Congratulations to ECE Associate Professors Kaushik Chowdhury, Raymond Fu, Matteo Rinaldi, ECE Professor Vincent Harris, and BioE Professor Jeff Ruberti whose projects were selected for Phase 1 2018 funding by Northeastern’s GapFund360 program. The GapFund360 helps Northeastern’s researchers bridge the gap between promising lab results and demonstrating a commercially viable prototype. We offer grants and programs designed to catalyze our state-of-the-art technologies, advancing innovation through prototyping, validation, and industry input.
Contactless Wireless Energy Transfer: Anywhere, Anytime Charging Surfaces
Principal Investigator and Team: Kaushik Chowdhury, Yousof Naderi, Ufuk Muncuk, Kai Li
This project involves the design of a software-hardware platform that transforms any surface into an Internet-connected and contactless wireless charger for multiple devices, including laptops and phones. Recent feedback from the prestigious “NYC Creative Destruction Lab” program with 27 acceptances out of 300 applications, and the third place at the competitive NSF Investor Relations workshop, both in 2018, point towards charging without physical contact as a breakthrough technology over the Qi-based full-contact chargers available today. We already have a laboratory prototype composed of a magnetic resonance (MR) beamforming transmitter with software controlled user-interface and MR receiver that charges a device 7.8″ from the transmitter.
The project, if funded, will accelerate the next version of MR-based charging surface development complete with an industrial design casing that delivers power safely over-distance and over-large-area to multiple-devices with high-rate. The steps involved are: (i) integration of an intelligent sensing and object detection component based on magnetic field tomography, (ii) design and build a programmable array of interconnected transmitter coils and amplifiers that avoid mutual interference, (iii) extensive testing with different candidate surface materials, and (iv) professional design, fabrication of the enclosure. This prototype will be demonstrated to potential customers and investors within a year.
Smart AI Trainer by Deep Learned Visual Intelligence
Principal Investigator and Team: Raymond Fu, Songyao Jiang, Fuming Guo: EchoPose
Video based human body parts localization is a fundamental but challenging topic in Artificial Intelligence, due to the complexity of human body movement, occlusions and limited computing resources. In recent years, deep convolutional neural networks have raised a significant achievement on human pose estimation in terms of accuracy. However, deep models are computation hungry which always require a powerful processor to run. Nowadays powerful personal computers are being replaced by portable smart devices such as smartphone. Strict requirement of high performance hardware limits the application of deep learning based software on mobile devices without powerful processors. We aims to create a novel patented deep learning technology that could provide high accuracy pose tracking and estimation algorithms running on CPU based mobile devices in real time. We invented a novel deep learning framework using an encoder-decoder like deep model to extract deep features from input frame and obtain the heatmap of human body joints. Through a novel optimization technique on the deep neural networks, our framework can further reduce the computational cost and parameters significantly by separating a pointwise high-dimensional convolution layer into several pointwise low-dimensional layers. Compared with any existing technologies in the market, our new technology shows superior accuracy while can still achieve real-time performances on mobile phones. It is a breakthrough for technology innovations and certainly leads to many game-changer AI products.
Cation Spin-Engineered Superparamagnetic Mn(Me)-ferrite (Me=3d TMs) Nanoparticles for MRI Contrast Agents and Targeted Magnetohyperthermia Cancer Remediation
Principal Investigator and Team: Vince Harris, Parisa Andalib
We propose a new technology to address a gap in the MRI market in the imaging of liver cancers. The MRI market has been facing a growing concern over the toxicity of the leading Gadolinium (Gd)-based MRI contrast agents (CAs). In 2017, the European Medical Agency banned the use of Gd CAs and the US FDA recently issued a warning to healthcare professionals that their use be restricted to limited cases and applied with the lowest doses possible. The need for low toxicity CAs for cancer imaging has placed a tremendous amount of pressure on producers to provide a safe alternative to Gd based CAs, specifically for liver cancer imaging.
Gap funding at this stage of development is essential. The pathway to development of a new product line for low toxicity agents based on Mn-ferrite is clear to us and involves what we refer to as Cation Spin Engineered nanoparticles. Several of the key steps have been developed and reduced to practice by our research team. However, we are without the financial resources to coalesce the components in order to realize our innovation. Timely funding will help us bridge, secure essential results, and prepare us to aggressively seek external support.
Global Resilience Institute Awardee: Battery-less Infrared Sensor Tags for Reliable Occupancy Sensing (BISTROS)
Principal Investigator and Team: Matteo Rinaldi, Zhenyun Qian, Sungho Kang
Sensor systems for human presence sensing and people counting will drastically improve the efficiency of heating, ventilation, and air conditioning (HVAC) in commercial buildings based on the demand. However, a user-transparent sensor system with the required accuracy, reliability, and cost to deliver such substantial energy savings is currently not available. Here, we propose an occupancy sensor technology that enables low cost and reliable indoor people counting for quick return on investment through dramatical savings on energy cost. The sensors utilize the energy of the infrared radiation emitted from a human body to operate and determine the presence of people within a detection range without consuming any electrical power. Therefore, the battery-less sensor tags can be installed virtually anywhere in a building without the need of periodic maintenance. This project seeks the development of a sensor prototype leveraging the recent proof-of-concept work demonstrated by the PI and Co-PI. A business model and a technology roadmap to the market will also be derived and refined during the project with the mentor. The expected technical deliverables from this project are the key missing parts for a convincing and intuitive demo to potential customers and investors hence extremely important for commercialization.
First in Animal Demonstration of Tendon/Ligament Repair and Replacement
Principal Investigator and Team: Jeff Ruberti, Adam Hacking
Goal: To demonstrate that a new collagenous regenerative patch, capable of delivering collagen to a damaged tissue and confining it to the damaged area will enhance ligament and tendon repair in animal models.
Market: The three most injured soft tissues are the anterior cruciate ligament (ACL), the rotator cuff (supraspinatus tendon) and the Achilles tendon. Each year in the US there are in excess of 500,000 reparative surgical procedures performed on these tissues. The current market for soft tissue repair is $2-3B and estimated to grow to $9-14B by 2024. Recently, a collagen-based device designed to augment the repair of supraspinatus tendon was purchased by Smith & Nephew for $125M up front in cash with another $85M in milestones.
State of Technolology/IP: Currently, each part of the device has been manufactured and tested both mechanically and for its ability to deliver collagen. There is an extensive patent portfolio comprising at least 5 issued patents and 3 patent applications wholly owned by Northeastern.
Approach: The collagen healing device will be manufactured at Northeastern and tested in two animal models (rat and rabbit) representing the top 3 indications for soft tissue repair (ACL replacement, repair of rotator cuff and Achilles tendon injuries)
