Advances in Medical Devices 4

The Intelligent Square Stepping Exercise System is designed to facilitate cognitive-motor rehabilitation for older adults with Multiple Sclerosis (MS). This innovative system includes a smart mat equipped with pressure sensors and real-time feedback capabilities, addressing the limitations of traditional square stepping exercises. The mat is lightweight, portable, and features an anti-slip surface for enhanced safety. A dedicated data processing unit processes sensor inputs and communicates with users via a mobile interface. Testing demonstrated accurate step detection, effective feedback delivery, and overall system reliability. The project achieved key milestones in usability, portability, and system integration, while identifying areas for future improvement, including manufacturing quality, sensitivity tuning, and user interface enhancements. This system offers a promising solution for safe and accessible rehabilitation for individuals with MS, addressing their unique needs.

Presenting Author: Manuel Hernandez. PhD
University of Illinois Urbana-Champaign

Bio: Manuel Hernandez received his Ph.D. from the University of Michigan at Ann Arbor in 2012. He is a Teaching Associate Professor in the Department of Biomedical and Translational Sciences at the University of Illinois Urbana-Champaign, and affiliate faculty in the Beckman Institute and Departments of Bioengineering and Health and Kinesiology. His fields of professional interest include aging, movement disorders, and systems neuroscience.

Co-authors: Xiaorui Gu, Prakhar Gupta, Junmin Liu, Hank Zhou, Brian Cisto, Mohammad Afzal Khan, Sam Mason, Robert Motl, and Emerson Sebastiao

The prevalence of physical health conditions brought on by poor sitting postures is rising, particularly among sedentary employees and students. Working with computers at the wrong viewing angle and distance can cause neck strain and eye-related deficiency syndrome. The proposed system will solve this problem by designing and developing a sensor-based posture recognition system (SPRS). The system has base components like the ESP32, HC-SR04—an ultrasonic sensor— and MPU6050—an inertial measurement unit sensor (IMU). The system will continuously monitor the viewing screen distance from the computer to the person working with the computer and the viewing angle of that person concerning the screen. The optimal value for maintaining viewing distance is 50 to 70 cm, and the viewing angle is 15 to 30 degrees. The outcome of the proposed work confirms that the system constantly tracks the angle and distance obtained from the sensors and provides alarming feedback to the users whenever the tracked values are not at the optimal level. In the future, the system will be developed without using any wearable sensor, which will give more convenience to the user. The proposed system will include machine learning concepts to classify the different sitting posture styles.

Presenting Author: Challa Revanth Kumar
Vellore Institute of Technology

Bio: Challa Revanth Kumar is currently pursuing a Master of Technology (M.Tech) in Biomedical Engineering at VIT University. He holds a Bachelor's degree in Mechanical Engineering from Pune University, and has developed a strong interest in interdisciplinary applications of engineering in the biomedical field.

His current research focuses on human posture monitoring systems aimed at preventing neck strain and eye-related issues using embedded technologies, sensor integration, and mobile-based feedback mechanisms.

Revanth has a strong academic and extracurricular background, having secured 3rd place and a Bronze Medal in the 2016 Science Olympiad, and 2nd place in a Running Competition in 2013. He is also actively involved in social initiatives, having volunteered for fundraising activities at Inamigos Foundation.

With a passion for innovation in health-focused technology, he aims to contribute to the advancement of smart healthcare solutions and wearable systems.

Co-authors: Manikandan Pulavendran and Sharmila Nageswaran

Congenital heart disease (CHD) is the most common and deadly birth defect, responsible for approximately 11% of infant mortality globally1 Approximately eight in 1,000 infants have CHD, while critical forms of CHD, or life-threatening CCHD, impact roughly two to four of every 1,000 births. Ultrasound is a proven, safe, and effective tool for diagnosing and assessing congenital heart defects (CHDs) and other pediatric heart conditions. Despite this, over 90% of young children lack access to pediatric ultrasound or echocardiography services in the places they are born or seen for routine care. Frontline clinicians in community and rural settings have limited tools to diagnose and manage life-threatening conditions in infants. By integrating ultrasound hardware components in a novel configuration and pairing with machine learning software, the RAPIDscan multi-model imaging device automates the image acquisition process with a repeatable, 30-second mechanical actuation “sweep” that simulates expert image capture from a single subcostal window placement to allow any level of user to screen and identify at-risk patients in minutes.

Presenting Author: Annamarie Saarinen, CEO and Research PI, Bloom Standard

Bio: Annamarie Saarinen is co-founder, CEO, chief research/regulatory officer for Bloom Standard. She has helped develop and launch 4 separate medical devices into the field - and has led development for the RAPIDscan automated pediatric ultrasound system since inception. Annamarie has a recognized background in public health policy, pediatric research and health economics. She has been a co-PI on 15 multi-center clinical research projects and co-author on 18 published manuscripts tied to diagnostic/screening and device development for pediatric patients. She is also the mother of a child diagnosed with critical heart defects as a newborn, and founded a global health NGO focused on early detection and treatment for children impacted by congenital, metabolic and genetic conditions. 

Co-authors: Alyssa Abo, Gwenyth Fischer, Todd Newman, Ulziikhishig Byambabayar and Rodrigo Robles Medellin

Neuromorphic computing, inspired by the human brain, provides an efficient framework for signal processing. This study presents a MEMS-based reservoir computing model that utilizes coupled MEMS devices to generate dynamic and high-dimensional responses. The proposed system is designed to classify human acceleration activity, specifically distinguishing between walking and running signals. By leveraging the advantages of MEMS technology, including compact size, low power consumption, and computational efficiency, our approach achieves a classification accuracy of 77%. This work demonstrates the potential of MEMS-based reservoir computing in wearable technology and real-time motion analysis applications.

Presenting Author: Mohammad Okour, University of Nebraska - Lincoln

Co-authors: Mohammad Megdadi, Mutaz Al Fayad, Abdallah Al Zubi, Sulaiman Mohaidat and Fadi Alsaleem