Definition of ankle

Anatomically, the ankle, or ankle joint, is the part of the foot that connects to the leg. It consists of seven tarsals plus the metatarsals of the foot and the bones of the lower leg.

List of common symptoms

1. Plantar fasciitis / plantar aponeurosis

Inflammation of plantar aponeurosis at the beginning of the medial calcaneus.

2. Anterior talofibular ligament tear

When the talus crus joint is flexed, after being subjected to extreme varus and adduction stress, the talus adducts and rotates in the ankle point, and the lateral ligament is damaged.

3. Talus osteochondral injury

Focal articular cartilage injury / defect with underlying bone edema, fracture and / or cyst formation.

4. Ankle fracture

The ankle joint is composed of the lower end of tibiofibula and talus; Fracture and dislocation are common injuries in orthopedics, which mostly occur after ankle sprain caused by indirect violence.

Summary

The application of biological intelligent manufacturing technology in the field of foot and ankle surgery is mainly divided into three links: first, the front-end surgical planning system mainly restores the accurate three-dimensional structure near the foot and ankle lesions through multi-modal medical images such as CT and MR, which is convenient for doctors to make visual and accurate surgical planning; The second is a high-sensitivity force feedback industrial robot, which is used to assist in surgical operation. It must have high-sensitivity force feedback control and motion control, so that it can realize safe and controllable operation in operation and will not cause any iatrogenic injury to patients; The third is image registration technology based on computer vision, which maps the results of surgical planning to the actual space, combines surgical planning with robot operation control, and realizes the series connection of the whole process. These three links can be summarized as "seeing clearly", "matching accurately" and "doing steadily".

In the future, intelligent manufacturing technology in the field of foot and ankle surgery will develop from the current auxiliary surgery to the leading surgery, so as to realize the intellectualization and mechanization of the whole process. Under the main framework of the above three links, continuously increase the control measures of multiple redundancy, enhance the stability and reliability of the system, and finally achieve the purpose of replacing manual operation, greatly reduce the workload of doctors, liberate doctors' hands, and make doctors become "commanders" rather than "operators".

Clear Sight“---Medical image processing combined with artificial intelligence

Based on the multi-modal foot and ankle region image, the multi-modal features are extracted through the marked data as the training set. Through the training of the deep convolution neural network as shown in the figure below, the parameter model of lesion region segmentation is obtained, and then the automatic and accurate segmentation of the unmarked image is realized. The extracted two-dimensional contour and region are restored to the standard 3D model through three-dimensional reconstruction, so that doctors can intuitively plan the operation of the model.

In the past, the three-dimensional reconstruction of medical image mainly focused on obtaining the edge information of image by interpolation method and edge detection technology, and building the three-dimensional entity of organ based on this; Or using the attributes of the organ to draw the possible shape of the organ in the CT data field; However, at present, the best effect of 3D reconstruction is marching cubes algorithm (MC algorithm for short), and its algorithm flow is shown in the following figure.

“Accurate registration“---Image registration combined with computer vision

The purpose of image registration is to map the restored three-dimensional structure to the actual space. More recently, it is to splice the adjacent scanned point cloud data together. The commonly used algorithm is the nearest point iterative algorithm ICP algorithm. Basic idea of ICP algorithm: ICP (iterative closest point) algorithm is a point set to point set registration method. As shown in the figure below, PR (red point cloud) and Rb (blue point cloud) are two point sets. The algorithm is to calculate how to use translation and rotation to make Pb and PR overlap as much as possible.

Assuming that two three-dimensional point sets X1 and X2 are given, the registration steps of ICP method are as follows: 1. Calculate the corresponding near point of each point in x2 in X1 point set; 2. Obtain the rigid body transformation that minimizes the average distance between the corresponding points, and obtain the translation parameters and rotation parameters; 3. Using the translation and rotation parameters obtained in the previous step for X2, a new set of transformation points is obtained; 4. If the average distance between the new transformation point set and the reference point set is less than a given threshold, the iterative calculation will be stopped, otherwise the new transformation point set will continue to iterate as a new x2.

The point cloud coordinates of the 3D structure at the reconstructed lesion are known, and the coordinates of the actual space need to be obtained by the sensor. At present, the highest precision space navigation device is the optical positioning instrument, which can achieve sub millimeter accuracy. Taking the product of a manufacturer as an example, its positioning range is two overlapping pyramid spaces at a certain distance from the instrument. In this range, the coordinates of a point in space (volume accuracy is sub mm) can be obtained through the marker, and the actual spatial point cloud of the structure at the lesion to be obtained in the registration is obtained by the active marker.

“Stable operation“---High sensitivity force feedback industrial robot

The safety of the surgical process is the top priority of the surgical robot platform. Using the high-sensitivity industrial robot to assist the surgical operation, it must have high-sensitivity force feedback control and motion control, and millisecond control response, so that it can realize safe and controllable operation in the operation without causing any iatrogenic injury to patients. The following are the application scenarios of the surgical robot products of the leading orthopaedic surgical robot manufacturers at home and abroad. At present, the surgical robots are basically in the auxiliary stage, and the surgical operation is still dominated by doctors. Most of these robots play the role of fixed instruments and cooperative devices.

The picture shows the assisted removal of bone tissue of knee joint by Stark Mako orthopaedic surgery robot

The picture shows the iliac joint bone nail implantation assisted by Tianzhi Aerospace Jiji orthopedic surgery robot

The picture shows Medtronic mazor orthopaedic surgery robot assisted spinal screw implantation

（Author： Luo Naiwen）