KLKarla Miriam Reyes Leiva

Karla Miriam Reyes Leiva

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Karla Miriam Reyes Leiva has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Anthropometry and diagnostic aware deep learning for exercise assessment

    by Karla Miriam Reyes Leiva, Pavla Nikelova, Martin Cerny

    Background: Correct technique during strength exercises such as squats and Romanian deadlifts (RDLs) is fundamental for performance and injury prevention. Objective: We introduce ADA (Anthropometry and Diagnostic Aware), a multimodal deep-learning framework that integrates IMU kinematics with anthropometric and diagnostic features to classify movement quality and predict movement related risk. Methods: Seventeen-sensor IMU data were collected from 15 healthy subjects performing correct and incorrect squat and RDL trials. A CNN-LSTM branch processed kinematic sequences and a fully connected branch processed static anthropometric/diagnostic inputs; feature fusion used attention weighting. Results: Incorporating anthropometry and diagnostic context increased sequence-level accuracy from 86.5% (kinematics only) to 94.8% (ADA) and enabled binary risk prediction at 97.8%. Personalized (transfer learning) fine tuning further improved accuracies (mean gains 3%–5% depending on window length). Conclusion: ADA demonstrates that subject-specific static features improve movement quality classification and risk stratification, supporting wearable-based personalized feedback in training and rehabilitation.

  • Design of an Artificial Vision System for Exercise Monitoring in Healthy Aging

    by Karla Miriam Reyes Leiva, Gloria Zapata, Maryame Heydari, Elena Villalba, Ricardo Imbert

    This study presents the design and pilot evaluation of a computer vision–based system for monitoring exercise performance in older adults, aiming to reduce frailty-related risks without the need for wearable sensors. Using MediaPipe and OpenCV, the system tracks posture and movement in real time and provides feedback on exercise execution. A pilot test was conducted with 14 volunteers performing seven exercises from the Vivifrail Spanish program (Wheel A). Performance was evaluated using perfomance analysis, yielding recognition rates between 91.06\% and 100\% across exercises. While the system showed high accuracy in detecting posture and repetitions, challenges such as camera positioning, clothing variability, and the absence of validation in the target population remain. These findings demonstrate the feasibility of computer vision for exercise monitoring and support its potential as an accessible tool for fall prevention and functional assessment in older adults. Future work will focus on clinical validation and integration into mobile platforms for home based use. This approach will allow older population to adequately perform exercises from the Vivifrail program, while professionals, as physiotherapists and geriatricians, can monitor their progress remotely, adjusting the program when needed.

  • Open Access

    SiViS: Simulated multi-patient physiological clinical states for advanced vital sign radar monitoring research

    by Karla Miriam Reyes, Ankit Gupta, Arthur Grosmaire, Silvie Procházková, Petr Matouch, Ivona Závacká, Josef Škroch, Martin Cerny

    Continuous, clinically reliable non-contact monitoring of multiple patients’ vital signs remains a significant technological challenge in care settings. This paper introduces a systematically constructed radar-based dataset designed to test methods for simultaneous multi-patient vital sign monitoring. Developed with Ostrava University Training Hospital, the dataset includes recordings from two-three advanced medical simulation mannequins (SimMan 3G Plus), emulating a wide range of physiological states—from healthy resting to acute emergencies (apnea, cardiac arrest, severe respiratory distress). We varied sensor geometry (top, frontal, lateral views at 1-4 m, 0°/45° angles) and radar parameters (ADC samples, chirp loops, ramp times, frame rates, gains), yielding over 100 uniquely configured sessions. Preliminary beamforming-based processing achieves mean heart-rate and breathing-rate errors near clinical thresholds (MAE ≈ 6.6 bpm for HR, 1.47 bpm for RR), demonstrating the dataset’s utility for developing advanced signal processing and machine-learning pipelines. In keeping with FAIR principles, all data are fully documented and publicly accessible, supporting reproducible research toward noninvasive multi-patient monitoring systems.


  • Decoding the Soccer Pass, a Kinematic Analysis Using Kinovea and Inertial Sensors

    by Carlos Diaz, Hermes Diaz, Karla Reyes

    Sports biomechanics is a specialized discipline that seeks to optimize sports performance through the analysis of biomechanical parameters. However, research conducted in the context of sports in low-resource countries is limited, as specialized systems for these analyzes can be expensive. This study presents a quantitative and descriptive research on the kinematics of the quality of the soccer-pass, by proposing a low-cost system solution using IMU sensors and Kinovea software for the evaluation of the quality of the pass. Inertial sensors were used to collect data on linear acceleration, angular velocity, and orientation of the athlete during the execution of the pass. These sensors were strategically placed on the players' bodies to detect movement during the technical gesture. The results obtained from the combination of Kinovea software and inertial sensors allowed precise kinematic measurements of angles and distances between key anatomical points, providing valuable information on the quality and effectiveness of the movements performed by athletes. Joint angles and distances were evaluated, revealing different approaches between subjects. Significant correlations were found between the speed, acceleration, and flexion of the performing knee. The analysis of anthropometric variables also showed correlations between Body Mass Index and knee flexion in both types of support. In the end, a significant relationship was revealed between speed and flexion of the supporting knee during the technical gesture. The proposed system has the potential to be used as a low-cost solution for the analysis of biomechanical parameters in athletes of soccer.

  • A 3D Printing Soft Prosthesis and an IoT Architecture for Controlling Gestures

    by Jose Luis Ordoñez-Avila, Darwin Daniel Lopez, Diego Ventura, Jose Molina-Palucho, André David Colindres, María Elena Perdomo, Karla Miriam Reyes Leiva, Julio Valentin Fajardo, Manuel Cardona

    Upper limb prosthesis have been evolving since 1967, progressing from mechanical prosthesis to robotic prosthesis, integrating electronics with emerging materials. This work aims to develop a low-cost integrated active upper limb prosthesis using additive manufacturing, controlled by the user with a myoelectric sensor to perform three hand gestures: power grip, two-fingered cup movement, and gripper actions. The prosthesis is based on the reduction of degrees of freedom, the analysis of five additive manufacturing materials, and the implementation of a neural network to control the gestures performed by the prosthesis. The precision obtained in the best model saved by the neural network was 91.25% and a recall of 90.75%. A key contribution of this work is the integration of soft 3D printing materials, which enhance object grasping by conforming to various shapes. In this case, grasping was achieved using a driven cable and a flex sensor in three gestures with different objects.

  • Open Access

    Estimating respiratory rate in freely moving users using independent component and multi-resolution analysis

    by Karla M. Reyes Leiva, Matous Vondal, Miao Yang, Martin Cerny, Ying Wang


  • Biomechanical Analysis of Functional Movement in Athletes Using Kinovea

    by Tracy Cristina Guevara García, María Fernanda Berlioz Parada, Karla Miriam Reyes Leiva


  • Assessing the Reliability of OpenCap and OpenSim as Open Source Softwares for Biomechanical Analysis in Neurological Rehabilitation: A Case Study

    by Arlin Mejia Bronfield, Karla Miriam Reyes Leiva

    The neurological rehabilitation focuses on enhancing functional recovery and improving the quality of life for people who have experienced injuries or diseases affecting the central or peripheral nervous system. This functional recovery includes a follow up of the kinematics of the patients limbs. The use of a open source software such as OpenSim, has been previously proposed as a tool for kinematic analysis, this software allows for highly specialized 3D musculoskeletal modeling, facilitates kinematic analysis and the assessment of force and angles in the lower and upper limbs of the human body. In this context, the propose of this research was to test the reliability of OpenSim for kinematic analysis during neurological rehabilitation. For this goal, the Motricity Index test was done by a group of three healthy participants, this values were used for comparison to the stroke patient who is currently undergoing neurological rehabilitation process. The results demonstrates all the limitations in the range of motion of the patient in comparison to the healthy group due his motor issues, such as muscle spasticity and weakness. This research shows the advantages and limitations of this software and its application in neurological rehabilitation. The goal is to contribute to the development of effective and personalized therapeutic strategies to improve the recovery process for these patients.


  • Implementing Research-Based Teaching Methodology, a Multi-disciplinary Case Study

    by Karla Miriam Reyes Leiva, María Elena Perdomo, José Luis Ordoñez-Avila