A Study On Healing Assessment Of Fractured Long Bone Based On Dynamic Fingerprints

More than 600,000 bone fracture incidents occur each year in China. Knowing the healing progress of fractured bones is not only an important indicator for the doctors to choose the follow-up treatments, but also can help patients to improve their post-operation life. Although there are numerous methods to assess fracture healing of long bones such as radiography, biomechanics and clinician’s experience, but in many cases, these methods are inaccuracy and the radiography is harmful itself. Actually, there is not "Gold Standard" for assessing the healing status of a fracture. In the biomechanical viewpoint, the fracture healing process can be described as the increase in stiffness and strength of the fracture site and it is completed only when these values reach what they were. Even though the local stiffness and strength are positively relative with the healing degree, but measuring them is not applicable in practice. In clinical study, doctors use the whole-bone stiffness as an indicator of the healing process but some scholars doubt whether the whole-bone stiffness are feasible to assess fracture healing of long bones. This thesis investigated the sensitivity of the whole-bone bending stiffness for assessing fracture healing of long bones based on a simplified beam model and the finite element model of an artificaial femur. In the mean time, this thesis also introduced three new methods based on the mode shape changes and studied the sensitivity of the three healing indicators, the Modal Strain Energy Change Ratio (MSECR), Curvature Mode Difference Damage Factor (CMDDF) and Flexibility Difference Curvature Damage Factor (FDCDF), by finite element analyses and modal experiment on the artificial composite femurs. These methods use the mode shape information from before-and after-damage model with easy implementation. From the simplified beam model, the results demonstrated that, at the initial healing stage, the whole-bone stiffness of the fractured bone is extremely sensitive to the variation of the callus stiffness; when the Young’s modulus of the callus reaches 15% that of the intact bone, the whole-bone stiffness rises up to 90% that of the intact bone. After that, the whole-bone bending stiffness increases slowly; it becomes less sensitive to the variation of the callus stiffness. These results support the hypothesis that the whole-bone stiffness is not reliable to assess the healing quality of a callus. The simplified model in this thesis provided a theoretical framework to explain why the whole-bone stiffness is insensitive to the healing process of fractured long bones in the late stage of healing. In structure engineering, vibration-based damage identification techniques have been developed for many years to monitor structural health and detect damages. The repair of fractured bone is an inverse process of structure damage, so the vibration-based damage identification techniques can be used in fracture healing assessment. After finite element analyses and modal experiments, the dynamic fingerprints assessments demonstrated that the healing indicators from the modal strain energy and modal curvature were sensitive to the damage severity and the location at the fracture healing stage. Before the Young’s modulus reach 30% that of the intact bone, the what index declines rapidly. After that the trend slows down and the index is insensitive to the callus modulus. The modal strain energy change radio and curvature mode difference damage factor could be two good potential indicators of bone healing assessment.