Deep-learning based 3D reconstruction of lower limb bones from biplanar radiographs for preoperative osteotomy planning.

PURPOSE: Three-dimensional (3D) preoperative planning has become the gold standard for orthopedic surgeries, primarily relying on CT-reconstructed 3D models. However, in contrast to standing radiographs, a CT scan is not part of the standard protocol but is usually acquired for preoperative planning purposes only. Additionally, it is costly, exposes the patients to high doses of radiation and is acquired in a non-weight-bearing position. METHODS: In this study, we develop a deep-learning based pipeline to facilitate 3D preoperative planning for high tibial osteotomies, based on 3D models reconstructed from low-dose biplanar standing EOS radiographs. Using digitally reconstructed radiographs, we train networks to localize the clinically required landmarks, separate the two legs in the sagittal radiograph and finally reconstruct the 3D bone model. Finally, we evaluate the accuracy of the reconstructed 3D models for the particular application case of preoperative planning, with the aim of eliminating the need for a CT scan in specific cases, such as high tibial osteotomies. RESULTS: The mean Dice coefficients for the tibial reconstructions were 0.92 and 0.89 for the right and left tibia, respectively. The reconstructed models were successfully used for clinical-grade preoperative planning in a real patient series of 52 cases. The mean differences to ground truth values for mechanical axis and tibial slope were 0.52° and 4.33°, respectively. CONCLUSIONS: We contribute a novel framework for the 2D-3D reconstruction of bone models from biplanar standing EOS radiographs and successfully use them in automated clinical-grade preoperative planning of high tibial osteotomies. However, achieving precise reconstruction and automated measurement of tibial slope remains a significant challenge.

Subject-Specific Alignment and Mass Distribution in Musculoskeletal Models of the Lumbar Spine.

Musculoskeletal modeling is a well-established method in spine biomechanics and generally employed for investigations concerning both the healthy and the pathological spine. It commonly involves inverse kinematics and optimization of muscle activity and provides detailed insight into joint loading. The aim of the present work was to develop and validate a procedure for the automatized generation of semi-subject-specific multi-rigid body models with an articulated lumbar spine. Individualization of the models was achieved with a novel approach incorporating information from annotated EOS images. The size and alignment of bony structures, as well as specific body weight distribution along the spine segments, were accurately reproduced in the 3D models. To ensure the pipeline's robustness, models based on 145 EOS images of subjects with various weight distributions and spinopelvic parameters were generated. For validation, we performed kinematics-dependent and segment-dependent comparisons of the average joint loads obtained for our cohort with the outcome of various published in vivo and in situ studies. Overall, our results agreed well with literature data. The here described method is a promising tool for studying a variety of clinical questions, ranging from the evaluation of the effects of alignment variation on joint loading to the assessment of possible pathomechanisms involved in adjacent segment disease.