Senior Design Projects
ECS193 A/B Winter & Spring 2022
Equine Limb Load Detection Phone App
| ********** | |
| David Hawkins | |
| UC Davis Department of Neurobiology, Physiology & Behavior |
Project's details
| Equine Limb Load Detection Phone App | |
| Despite the best efforts of coaches and jockeys, equine overuse injuries stemming from training and racing continue to end the seasons and lives of racehorses across the globe. During the course of a training block or racing season, the repetitive trauma accumulated through the thousands of strides taken by a racehorse slowly breaks down muscle, tendon, and bone integrity. Without proper recovery and training, microdamage accumulates in musculoskeletal structures resulting in injury and lameness, or catastrophic failure. Since horses cannot communicate when they are overworked or beginning to develop these injuries, new approaches are needed to quantify the training loads on horses, their loading behaviors during layup, and the association of these loads on injury risk and recovery. | |
| We propose development of a phone App that can utilize the phone accelerometer and possibly GPS information to detect acceleration pattern changes that may be prognostic for injury. A cell phone would be secured to the saddle of a horse during training and possibly the bare back during rest. The phone App would record the accelerometer data, analyze that data to detect strides and acceleration “patterns,” and use ML/AI algorithms to detect anomalies in those patterns over weeks/months of training sessions and/or rest periods that may be prognostic for overuse injuries. Subtle changes in acceleration patterns may reflect the onset of injury that is not yet detectable by the trainer. | |
| 1. Stand-alone App available via the App Store or GitHub, or a web-app running in a web-browser that can do the following: A. Collect phone accelerometer data at the fastest sampling rate available (e.g. Androids ~ 400 Hz). B. Save accelerometer data and process these data to quantify limb acceleration profiles. C. Provide a user-friendly interface with the ability to display training session outcomes that are easily understandable by the user. D. Use ML/AI approaches to detect acceleration profile anomalies across training sessions and/or rest periods. 2. App beta tested by users identified by the clients. 3. Clearly and thoroughly documented source code maintained in a GitHub repository. Project goals and deliverables will be refined as part of early discussions between clients and the student team. | |
| The ideal team will have: - an understanding of basic principles of physics, - experience with phone application development, - python or equivalent programming experience, - familiarity with signal processing methods, - experience with ML/AI algorithms and approaches. | |
| ********** | |
| 30-60 min weekly or more | |
| Client wishes to keep IP of the project | |
| Attachment | Click here |
| No | |
| Team members | N/A |
| N/A | |
| N/A |