traditionally, conventional radiographs of the hip are used to assist surgeons during the preoperative planning process, and these processes generally involve two-dimensional X-ray images with implant templates. Unfortunately, while this technique has been used for many years, it is very manual and can lead to inaccurate fits, such as “good” fits in the frontal view but misalignment in the sagittal view. In order to overcome such shortcomings, it is necessary to fully describe the morphology of the femur in three dimensions, therefore allowing the surgeon to successfully view and fit the components from all possible angles.
Objective
The objective of this study was to efficiently describe the morphology of the proximal femur based on existing anatomical landmarks for use in surgical planning and/or forward solution modeling.
Methods
Seven parameters are needed to fully define femoral morphology: head diameter, head center, neck shaft axis, femoral canal, proximal shaft axis, offset, and neck shaft angle. A previous algorithm has been developed in-house to automatically locate anatomical landmarks of patient specific bone models. Once the bone model has been aligned and scaled based on these landmarks, the femoral head diameter and center are calculated by iteratively fitting a sphere to the corresponding femoral head point cloud. An iterative cylindrical fitting algorithm is used to describe the neck shaft axis. The femoral canal is determined using three steps: 1) the femur is sliced at 10mm increments below the lesser trochanter, 2) the femoral canal boundary is determined at each slice, and 3) the largest circle is fit within each slice's canal boundary. The proximal shaft axis is described by fitting a line to the canal circle center locations. Offset is defined as the distance from the head center to the proximal shaft axis. Finally, the neck shaft angle is the angle between the neck shaft axis and the proximal shaft axis.
Results
The goal pertaining to femoral component morphology is to provide meaningful information that can be used to determine how the femoral stem fits within the canal. Regardless of differences in bone sizes and geometries, the algorithm has proven to be successful in describing the femoral morphology of a patient-specific bone model.
Discussion
These results lay the groundwork for an automatic stem fitting algorithm, which is described in a subsequent abstract. The morphology knowledge of the femoral head, femoral neck, femoral canal, and various axes can be coupled with known THA component parameters (such as offset, neck length, neck shaft angle, etc.) to allow our algorithms to predict the “best selection” and “best fit” for the femoral stem. This can also be applied to the acetabulum and can then be used as a surgical planning tool as well as a parameter when modeling postoperative predictions.
For any figures or tables, please contact authors directly.