Wearable Pulse Oximetry & Motion Sensing with 3D Printed Soft Electronics

PI: Carmel Majidi, Mechanical Engineering, cmajidi@andrew.cmu.edu
Adam Feinberg, Biomedical Engineering, feinberg@andrew.cmu.edu
Website: Majidi: https://www.cmu.edu/me/people/majidi.html
Feinberg: https://www.cmu.edu/engineering/materials/people/faculty/bios/feinberg.html

Executive Summary: Diabetics are at high risk for circulatory disorders that can interfere with wound healing and lead to cardiovascular diseases and neuropathies.Early detection and prevention can be dramatically improved with the aid of continuous, real-time monitoring of vitals like pulse rate and oxygenation.However, existing portable technologies for pulse oximetry are bulky and cannot be comfortably worn during sleep and daily activities. To enable 24/7 monitoring, the team will introduce a soft, lightweight pulse oximetric sensor that is custom-fitted to the foot and integrated into a compression sock with sensors for monitoring leg motion and a Bluetooth module for wireless connectivity.The sensing elements will be produced with a novel 3-D printer developed by the Regenerative Biomaterials & Therapeutics Group (PI: Feinberg) and incorporate stretchable circuit architectures pioneered by the Soft Machines Lab (PI: Majidi). Microfluidic wires or conductive fabrics will be used to connect the sensing nodes to a rigid body-mounted module that contains a microcontroller, wireless transceiver, antenna and battery.

Programmable Device for Optimal Non-Invasive Subcutaneous 2-D and 3-D Imaging

PI: Srinivas Narasimhan, Robotics Institute, srinivas@andrew.cmu.edu
Artur Dubrawski, Robotics Institute, awd@cs.cmu.edu
Website: Narasimhan: http://www.cs.cmu.edu/~srinivas/
Dubrawski: http://www.autonlab.org/autonweb/10223.html

Executive Summary: Existing commercial systems for subcutaneous imaging produce poor quality images, limiting their value in medical diagnosis and patient care.In this project, a prototype of a high quality non-invasive optical imaging device is being built to capture subcutaneous structures.Researchers are utilizing adaptive optics that dynamically modulate illumination and sensing to produce images and 3-D volumes with high contrast.Algorithms have been developed that optimize the lighting and imaging dynamically for every subject based on the individual’s anatomy (tissue thickness and properties, micro-vascular structure, etc.)

Intelligent Delivery of In-Home Hospice and Palliative Healthcare

PI: Zachary B. Rubinstein, Robotics Institute, zbr@cs.cmu.edu
Stephen F. Smith, Robotics Institute, sfs@cs.cmu.edu
Website: Rubinstein: http://www.cs.cmu.edu/~zbr/
Smith: http://www.cs.cmu.edu/~sfs/

Executive Summary: IDSHealthcare is working to provide real-time decision-support and scheduling for the efficient delivery of in-home hospice and palliative healthcare. The system is being designed to leverage dynamic scheduling technology together with two-way communication between the healthcare providers and the patients to provide efficient, cost-effective allocation of resources to patients and empower them throughout their case process. The project team is working to implement IDSHealthcare and integrate it into the communication and user-interface infrastructure of Celtic Healthcare, and will demonstrate the system’s feasibility and benefits by conducting a pilot at Celtic Healthcare to assist in the scheduling and management of care delivery for actual patients in Southwestern Pennsylvania.

Combining External and Implantable Sensors with Machine Learning to Detect Changes in the Health Status in Patients with Systolic Heart Failure

PI: Asim Smailagic, ICES, asim@cs.cmu.edu
Website: http://www.cs.cmu.edu/~./asim/

Executive Summary: Congestive heart failure is a leading cause of mortality, morbidity and hospitalization in the United States. The project team is developing and pilot testing instrumentation and data analysis software that alerts clinicians to patient decline well before the point that hospitalization is called for. The system will use external and implantable sensors to sample patient health parameters much more frequently than is possible during episodic patient-clinician encounters, and will apply modern analytics with predictive power to the resulting data streams. Clinician-facing data management and visualization interfaces will be well matched to clinical work-flow and thought processes.

Contact Us

  Phone: 412-268-3978
  Email: lwinter@andrew.cmu.edu