METHODS: Sixteen computed tomography scan of SC patients (8 months-6 years old) were imported to Materialise Interactive Medical Image Control System (MIMICS) and Materialise 3-matics software. Three-dimensional (3D) OC models were fabricated, and linear measurements were obtained. Mathematical formulas were used for calculation of OC volume and surface area from the 3D model. The same measurements were obtained from the software and used as ground truth. Data normality was investigated before statistical analyses were performed. Wilcoxon test was used to validate differences of OC volume and surface area between 3D model and software.
RESULTS: The mean values for OC surface area for 3D model and MIMICS software were 103.19 mm2 and 31.27 mm2, respectively, whereas the mean for OC volume for 3D model and MIMICS software were 184.37 mm2 and 147.07 mm2, respectively. Significant difference was found between OC volume (P = 0.0681) and surface area (P = 0.0002) between 3D model and software.
CONCLUSION: Optic canal in SC is not a perfect conical frustum thus making 3D model measurement and mathematical formula for surface area and volume estimation not ideal. Computer software remains the best modality to gauge dimensional parameter and is useful to elucidates the relationship of OC and eye function as well as aiding intervention in SC patients.
RESULT: An automated 3D modeling pipeline empowered by an Artificial Neural Network (ANN) was developed. This automated 3D modelling pipeline enables automated deformation of a generic 3D model of monogenean anchor into another target 3D anchor. The 3D modelling pipeline empowered by ANN has managed to automate the generation of the 8 target 3D models (representing 8 species: Dactylogyrus primaries, Pellucidhaptor merus, Dactylogyrus falcatus, Dactylogyrus vastator, Dactylogyrus pterocleidus, Dactylogyrus falciunguis, Chauhanellus auriculatum and Chauhanellus caelatus) of monogenean anchor from the respective 2D illustrations input without repeating the tedious modelling procedure.
CONCLUSIONS: Despite some constraints and limitation, the automated 3D modelling pipeline developed in this study has demonstrated a working idea of application of machine learning approach in a 3D modelling work. This study has not only developed an automated 3D modelling pipeline but also has demonstrated a cross-disciplinary research design that integrates machine learning into a specific domain of study such as 3D modelling of the biological structures.