Design Of Infant Head Shape Fast Detection Equipment And Deformity Diagnosis Scheme

Cranial deformation can be caused by many reasons. Mostly it is because that infant's head continuously lies on a planar surface causes that part of the cranium gets flattened during the growth while some other parts free of stress expand correspondingly and finally make the whole head look asymmetry. In recent years, the increase of cranial deformation cases raises the interest of the researchers in this field. A traditional way to measure the deformation is testing the length, width and some other indexes using tools such as tapes and Anthropometric calipers. Only a few indexes can be tested through this method and the precision is low. Thus the judge of the head deformities has not come to an agreement. Most of the time, the diagnosis depends on the clinician's subjective feeling. Also, orthopedic devices are relatively expensive, difficult to adjust, and lack of generality.Firstly this paper studies the method of extracting the profile of the head from CT image data. The method of weighted median filtering is used to suppress the noise. The CT image is enhanced by the method of adaptive histogram equalization. Corrosion and expansion is adopted to remove the redundant details in the image. Then, the sobel operator is used to extract the profile. The contour of the head is fitted to an ellipse curve by using the least square method. The elliptic equation is calibrated with the method of quadratic regularization. After calibration, parameters of ellipse can be achieved directly from the elliptic equation. Through the analysis, it is found out that the length and width of the ellipse can be used to estimate the length and width of the head. Then the cephalic index can be calculated to determine the coefficients of cranium and identify the kind of deformation; the angle between the axis of symmetry and the ellipse major axis can access whether the patient is suffering the deformational plagiocephaly.On the basis of the research above, a low cost equipment which can detect the health condition of infant's head rapidly is put forward. A head shape detection module and some conventional detections modules are designed. ARM chip is used as the main control chip. The head shape detection module consists of infrared distance measuring sensors, step motors and encoders. The equipment has a communication module to communicate with the master computer. When the infant is diagnosed as suffering from a cranial deformation, data can be transferred into the master computer to do relative analysis. Meanwhile, the corresponding PC software is also designed.If the infant is diagnosed with illness cranial deformation, more detailed examine should be token. The 3D model is built from the point cloud of the head. Also the model is simplified for the diagnosis analysis. Based on the infant's skull anatomy, the skull model is the divided into 5 areas. Aiming at different characteristics and causes of deformational brachycephaly, deformational scaphocephaly and deformational plagiocephaly, different corresponding load sets are chosen to apply on the finite element model of skull. Then, the stress and strain condition is analyzed for providing the fundamental data for the choice of orthopedic plan and design of orthopedic instruments later.Last, based on changes of infant's head parameters at different times, the growth model of the infant's head is set up. Furthermore the mapping relationship among head contour points in different stages are established. Also, the mapping relationship between the infant's head contour points of different cranial deformation and that of the healthy one. Using deformational plagiocephaly as an example, the comparison between the profiles of deformation and health is carried out to identify the specific area to add load. Using the growth model, the profile of the head after orthopedics is forecasted and analyzed. Based on it, the design scheme of the orthopedic instrument is formed.