Les recherches scientifiques d'Ekico
Shoulder Tendinopathy: Advanced technological monitoring for an effective rehabilitation
BACKGROUND AND OBJECTIVES: Rio, a 2-year-old intact male border collie, developed biceps brachii tenosynovitis with insertional microavulsion following trauma monitored with wearable IMUs, aiming to document change across treatment stages and compare trends with musculoskeletal ultrasound
CONCLUSIONS: Tendiboots™ allows for a comprehensive approach to the evaluation of canine patients, detecting subtle irregularities, monitoring progress, and providing objective data useful for future treatments. Using a treadmill during the data acquisition phase can be useful to reduce the variables of free walking
Objective assessment of limb asymmetry using imu sensors: a sensor-based method for detecting ground reaction force asymmetries in over 1000 dogs
This e-poster presents an objective, sensor-based approach to limb asymmetry assessment in dogs at the trot, designed for routine clinical use.
We report a Grade 0 population reference built from a large real-world cohort recorded by veterinarians and rehab professionals, then show how the reference is surfaced in app for fast interpretation and longitudinal follow up. Symmetry is computed as the absolute percent difference between paired limbs, summarized with percentiles to define a normal range. While figures illustrate peak ground reaction force per kg, the same computation applies to other gait parameters available in the app. The scope of this e-poster focuses on Grade 0 statistics for clarity, with clinical screenshots illustrating higher-grade cases; the broader multi-grade dataset referenced in the accepted abstract exceeds 1000 dogs. This reference is intended for triage and monitoring, not as a single visit diagnostic cut off. Keywords : canine gait, lameness, limb asymmetry, IMU, biomechanics, reference percentiles, real-world data, clinical decision support
Detection of lameness in dogs and identification of the affected limb using a feedforward neural network classifier: a preliminary study based on field data
This preliminary study examined the feasibility of using an automated classifier to diagnose lameness and determine the affected limb in dogs of various breeds, weighing between 7 and 65kg. Early and accurate detection of potentially multifaceted gait abnormality in dogs is as challenging as it is essential in the process of effective treatment and management strategies, especially in the case of low-grade lameness or in asymmetrical individuals.
This study developed and tested the performance of a feedforward neural network classifier, trained on field data collected by users of the dedicated app and connected sensor-based system. The aim is to provide an effective, easy-to-use tool to help veterinary clinicians determine the precise location of lameness in dogs.
Evaluation of Different Induced Lameness Grades in Dogs: A Preliminary Comparative Analysis Using a Novel Sensor-Based System
Canine lameness demands accurate assessment methods to guide clinicians in diagnosis, treatment planning, and rehabilitation. Recent technological advancements, such as sensor-based systems, have significantly improved the ability to objectively evaluate and quantify lameness.
This study presents an innovative approach to the evaluation of induced lameness grades, leveraging a novel analysis software using a sensor-based system and 3D printed semi-spheres of varying diameters.
