
SHINE: Smart Health Innovation & Neonatal Engineering
Background
This course is designed for students interested in the intersection of healthcare and technology, including the field of neonatal care. Advanced Internet-of-Things sensing, edge AI/ML, and ubiquitous connectivity are reshaping healthcare at every stage of early life. Wearable monitors, smart bassinets, and camera-equipped cribs now stream high-resolution physiology and behavior data in real time; cloud-edge analytics can flag subtle warning signs hours before conventional vitals drift; and secure telepresence tools keep families virtually bedside even when geography or infection protocols separate them. These possibilities, however, collide with a stubborn reality: in many clinics data still sit in siloed devices, proprietary dashboards, or handwritten charts, leaving caregivers to stitch together a fragmented picture under time pressure. Without common standards or adaptive decision support, the promise of precision newborn and pediatric care remains under-delivered—especially in resource-constrained settings where staff and equipment are already stretched thin. Our team SHINE brings together students from computing, engineering, and the health sciences to innovate, design and develop end-to-end smart-health solutions that marry IoT hardware, machine-learning analytics, human-computer interface, and human-centered software. While neonatal intensive-care needs remain a flagship domain—given the team’s access to the neonatal intensive care unit (NICU) partners—we intentionally extend our scope to maternal health, pediatric monitoring, and broader acute-care scenarios so innovations can scale across the continuum of early life and beyond.
Our main areas of interest are:
- Development of connected sensing devices (wearables, bedside modules, ambient sensors) that capture vital signs and behavioral cues for infants, children, and mothers.
- AI/ML models that predict clinical risk (e.g., readmission, respiratory compromise, postpartum complications) and optimize resource allocation in critical-care environments.
- To design and implement intelligent, connected systems that enhance neonatal care while also fostering parental engagement and optimizing clinical resources in the NICU.
- Interactive applications and interfaces—mobile, web, XR—that keep families engaged, support clinician workflow, and strengthen parent-patient bonding even when physical presence is limited.
- Data-driven evaluation of how smart-health systems affect outcomes, clinician workload, and health-equity gaps.
Potential Projects
BabyTalk - Remote Reading for Newborns
This project aims to address the emotional distress and lack of parent-infant bonding caused by prolonged NICU stays through a real-time messaging, singing, and reading experience for parents with infants in the NICU. The proposed solution involves creating a system that allows parents to record and send messages, which will be played for the baby by bedside nursing staff during designated hours. This system is expected to improve neuromotor, cognitive, and language development in infants, as well as reduce parental stress and anxiety. We intend to develop of an mobile application and the corresponding NICU system that help mothers who can't be physically present with their babies in the NICU in supporting the parent-infant bonding. The application/system could live-stream video of the baby, play recorded lullabies from the mother, or provide a platform for nurses to share updates. The proposed solution involves designing a web application with a parent interface, nurse interface, and a raspberry pi with a small speaker attached, which will play messages for the infant. All three components will be connected to a virtual machine hosted on AWS, which will serve as a backend to bridge all three interfaces. The system is expected to positively impact parent-infant bonding during prolonged NICU admission time and has the potential to be used commercially by NICU bed/device companies if managed to appear small and able to be integrated into infants' beds/incubators. Overall, this project aims to improve the well-being of both parents and infants in the NICU and has the potential to have a significant impact on the healthcare industry.

MediTalk - Real-time AI-Powered Medical Interpreter System
This project seeks to develop an accurate, time efficient language interpreter system that can be integrated in everyday workflow and is easily accessible in real time during provider to patient interaction. More specifically, the project's primary objective is to create an AI-Powered English to Spanish language simultaneous medical interpreter system, leveraging the capabilities of sophisticated language models like ChatGPT. The system's design encompasses the integration of dependable speech-to-text and text-to-speech services to accurately convert spoken words into a manageable format for the AI model as well a simple user interface. The AI model, a customized version of ChatGPT, will be trained on English-to-Spanish medical conversations. It will be programmed to understand and use complex medical terminologies and adapt its conversational style based on the patient's age or profession, thus offering a personalized interaction. A unique aspect of this system is its ability to simplify and explain complex medical terminologies when the patient faces difficulties in understanding them, further enhancing the comprehensibility of the communication. By successfully developing such an advanced system, we aim to significantly enhance communication between healthcare providers and patients with limited English proficiency. This project not only has the potential to improve patient satisfaction and healthcare outcomes but also to mitigate language-related disparities in healthcare. Its successful implementation could serve as a model for other sectors where language barriers impede productivity. This project provides students with a unique opportunity to gain understanding of AI technology along with practical experience in its implementation for real-world challenges. Students will also learn how to use and integrate cloud-based services such as speech-to-text services. This project will develop their problem-solving skills, technical acumen, and understanding of the societal impact of AI, preparing them for diverse roles in the AI and healthcare technology sectors.
Other Projects
- Development of an application that helps mothers who can't be physically present with their babies in the NICU in supporting the parent-infant bonding. The app could live-stream video of the baby, play recorded lullabies from the mother, or provide a platform for nurses to share updates.
- Development of an AI-based system to predict safe discharge timings for neonates.
- Development of IoT integrated devices such as a music-activated baby bottle to encourage feeding in preterm babies, and a vibrating mattress to stimulate respiration in preterm babies.
- Design of an AI system for optimal incubation needs prediction and resource allocation.
- AI-based risk calculation for early delivery and intervention prediction in early onset severe Intrauterine Growth Restriction (IUGR) cases.
Contact: Prof. Tamer Nadeem (tnadeem@vcu.edu), Dr. Miheret Yitayew (Miheret.Yitayew@vcuhealth.org)