Projects | Academic

Footwear Predictive Model

Wed, 01 Dec 2021 00:00:00 +0000

Note: The work in this study was submitted for peer-review to the Footwear Science journal, while preliminary data were presented at the 15th biennial Footwear Biomechanics Symposium. Since the work is still under review, results will not be presented - however preliminary data can be viewed in the “Can We Predict Cushioning Perception from the Mechanical Properties of Shoes?” slides in the “Featured Conference Presentations” section. Code used in this project can be found under the “Footwear-Predictive-Model” repo on my GitHub.

Overview: Footwear companies often conduct research to investigate how running biomechanics or comfort are affected by the mechanical properties of the shoe. However, little research has been done on how runner satisfaction, or how well the shoe is perceived, is influenced by footwear mechanical properties. Therefore, we sought to build a model that could predict runner satisfaction from mechanical properties, which would be a useful tool for footwear companies. Data in this study were from a database compiled across multiple studies in the Brooks Running Research Lab in which subjects ran in shoes on a treadmill and then answered questions related to how well they enjoyed the shoe. The database contained information on 87 subjects, mechanical testing data on 61 unique shoes, and satisfaction ratings from 615 subject-shoe pairings. Satisfaction in a shoes was defined in 3 ways: degree-of-satisfaction on 7-point Likert scale, overall satisfaction on a 3-point Likert scale, and willingness-to-purchase the shoe (yes/no response). Overall satisfaction was derived from the 7-point Likert scale by collapsing all dissatisfied scores (1-3) into a single score and all satisfied scores (5-7) into a single score. Random forest and elastic net logistic regression models were built using each definition of satisfaction as the outcome variable, for a total of 6 models. Predictors used for the models were the mechanical properites measured for each shoe, in addition to subject age, gender, and body mass. For each model, an 80/20 split was used on the data to obtain training and validation sets in order to build the models and then test accuracy, defined by no-information rate (proportion of data belonging to the largest class). All analyses were performed in R, with models built using the ‘caret’ package.

Data Management & Exploratory Analysis

Thu, 30 Sep 2021 00:00:00 +0000

Note: The code for these projects can be found in the “misc-data” repository on my GitHub. While these projects don’t belong to any particular research study, they are responsible for helping me develop my skills in working with large datasets.

Project 1: This project utilized the same dataset as the predictive model study. Here, we sought to determine if types of runners can be identified from running biomechanics using a linear discriminant analysis in R. Types of runners were defined by the industry partner and were based on running goals and/or volume of running performed. Biomechanical data came from studies in which subjects ran on a force-instrumented treadmill in a modified sock liner.

Project 2: This project involved the creation of a new data table from 3 existing datasets. The first dataset was an Excel file with data combined across multiple prototype shoe-testing studies. Examples of information contained in row were: shoe tested, study name (usually the name of a prototype shoe follow by its generation and iteration), season & year (e.g. Fall 2020), subject name, subject characteristics (body mass, height, age), mechanical properties of shoe tested, subject perception of shoe tested, and other survey data. The second dataset was a set of folders that contained MATLAB files. File names were constructed so that embedded within each were the study name, season & year, and subject initials. Each file contained biomechanical data across multiple shoes for one subject for a single study. The third dataset was an Excel file that had more biomechanical data, but for a control condition in each study. Each row contained the following information: subject initials, study name, season & year, and the biomechanical data. I was responsible for merging all 3 datasets into a master Excel file. I wrote code in MATLAB to do the following (more detail can be found in the code):

Locate first dataset and load it into MATLAB.
Create a few new columns from existing columns. Combine “prototype name” and “prototype generation” into a single column. Pull first letter from “first name” and “last name” columns to create subject initials. Combine “season” and “year” into a single column.
Loop through the folders containing the MATLAB files of biomechanical data. From the file names, extract the following: prototype shoe name + generation, prototype iteration, season + year, and subject initials.
Load in the data and pull out names of shoes tested and their associated biomechanical variables of interest. This created a table containing information from Step 3 and biomechanical data.
Due to irregularities and typos, quite a few names of tested shoes and prototype shoes had to be corrected from the MATLAB file. This was achieved by looping through all shoe names and seeing if they existed in a master list of shoe names from the first dataset. If they did not, then the incorrect name was replaced by the correct name by either selecting it from the master list or manually entering the name.
A combined table was then created by merging the first dataset and the table of biomechanical data using tested shoe name, subject initials, prototype shoe name + generation, and season + year.
More code was then written to combine the above table with the third dataset, once again using tested shoe name, subject initials, prototype shoe name + generation, and season + year.

Project 3: The overall goal of this project was to use GPS watch data to answer research questions for our industry partner. GPS watch data was collected from 1000+ runners over several as they went on their routine runs, and when they completed challenges set forth by the company – for a total of 150,000+ unique runs. Due to the timing of project with the end of my postdoctoral fellowship, I was only able to complete an exploratory data analysis regarding running patterns. To do this, I also needed to consolidate data across multiple datasets within a SQLite database. Unless noted, most work was done in Python using Jupyter Notebooks. Some functions I performed during my EDA:

Used Python and SQLite3 to create CSV files from each table within a SQLite database so the data could be used in R, Excel, or Python – performed in Spyder.
Created custom functions to: parse out times of day for use in histograms, parse out runner IDs from strings, and define runner groups from answers to a survey question.
Merged data from datasets using unique identifiers, such as runner ID and task/challenge ID.
Created histograms to check frequency of runs.
Created histograms to visually check for outliers or nonsense data (e.g. value of 0 when 0 is not possible) in variables of interest – cleaned data when necessary.
Ran simple linear regressions to test for relationships using variables of interest. Subsequently generated histograms of Pearson correlation coefficients, using average and 95% confidence intervals to confirm presence of relationships.
Derived a categorical variable from a continuous variable, and which was used to run ANOVAs with Tukey post-hoc tests.
Using R - run linear mixed models to determine if specific running metrics are influenced by time of day for a run
Using R - run linear discriminant analysis to determine if types of runners can be identified from running metrics
Using R - use principal components analysis to see if sub-groups can be distinguished within a given type of runner

Prototype Footwear Testing

Sun, 01 Aug 2021 00:00:00 +0000

Note: Details omitted due to classified nature of studies

Overview: In this set of studies, 4-6 sets of prototype shoes were given to us by an industry partner to test for differences in running biomechanics, and to evaluate runner feedback. In each study, runners were outfitted in a set of retroreflective markers for motion capture and asked to run on a force-instrumented treadmill. For each shoe, runners ran on the treadmill for a set amount of time at a set 8 pace, and then answered survey questions regarding the shoes. Motion capture and force plate data were used to analyze running biomechanics, while survey data were used to quantify runner perception of the shoes.

Injuries in Trail Running

Thu, 01 Jul 2021 00:00:00 +0000

Note: This work was submitted for peer-review to the Journal of Sport and Health Science, so data will not be presented.

Overview: The purpose of this study was to quantify injuries related to trail running, as well as the type of footwear used during trail running. A survey was distributed to 1000+ runners with questions regading injuries sustained during trail running, types of trails used during running, types of terrains of said trails, and footwear preference during running. Chi-square tests were used to determine differences in injury frequency, as well as footwear preference. Multiple marginal indepedence tests describe any associations between injuries and trail/terrain type. All statistical analyses were run in R.

Bone Morphological Changes to Loading History during Growth

Wed, 26 Feb 2020 00:00:00 +0000

Note: This work is currently under manuscript preparation so data will not be presented. Prelminary work of this study was presented at the 40th and 43rd Annual Meeting of the American Society of Biomechanics, the slides of which can be viewed from the “Featured Conference Presentations” section. This work was also funded through a NIH R21 grant and a grant through Center for Human Evolution and Devlopment at Penn State.

Overview: It is well-established that exercise during growth increases bone strength. Previous work in mice suggests that bone shape is similarly plastic, but very little research exists on the subject. Bone shape is important, as it acts like a system of levers to help us move. For example, bony protuberances and bone length can influence moment arm lengths, which impacts locomotor capability. The purpose of this study was to investigate if exercise during growth promotes changes to bone shape in addition to bone strength.

For this study, 30 helmeted guineafowl were split into a sedentary (SED) and exercise (EXE) group. SED birds were housed in small pens that restricted their ability to move. EXE birds were housed in large pens, which allowed plenty of movement freedom. EXE birds were also trained for 30 minutes per day, 4x per week, in which they performed short bursts of high-acceleration running. The protocol last for 14 weeks, at which point birds had almost reached skeletal maturity and were sacrificed. Bird carcasses were scanned using a DXA scanner to get a measure of full-body bone mineral content and density. The tarsometatarsus (TMT), analogous to a fused human foot excluding toes, was extracted scanned using a microCT scanner. Bone strength was assessed using cross-sectional properties at 50% of diaphysis length using ImageJ and the BoneJ plug-in. Bone shape was characterized using a 3D array of 159 custom landmarks (created using Avizo and ViewBox) for each TMT. Geometric morphometrics (i.e. general procrustes analysis followed by principal components analysis) were used to characterize differences in bone shape. Other measures of bone shape, such as length and hypotarsus width (analogous to the human heel or tibial tuberosity at the knee but for the ankle), were also taken. Procrustes ANOVA and t-tests on PC1 scores were used to test if groups could be distinguished. T-tests with Bonferonni corrections were used to test for group differences in linear measurements of bone, body size and composition, and bone composition. Geometric morphometrics and statistical analyses were performed in R.

Moment Arm Changes to Loading History during Growth

Wed, 26 Feb 2020 00:00:00 +0000

Note: This work is currently under manuscript preparation so data will not be presented. Preliminary work for this study was presented at the 44th Annual Meeting for the American Society of Biomechanics.

Overview: Moment arms are often linked to locomotor specialization due to their influence on torque generation, muscle length changes, and muscle shortening velocity. For example, animals adapted for sprinting (cheetas, greyhounds, human sprinters) were found to have small moment arms in their limbs, while animals known for digging tend to have larger moment arms. Small moment arms impair torque generation, but reduce muscle fiber strain and shortening velocity, which are beneficial for power production and rapid limb swing. Large moment arms, meanwhile, amplify a given muscle force to increase torque generation, and thus, are better suited to situations where strength is more important than speed. However, whether moment arms are immutable in response to exercise is largely unknown. Moment arm length is largely determined by joint structure, whose components of muscle and bone are known to be plastic in reponse to loading. Therefore, the purpose of this study was to investigate moment arm adaptability in response to loading history during growth.

In this study, 24 guineafowl were split evenly into 3 groups at 2 weeks-old. An exercise (EXE) group was stored in a large pen that allowed to ample opportunity to run aroud and jump to perches. A sedentary (SED) group was house in a small pen with a low ceiling to discourage movement, including jumping. The third group, called the botox (BTX) group, were house similarly to the SED birds, but also received bilateral intramuscular injections of botox into their plantarflexors to induce paralysis (i.e. muscle unloading). EXE and SED birds received injections of a saline solution instead. Injections were started at 7-8 weeks of age, and were given every 5 weeks for 20 weeks until birds had reached ~6 months of age. Birds were then sacrificed and the left hindlimb was isolated, with all musculature except plantarflexor tendon removed. Moment arms were measured using a tendon travel protocol, which is based on the concept of virtual work. This method assumes that joint work and muscle work are equal, so that moment arms are the ratio of the linear displacment of the dnon to the joint angle displacment (force is equal). Retroreflective markers were attached to the bones in order to use 3D motion capture to track joint movement. A linear transducer was attached to the tendon to measure linear displacement and to keep the tendon under constant tension. Each bird limb underwent 3 trials of 8-10 flexion/extension cycles. Linear vs. angular displacement plots were generated from data compiled across all 3 trials, and a 3rd order polynomial was fit to the data to calculate moment arms. A linear mixed model was used to test for group differences, as well as any group x angle interactions. Moment arm analyses were performed in MATLAB, while statistical analyses were run in R.

Muscle Architecture Changes to Loading History during Growth

Fri, 26 Oct 2018 00:00:00 +0000

Note: The work in this study was published in the Journal of Biomechanics in 2018. Work in this study was also presented at the 41st Annual Meeting of the American Society of Biomechanics, receiving the Journal of Biomechanics Award.

Overview: Muscle function is often studied in relation to locomotor function. For example, sprinters are known to have longer muscle fascicle lengths than non-sprinters, and longer fascicle lengths are correlated with faster times amongst sprinters. Longer fascicles are beneficial to sprinting to due their force-length and force-velocity effects that promote power production. However, it is unknown if the sprinter morphology is a result of genetics or training during adolescence. Therefore, the purpose of this study was to study the effects of high-acceleration training during growth on muscle morphology.

For this study, 30 helmeted guineafowl were split into a sedentary (SED) and exercise (EXE) group. SED birds were housed in small pens that restricted their ability to move. EXE birds were housed in large pens, which allowed plenty of movement freedom. EXE birds were also trained for 30 minutes per day, 4x per week, in which they performed short bursts of high-acceleration running. The protocol last for 14 weeks, at which point birds had almost reached skeletal maturity and were sacrificed. Bird carcasses were scanned using a DXA scanner to get a measure of full-body bone mineral content and density. Two muscles with differing architectures and functions were harvested for investigation. The iliotibialis lateralis pars postacetabularis (ILPO) is a large, hip extensor muscle ith long, parallel fibers and a short tendon. The lateral gastrocnemius (LG) is a plantarflexor at the angle with short, pennated fibers and a long tendon. During running, fibers in the ILPO undergo large length changes while fibers in the LG maintain their length. After formalin fixation, muscles were dissected in separate regions: ILPO into anterior and posterior regions, and LG into proximal, middle, and distal regions. Three muscle fibers were extracted, via digestion in nitric acid, were measured from each section using a macro-magnification camera, ImageJ, and MATLAB. Three sarcomere lengths were also measured from each muscle section using laser diffraction. Other measures included: muscle masses on fresh specimen, bone lengths X-ray, pennation angle, and physiological cross-sectional area (PCSA). Using R, a blocked ANOVA was used to test for exercise effects on optimal fiber lengths, limb length-normalized optimal fiber lengths, sarcomere length, and pennation angle while accounting for natural variations within the muscle (e.g. anterior fibers in ILPO are naturally shorter than posterior fibers). Repeated measures ANOVA was used to test for body mass differences across the experimental protocol using SAS. Meanwhile, t-tests with Bonferroni corrections were used for all other measures and also performed in R.

SED birds had 6% greater body mass than EXE birds, but body composition was not different between groups. SED birds also had 3% longer limbs. Muscle mass and pennation angle were not different between groups. EXE birds were found to have 12% longer optimal fiber lengths and 15% limb-length normalized optimal fibers lengths in ILPO. In the LG, a nonsignificant (p = 0.068) difference of 14% was found for optimal fiber lengths, while normalized optimal fiber lengths were significantly different. PCSA in the SED were larger in both the ILPO and LG (16% and 12% respectively), but these differences were not significant (p = 0.123).

The results of this study suggests that adult muscle morphology may be influenced by exercise during growth. Specifically, morphology may reflect the type of exercise performed. High-acceleration training induced longer muscle fascicles, which are preferably for high power movements. Meanwhile, muscle size was either unchanged or smaller as a result of training. Together, it seems that muscle priortized adaptations for strain and power production over isometric strength.

Relationship Between Muscle Stiffness & Hamstring Strain Injuries

Sun, 12 Jul 2015 00:00:00 +0000

Note: This work is currently in progress, but premilinary data was presented at the 25th Congress of the International Society of Biomechanics. Data from this project was combined with data from another research group for another study, which was presented at the same conference.

Overview: Hamstring strain injuries are common injuries but their mechanism is still unknown. It’s likely that muscle strain (i.e. length changes), instead of force, is the culprit. However, when measured at the time of injury, neither seem sufficiently high enough to cause injury. It is possible that microdamage occurs in the muscle and may lead to injury, similar to fractures in bone. The purpose of this study was to determine if hamstring strain injuries are related to changes in hamstring stiffness, as quantified by shear modulus using ultrasound elastography. Hmastring stiffness was prospectively tracked throughout the season in collegiate sprinters and jumpers, with subjects coming into the lab once per month. Nine sites were identified on each hamstring of both legs: a proximal, middle, and distal portion of each of the biceps femoris long head, semitendinosus, and semimembranosus. On each visit, three shear modulus measurements were taken at each of the 18 sites to get a mean value for each region of each muscle.

While none of the athletes were injured at the beginning of the study, only six were free of any history of hamstring injuries. Throughout the course of the study, only one athlete (male) experienced a hamstring strain, which occurred one week post-initial scans. The athlete indicated that they felt the strain in the middle portion of their semimembranosus on their right leg near ultrasound site. Using this athlete as a case study, we found that in the injured semimembranosus, the midbelly showed the highest shear modulus values, which were roughly 50% higher than the proximal portion and 3x higher than the distal portion. The injured semimembranosus mid-belly also had shear modulus values 2.5-3x higher than the semimembranosus mid-bellies (right leg) of three non-injured males. Distal portions, meanwhile, showed similar values across the injured and non-injured males. While this is a case study and more work is needed, these data provide evidence that hamstring stiffness may be altered prior to strain injury. Possible explanations for increased stiffness include: 1. edema at the injury site leading to increased stiffness or 2. higher tension in areas adjacent to injury due to damage at the injury.

Relationship Between Muscle Stiffness & Sarcopenia

Tue, 07 Jul 2015 00:00:00 +0000

Note: Master’s thesis work.

Overview:
Sarcopenia, the loss of muscle mass with age, is a common problem in the elderly population. Although exercise is a common prescription, older adults have a decreased ability to gain muscle compared to young adults. Evidence also suggests muscles in older adults are stiffer, due to increased glycation in the extracellular matrix. Strain is a mechanical signal for hypertrophy and increased stiffness may cause muscle cells to experience less strain for any given load. Therefore, we hypothesized that response to exercise is impaired in older adults because of increased stiffness.

For this study, 19 rats were split into young (12 months) and old (32-33 months) groups. Each rat underwent 3 sets of 10 maximum eccentric dorsiflexions utilizing electrical stimulation and a dynamometer. Following sacrifice, two dorsiflexor muscles were harvesed - the tibialis anterior (TA) of both legs and extensor digitorum longus (EDL) of the non-exercised leg. The EDL underwent stiffness testing to determine the Young’s modulus of the muscle. The TAs were used in a Western blot analysis to determine cellular response. Focal adhesison kinase (FAK) is a protein that is phosphorylated with stretch, making it a good indicator of exercise response. The ratio of phosphorylated to total FAK was used a measure of cellular to response to exercise and termed FAK activity.

As a result, we found smaller and stiffer muscles in the old rats. There was also greater variability in these measures in the old rats. A negative relationship was also seen between muscle stiffness and size. In older rats, a negative relationship was seen between FAK activity and muscle stiffness, while a positive relationship was seen between FAK activity and muscle size.

Our results describe relationships between muscle size, muscle stiffness, and response to exercise in old age. This study implicates increased muscle stiffness as possible mechanism behind the development of sarcopenia. However, while results suggest the possibility of a relationship between sarcopenia and muscle stiffness, this study lacks the ability to establish a causal link.