Date on Master's Thesis/Doctoral Dissertation


Document Type

Master's Thesis

Degree Name

M. Eng.



Degree Program

JB Speed School of Engineering

Committee Chair

Voor, Michael

Committee Co-Chair (if applicable)

Roussel, Thomas

Committee Member

Roussel, Thomas

Committee Member

Caruso, John

Author's Keywords

bone remodeling; device validation; microstrain; finite element analysis; regression analysis; hip fracture


Lateral falls often lead to hip fracture particularly in the elderly who have low bone mineral density. These fractures frequently lead to indirect mortality soon after injury. Normal use over the course over a lifetime leads to optimized adaptation of the bone in the proximal femur according to normal loading. A lateral fall generates non-normal, lateral loading at the proximal hip where the bone has not adapted to withstand such loading. The resulting fracture is generated by reversed strains on the bone tissue in the femoral neck in the hip in contrast to vertical, quotidian loading. Modern practices for preventing hip fractures are largely supplements or medications while standard exercise and vibration therapies are also used. Preventative measures such as these may help but it is apparent that more is needed.

A supplementary exercise device intended to stimulate lateral, localized bone formation at the hip while providing user quality feedback could be a promising solution to overcoming such high hip fracture rates. The device consists of: a main body in the form a lap plate, two adjustable pad arms that optimally position two impact pads adjacent to the user’s lateral proximal femur, and a knee plate that helps maintain the device in the optimal position on the lap relative to the hip joint. Anabolic thresholds for strain magnitude and strain rate are both shown to be critical metrics for stimulating bone remodeling such as what occurs in the femoral neck. The objectives of this second-generation prototype are to design, fabricate, and validate a versatile device with user performance feedback to allow the user to comfortably achieve the biological thresholds for appropriate anabolic bone remodeling in the femoral neck.

The prototype design was based on anthropometric data representing the typical archetype. Strength of key elements was analyzed via manual calculations and finite element analysis (FEA). Ideal sensor placement was also analyzed via FEA to maximize sensitivity. Quasi-static testing in an MTS machine across the breadth of relevant user settings is performed to translate strain gage output to pad force. The discrete results of this testing were then used to generate a 95% two-sided regression model of continuous predictors for accurate force measurement based on user-specific setting inputs. Custom software was developed to process the raw data and provide user feedback. Additional accelerometer data was processed as a potentially simpler alternative to strain data for feedback to the user regarding proper exercise effort. Dynamic testing was collected on 10 volunteers who perform a swift hip abduction using the prototype which creates three-point bending in the femur that generates strain in the femoral neck. Additional tests were performed to optimize data outcome based on user factors. Pad force rate was converted to theoretical bone strain rate based on data provided by the first-generation device study. Strain and strain rate data were compared to the accepted biological thresholds to stimulate remodeling taken from the prevailing bone biomechanics literature.

A prototype was successfully fabricated after calculations were used to validate design integrity. This device was proven functional in acquiring dynamic data via custom software after use by volunteers of varying anthropometry. Before this dynamic data was acquired, preferred strain gage placement was determined to provide the most sensitive measure of pad impact force by calculating stress profiles at minimum and maximum settings and the regression model was validated via four-in-one plot analysis with an R2 of 99.97%. Ideal instruction and performance of the exercise using the device were refined though a series of sub-studies evaluating data acquisition and data metrics. These sub-studies suggested optimal feedback is achieved through an appropriate knee arm setting and a narrow pad arm setting using extra padding under the instruction to swiftly drive through the pad to a 60 beats per minute metronome without pushing down on the plate. Volunteer data revealed an average peak value of 499.5 N surpassing the 350 N minimum and 450 N suggested force to achieve strain magnitudes above the 1000 µε osteogenic threshold. Similarly, the average strain rate of the volunteers averaged 21509.6 µε/s far exceeding the 10000 µε/s bone remodeling threshold. These findings suggest that this device has the potential induce anabolic bone remodeling at the hip, thus encouraging more study toward aims of reduced hip fracture rates. Acceleration data did not prove to be an alternative to strain data for user feedback.