Background: Minimally invasive surgery is widely used for managing fractures. When using the intramedullary nailing for bone fixation, surgeons must insert interlocking screws to prevent rotation of the bone fragment; however, it is difficult to determine the exact screwing position of intramedullary nails inserted into the bone. Conventionally, the distal interlocking nail surgery is performed under intermittent X-ray shooting. Nevertheless, this freehand fluoroscopic technique is technically demanding and time-consuming. Currently, the failure rate of this surgery is more than [Formula: see text], and the location error requires to be controlled within 2[Formula: see text]mm. Purpose: To develop a deep-learning approach for locating the intramedullary nail’s holes based on 2D calibrated fluoroscopic images. Methods: The projection of the hole’s axis is deeply regressed in the first step. Then, the hole’s 3D axis is derived by computing the intersection line of two planes determined by the projection of the axis and the X-ray source, respectively. The benefit of the data-driven manner is that our method can be applied to the arbitrary shape of the hole’s contour. Besides, we extract hole’s contour as the distinctive feature, so as to reduce the space of the training data in a large scale. Results: Our approach is proved to be efficient and easy to be implemented, and it has been compared with traditional location method in phantom experiments. The location accuracy error of the traditional method is [Formula: see text][Formula: see text]mm, [Formula: see text], and the location error of this method is [Formula: see text][Formula: see text]mm, [Formula: see text]. Furthermore, the traditional method takes an average of 10[Formula: see text]min to complete the location, while our method takes only 4[Formula: see text]min. In addition, to further verify the robustness of our method, we carried out a preclinical study involving different neural networks for locating the hole’s axis. Conclusion: Whether in terms of time consumption or accuracy error, our method is significantly better than traditional method, and the efficiency has been significantly improved. Therefore, our method has great clinical value. In addition, our approach has potential advantages over the X-ray guided freehand solution in terms of radiation exposure, and it has tremendous application prospects.

There is a lack of volume preserving and reasonable deformation of human muscles during bones and joints movement in the field of digital orthopedics. A novel approach for modeling of human muscle and its deformation was put forward to effectively assist doctors in guiding patients to carry out rehabilitation exercises. Firstly, based on Magnetic Resonance Imaging (MRI) data, the generated slice images were used to extract the outer contour lines and then the corresponding contour lines and optimal matching points of the adjacent layer images were connected to construct the three-dimensional (3D) geometric models of the muscles; Secondly, the mapping relationship between parameters can be established through hierarchical definition of the muscle characteristics to realize the volume-preserving deformation of muscle; Finally, the movement of human joints can be realized based on the constraint range of joint movement, and the vector-valued dynamic fourth-order differential equation was proposed to make the characteristic curve dynamically simulate the process of muscle deformation, thereby forming the corresponding relationship between bone movement and muscle deformation. The effectiveness and feasibility of this method have been verified in our experiments with biceps brachii and triceps brachii as examples. The maximum volume errors of biceps brachii and triceps brachii during the deformation process were less than 0.6%, which can be ignored within a certain allowable error range, reflecting that the parametric method was used to realize the reasonable volume-preserving deformation of human muscle.

In clinics, the reduction of femoral intertrochanteric fractures should meet the medical demands of both axis alignment and position alignment. State-of-the-art approaches are designed for merely position alignment, not allowing for axis alignment. The axis-position alignment can be formulated as a least square optimization problem with the inequality constraints. The main challenges include how to solve this constrained optimization problem and effectively extract the semantic of the randomly fractured bone pieces. To address these problems, a semi-automatic data-driven method is introduced. First, the medical semantic parameters are computed, at the beginning of when the 3D input pieces’ anatomical areas are labeled by using the deep neural network. A statistical shape model is leveraged to generate the synthetic training data so as to learn the anatomical landmarks of the pieces, greatly reducing the labeling costs for training. The final reduction position of the pieces is obtained through iterative axis alignment and position alignment. Our method is evaluated by three baselines, i.e., the manual assembly of the orthopaedic specialists and two typical bone assembling methods. The presented method solves an optimization problem for assembling intertrochanteric fracture by axis-position alignment. All cases can be successfully assembled with the developed algorithm which is proved to be capable of reaching the clinical demand.