WSE and APL have selected four teams as the inaugural winners of the SURPASS initiative. Chosen from a field of strong contenders, these teams’ proposals have answered the challenge to go beyond possible to seek ambitious and innovative solutions to some of the world’s most pressing societal problems.
Between Earth & Space, The Next Strategic Flight Regime (BEAST)
Melissa Terlaje, Ph.D.
Kevin Hemker, Ph.D.
The team behind BEAST envisions a future where sustained flight in the upper stratosphere is not only possible but also commonplace. While only rockets and missiles are currently able to reach this altitude, air-breathing flight vehicles could be the answer to this level of transportation, enabling the shuttling of people and goods to and from space, providing new modes of surveillance, and paving the way for military missions and capabilities.
To achieve this lofty goal, a team of researchers and experts from WSE’s Department of Mechanical Engineering and APL’s Air and Missile Defense Sector (AMDS) and Research and Exploratory Development Department (REDD) theorizes that it will need to create radical advances to existing technology and capabilities to enable these types of flights, including enhancing the vehicle’s lift and control, creating high-impulse combustion, and applying advanced thermal management of the flight vehicle.
The team detailed a three-pronged approach focused on: (1) vehicle lift and maneuverability, which will require state-of-the-art modeling of hypersonic flow and turbulence; (2) enhanced thrust, via the development of activated microporous liquid fuels; and (3) thermal transport and morphing.
Back Row, Left to Right:
David Van Wie, Ph.D. (APL); Michael Brupbacher, Ph.D. (APL), Joseph Katz, Ph.D. (WSE), Tamer Zaki, Ph.D. (WSE)
Front Row, Left to Right:
Melissa Terlaje, Ph.D. (APL), Kevin Hemker, Ph.D. (WSE), Morgan Trexler, Ph.D. (APL)
Photoacoustic Retinal Prosthesis
Emad Boctor, Ph.D.
Seth Billings, Ph.D.
Millions worldwide experience some form of visual impairment or blindness. Leading causes of blindness include glaucoma, macular degeneration, and diabetic retinopathy. While surgery is an option for patients with cataracts and preserved retinal function, those impacted by irreversible blindness have little hope.
Although artificial retinas and visual prosthetic technologies have evolved over the last decade, current approaches rely on invasive electrode arrays for retinal stimulation and provide limited resolution and temporary benefits that degrade over time.
PIs Emad Boctor, associate research professor at the Whiting School of Engineering’s Laboratory for Computational Sensing and Robotics, and Seth Billings, a project manager in APL’s Research and Exploratory Development Department, working with a team comprised of WSE and APL researchers—including WSE’s James Spicer and Russell Taylor; SOM’s Peter Gehlbach, Martin Pomper, and Raymond Koehler; and APL’s Francesco Tenore, Breanne Christie and David Shrekenhamer—are developing an innovative photoacoustic retinal stimulation (PARS) approach with the potential to help a subset of patients with severe vision loss and degenerative disorders of the outer retina.
As part of their SURPASS proposal, the team detailed a less-invasive wearable technology that would stimulate retinal tissue and restore functional vision. The capability, known as PARS, would involve specially made eyewear with an integrated camera and photoacoustic (PA) laser scanner that can process a live-camera image to initiate a visual simulation pattern. The PA laser scanner then quickly pulses this pattern onto a PA-sensitive implant layer that has been situated in the back of the eye.
To date, “vision restoration devices have required implanting an invasive, active stimulation device into the retina. This approach creates major limitations,” says Boctor. While PARS still requires a passive implant, it lies on the retinal surface without penetration. “A further improvement has been the relocation of the entire stimulation device and its power source outside of the eye. This repositioning allows upgradability, and adaptation of the device during disease progression, as well as high-resolution excitation,” Boctor says.
Boctor tapped Billings, whose research at APL focuses on AI and computer vision with applications in autonomy, robotics, rehabilitative technology, and interventional and diagnostic medicine, to bring this concept to life.
Billings hopes to validate this novel approach with further studies supported by SURPASS funding.
Left to Right:
Francesco Tenore, Ph.D. (APL); Emad Boctor, Ph.D. (WSE); Peter Gehlbach, M.D., Ph.D. (SOM); Seth Billings, Ph.D. (APL); James Spicer, Ph.D. (WSE); Breanne Christie, Ph.D. (APL); Jeeun Kang, Ph.D. (WSE)
Raymond Koehler, Ph.D. (SOM); Martin Pomper, M.D., Ph.D. (SOM); David Shrekenhamer, Ph.D. (APL); Russell Taylor, Ph.D. (WSE)
CEREBRO: Enabling the Next Step of Human Evolution
Amy Foster, Ph.D.
Nicholas G. Povlopoulos, Ph.D.
According to the World Health Organization, one in six people suffers from a neurological disorder. Despite this fact, there is still no way to routinely monitor brain health efficiently and cost-effectively with limited invasiveness. CEREBRO seeks to change that.
Surveying the clinical landscape, the team learned that there is currently no wearable and portable device that is able to easily decode subtle cognitive information across the whole brain during natural everyday activities. Even the most current high-tech and least-invasive brain imaging technology is too cost-prohibitive to allow for mainstream usage.
One of the CEREBRO team’s goals is to make non-invasive brain scans readily available so that they may become as common as routine checkups. The team envisions a future where a data pool of brain-health metrics from millions of brains may help doctors and scientists unlock the mysteries of neurological disorders (such as Alzheimer’s disease) years before onset, and enable non-invasive applications from human performance monitoring to brain computer interfacing. The technology would be in the form of a wearable, portable device for neuroimaging not yet possible because of a “missing link in the technology chain”—a new sensor that can measure a brain’s continually broadcasting magnetic fields in an unprecedented, highly sensitive manner. The sensor works by translating brain signals to the optical domain where extremely small signals can be measured with high precision.
Back Row, Left to Right:
Jacob Khurgin, Ph.D. (WSE); Mark Foster, Ph.D. (WSE); Spencer Langevin (APL); Jeremiah Wathen, Ph.D. (APL); Griffin Milsap, Ph.D. (APL)
Front Row, Left to Right:
Amy Foster, Ph.D. (WSE); Nicholas G. Pavlopoulos, Ph.D. (APL); Konstantinos Gerasopoulos, Ph.D. (APL)
Elisabeth Marsh, M.D., Ph.D. (SOM)
Organoid Intelligence: Synthetic Biological AI
Erik Johnson, Ph.D.
Thomas Hartung, Ph.D., M.D.
AI is quickly permeating society in many ways, from our shopping experiences to the information and news we receive on the internet. However, artificially replicating the function and efficiency of the human brain remains elusive. The essential question for this team is: Can you manufacture intelligence and maintain long-term learning in an artificial brain?
“Despite emerging machine learning models to process images, videos, and text with high performance, progress is requiring exponentially increasing resources without a clear path to human-like intelligence,” said PI Thomas Hartung of WSE and BSPH. “This SURPASS proposal is focused on unlocking the power of biological computing as a potential capability to leap-frog the current state of AI.”
To do so, the team wants to utilize organoid intelligence (OI). Grown in a petri dish under controllable conditions, an organoid is a tiny tissue culture derived from stem cells that can be trained and observed through microelectrode interfaces. By studying an organoid’s behavior, the team can learn about changes in biological neural networks, which provides insight into the synaptic learning that is critical to understanding biological intelligence.
Because this is an emerging field, the team anticipates addressing ethical challenges tied to human stem cell bioengineering. There are also technology gaps to consider. Breakthroughs in neural interface are needed to determine whether input and output signals can be properly delivered to and from an organoid. Current AI technologies that are needed to understand and model brain organoids are also lacking.
Hartung and his group in JHU’s Department of Environmental Health and Engineering have developed a 3D brain-organoid model derived from induced pluripotent stem cells. The model incorporates brain immune cells, known as microglia, which are necessary for neural development and learning. The team aims to grow these brain organoids in phases, increasing biological mass to the ultimate goal of 1 billion neural cells. APL’s PI, Erik Johnson, said he hopes to have working prototypes within 18 months, but acknowledged that longer-term applications of OI for biological sensing and computing may still be many years away.
Left to Right:
Thomas Hartung, Ph.D., M.D. (WSE); Erik Johnson, Ph.D. (APL)
Lomax Boyd, Ph.D. (BSPH); Brian Caffo, Ph.D. (BSPH); David Gracias, Ph.D. (WSE); Erin Hahn, J.D. (APL); Tim Harris, Ph.D. (WSE); Jeffrey Kahn, Ph.D. (BSPH); Bart Paulhamus, D.Eng. (APL); Lena Smirnova, Ph.D. (WSE/BSPH); Francesco Tenore, Ph.D. (APL); Brock Wester, Ph.D. (APL)