Computerized reconstruction of fragmentary skeletal remains.

This research presents a new software, "Fragmento", for accurate analyses of fragmentary human skeletal remains and facilitation of three-dimensional (3D) fragmentary matching and full bone reconstruction. The framework utilizes the power of statistical bone atlases to create 3D templates for bone matching and to interpolate missing anatomy for full bone reconstruction. Developed tool has enhanced features allowing the user to visualize, review and scale all scanned skeletal remains within a 3D statistical template, merging accepted registered elements to provide a fully reconstructed bone. A three stage validation was performed on Fragmento: Stages I and II used simulated fragmentary data which was compared to full bones with an error less than 3mm; Stage III compared output from geographic information system (GIS) software with comparable results. This validation process demonstrates the robustness and utility of Fragmento as tool for 3D fragmentary bone matching and full bone reconstruction.

A three-dimensional analysis of bilateral directional asymmetry in the human clavicle.

This study presents a novel three-dimensional analysis using statistical atlases and automated measurements to assess diaphyseal morphology of the clavicle and its relationship to muscle asymmetry. A sample of 505 individuals (285 males, 220 females) from the William McCormick Clavicle Collection was CT scanned, segmented, and added to a statistical bone atlas that captures correspondence between homologous points on the bone surfaces. Muscle attachment sites were localized on the atlas and then propagated across the entire population. Cross-sectional contours were extracted at 5% increments along the entire bone, as well as at muscle attachment sites and the clavicle waist; maximum and minimum dimensions of each cross-sectional contour were calculated. In addition, the entire three-dimensional surface was examined for asymmetry by analyzing the magnitude and directional differences between homologous points across all bone surfaces in the dataset. The results confirm the existing studies on clavicle asymmetry, namely that the left clavicle is longer than the right, but the right is more robust than the left. However, the patterns of asymmetry are sexually dimorphic. Males are significantly asymmetric in all dimensions and at muscle and ligament attachment sites (P < 0.05), whereas female asymmetry is more variable. We hypothesize that this is related to absolute and relative differences in male muscle strength compared to females. However, an area with no muscle attachments on the posterior midshaft was significantly asymmetric in both sexes. We suggest that this is a curvature difference caused by opposing muscle actions at the medial and lateral ends of the bone.

Three-dimensional morphology of the knee reveals ethnic differences.

BACKGROUND: Studies have demonstrated sex differences in femoral shape and quadriceps angle raising a question of whether implant design should be sex-specific. Much of this research has addressed shape differences within the Caucasian population and little is known about differences among ethnic groups. QUESTIONS/PURPOSES: We therefore asked: Do shape differences in the distal femur and proximal tibia exist among different ethnic groups and between the sexes in each ethnic population? And if ethnic differences exist, do they have a clinical impact on current TKA design? SUBJECTS AND METHODS: We analyzed 1000 normal adult knees (80 African American, 80 East Asian, and 860 Caucasian). Three-dimensional surface models were created for each bone and added to three-dimensional statistical bone atlases. Statistical shape analysis was conducted with a process combining principal components and multiple discriminate analyses. Eleven femoral and nine tibial measurements were calculated. RESULTS: We found differences in mean measurements between the sexes and ethnicities. Males had larger knees, with a mean 5-mm-larger anteroposterior dimension than females in all ethnicities. African American females had a 7.4-mm-deeper patellar groove, 2.3-mm-smaller tibial mediolateral dimension, and 2.5-mm-larger tibial anteroposterior dimension than Caucasian females. African American males had a 4.3-mm-larger femoral anteroposterior dimension, 10.1-mm-larger tibial mediolateral dimension, and 6-mm-larger tibial anteroposterior dimension than Asian males. CONCLUSIONS: We identified differences in three-dimensional knee morphology among Caucasian, African American, and East Asian populations. Clinical studies will be required to determine whether these differences are important for implant design.

Automatic and manual methodology for three-dimensional measurements of distal femoral gender differences and femoral component placement.

Our study aimed to introduce an automatic three-dimensional method for measuring the distal femur and identifying potential gender differences and the effects on femoral component placement in total knee arthroplasty. Three hundred forty-two femora were scanned with computed tomography. Automatic and manual bone resection and component placement were compared using a virtual resection tool. For standard component use, 77.3% of the femora were male and 23.1% were female. For gender-specific component use, 91% were female and 7.3% were male. Surgeon errors in both component translation and rotation existed but were minimal. From these results, gender alone did not dictate component use in primary total knee arthroplasty. The restoration of femoral condylar profile in 3 dimensions can be obtained by accurately measuring patient distal femoral anatomy and the appropriate femoral component design selection. Additional bone cuts, soft-tissue maneuvers, and adverse outcomes in fitting the patient to the femoral component may be avoided.