Application Of Fractal Dimension In Oral Clinical Prognosis CBCT Radiology Study

Objectives: to propose a new scale for observing prognosis of alveolar bone grafting using CBCT,and to create a new methodology to deal with clinical radiological sample chaos,of which FDR(Fractal dimension ratio)family performancewell in describing the bone regeneration procedure after alveolar cleft bone grafting,non-invasive,also of good usability.Method: · A traditional easy-to-use operator selected to calculate the FD(Fractal Dimension)of ROI(region of interest).Canny operator is selected to extract edge morphologic structure of small ROI,because of its usability and thin line edge.For non image processing professional researcher,we try to combine this widely-used edge extract operator with improved experimental design and statistical method,as an alternative of writing complex image processing program.· Samples were carefully selected with clear edge of the magnification visual field.19 CBCT images of 14 patientswere enrolled,and we adjust gray-scaleand magnification,make the image edge character of homogenization and clear.· FDR and its derivatives is proposed to describe the structure differences between higher and lower bone density.To control the perturbation bringing by noise and the influence of individual difference on FD value,we use the FDR of 2 different areas in one CBCT image,describing the relatively complexity of the micro-bone structure of ROI(Region of Interest)in this study.2 members of FDR family:INFDR(interested & noise area fractal dimension ratio)and ISCBFDR(interested & self cancellous bone area fractal dimension ratio)are proposed to describe the The relatively structure complexity change.· Spearman Rank correlation coefficient was used to detect the relationship of FDR family andmicro-structure of alveolar bonein CBCTimage's ROI.ROI's FD number calculated via box-count algorithm to get INFDR and ISCBFDR sets,and we score the densification degree buy human judge.Finally,INFDR or ISCBFDR's correlation with alveolar bone densification degree is detected by Spearman Rank test.· 3 patient's longitudinal data lines to show how micro bone structure change after surgery 3 patients' longitudinal ISCBFDR is plotted to show the change of micro replanted bone structure,which is coincided with clinical experience.Results: · Clinical CBCT radiology sample has inferior quality mainly because of its low resolution and hard to follow-up,for which medical researchers hardly to get the effective FD without enough image processing experience.· Canny operator is a widely used edge detect algorithm with simple implementation in MATLAB.· FDR counted by box-count can be used to describe the relatively differences between FD,with the noise bring by Canny operator.· INFDR shows significant positive correlation with bone's densification degree under a certain CBCT resolution range.· ISCBFDR reflects alveolar replanted bone's microstructure change.The way ISCBFDR changes coincident with clinical facts.Conclusion: This study explained the difficulty of propelling fractal theory applied onto clinical imaging and the dispute in correlation between FD and tissue form in a sample perspective.A new structure of FD statistics combined Canny edge detect operator with FDR was formed to explore and evaluate the micro bone structure change after alveolar cleft bone grafting surgery for non-professional imaging processors doing fractal clinic imaging research.Coincidence between FDR family and clinical facts shows the great potential of FDR family being used in clinical auxiliary diagnosis and treatment evaluation.