Neurological disorders such as epilepsy, Alzheimers, Parkinson’s and certain forms of dementia are the leading cause of disability and second-leading cause of disease worldwide. These disorders disproportionately affect low-resourced communities due to lack of access to specialized healthcare, and many of these complex diseases lack curative solutions. The need to address neurological disorders is high, yet current diagnostics and treatments are not effective for preventative or personalized care and are not accessible or affordable enough to meet the needs of more than 3 billion people living with neurological disorders.
Flavia Vitale, Associate Professor in Bioengineering in Penn Engineering and in Neurology in Penn Medicine, works to meet this need, developing accessible and affordable solutions for the diagnosis, treatment and rehabilitation of people with neurological disorders.
“I started my research career in biomedical engineering hoping to one day help humanity,” says Vitale, who is also a 2024 recipient of a National Science Foundation (NSF) CAREER Award for her work. “But it wasn’t until I gained a more diverse skill set during my doctoral and postdoctoral research across chemical engineering and materials science that I was able to do that in a real way.”
Vitale’s multidisciplinary skills are what allow her to develop devices that help people living with brain disorders. The CAREER Award is now helping her further apply those skills and actualize some of her first long-term research projects at Penn.
“This CAREER Award will support my lab’s current research in leveraging innovation in materials and fabrication approaches to develop devices that are able to interface with and control different chemical and electrical signals inside the brain,” she says.
Focused primarily on understanding the brain activity involved in epilepsy-induced seizures, Vitale aims to design and develop brain-interface devices to pinpoint and suppress uncontrolled brain activity to prevent seizures from happening. Her work will lead to revolutionary health care for the 30% of epilepsy patients whose conditions are drug resistant. Currently those patients either wait out the uncontrolled brain activity and oftentimes life-threatening convulsions, or hope to be eligible for invasive surgeries to remove the part of the brain where seizures originate or to implant the seizure-controlling devices that are currently available.
How do we measure chaos and why would we want to? Together, Penn engineers Dani S. Bassett, J. Peter Skirkanich Professor in Bioengineering and in Electrical and Systems Engineering, and postdoctoral researcher Kieran Murphy leverage the power of machine learning to better understand chaotic systems, opening doors for new information analyses in both theoretical modeling and real-world scenarios.
Humans have been trying to understand and predict chaotic systems such as weather patterns, the movement of planets and population ecology for thousands of years. While our models have continued to improve over time, there will always remain a barrier to perfect prediction. That’s because these systems are inherently chaotic. Not in the sense that blue skies and sunshine can turn into thunderstorms and torrential downpours in a second, although that does happen, but in the sense that mathematically, weather patterns and other chaotic systems are governed by physics with nonlinear characteristics.
“This nonlinearity is fundamental to chaotic systems,” says Murphy. “Unlike linear systems, where the information you start with to predict what will happen at timepoints in the future stays consistent over time, information in nonlinear systems can be both lost and generated through time.”
Like a game of telephone where information from the original source gets lost as it travels from person to person while new words and phrases are added to fill in the blanks, outcomes in chaotic systems become harder to predict as time passes. This information decay thwarts our best efforts to accurately forecast the weather more than a few days out.
“You could put millions of probes in the atmosphere to measure wind speed, temperature and precipitation, but you cannot measure every single atom in the system,” says Murphy. “You must have some amount of uncertainty, which will then grow, and grow quickly. So while a prediction for the weather in a few hours might be fairly accurate, that growth in uncertainty over time makes it impossible to predict the weather a month from now.”
In their recent paper published in Physical Review Letters, Murphy and Bassett applied machine learning to classic models of chaos, physicists’ reproductions of chaotic systems that do not contain any external noise or modeling imperfections, to design a near-perfect measurement of chaotic systems to one day improve our understanding of systems including weather patterns.
“These controlled systems are testbeds for our experiments,” says Murphy. “They allow us to compare with theoretical predictions and carefully evaluate our method before moving to real-world systems where things are messy and much less is known. Eventually, our goal is to make ‘information maps’ of real-world systems, indicating where information is created and identifying what pieces of information in a sea of seemingly random data are important.”
Enamored by the chemical processes of life, Yihui Shen, J. Peter and Geri Skirkanich Assistant Professor of Innovation in Bioengineering, started her research career as a chemist studying the way that proteins fold and the intricate dynamics underlying life processes.
“As an undergraduate, I studied physical chemistry, thinking that one day I’d be addressing challenges in hardcore STEM fields,” she says. “It wasn’t until I observed the dynamics of a single protein molecule that I fell in love with microscopy. I realized that this imaging tool could not only help us observe biological processes on a small scale, but it could also provide new insight at the interface of engineering, chemistry and physics and solve problems on a large scale.”
When Shen turned her attention to microscopy, the field itself was advancing quickly, with improvements being made and new techniques being released every month. Without missing a beat, Shen dove deeper into the most current tools available when she joined Dr. Wei Min’s lab at Columbia University as a doctoral student.
“Professor Wei Min is a pioneer in a new imaging technique called coherent Raman imaging,” says Shen. “In this type of microscopy, we focus light on a very specific point in the cell and measure the amount of scattered light that comes back after exchanging energy with the molecular vibration. This approach allows us to visualize the spatial distribution of different molecules, the very chemistry of life I had studied as an undergraduate, at a high enough resolution to gain insights into biological processes, such as tissue organization, drug distribution and cellular metabolism.”
With this new tool under her belt, Shen was able to ask the kinds of questions that could connect the use of this observation tool to practical applications for real-world challenges.
“I started thinking outside the box,” says Shen. “What if we could observe the chemical exchanges involved in metabolism as they are happening on the scale of a single cell, and then use that insight to pinpoint the exact metabolic pathways and molecules that facilitate tumor growth and disease?”
Team Current Care (Andrew Lee, Antranig Baghdassarian, Johnson Liu, Leah Lackey, Brianna Leung, and Justin Liu), took home the $3,000 Grand Prize in the Cornell Hackathon.
Brianna Leung, a rising senior majoring in Bioengineering and minoring in Neuroscience and Healthcare Management at the University of Pennsylvania, led a diverse team of student scientists and engineers to resounding success at the 2024 Cornell Health Tech Hackathon, where the team won the $3,000 Grand Prize.
Held in March 2024 on Cornell’s campus in New York City, the event brought together students from 29 different universities for a weekend of finding “hacks” to patient wellness and healthcare issues inspired by the theme of “patient safety.”
ADAPT members enjoy a pancake-making marathon in preparation for their pancake sale.
Leung serves as President of Penn Assistive Devices and Prosthetic Technologies (ADAPT), a medical-device project club whose members pursue personal projects, community partnerships and national design competitions. Penn ADAPT’s activities range from designing, building and improving assistive medical devices for conditions such as cerebral palsy and limb loss, to community engagement activities like their semesterly 3D-printed pancake sale.
In her role, Leung has increased the program’s hackathon participation to give club members greater exposure to fast-paced, competition-based design. She also leads the HMS School project, which develops and manufactures switch interfaces for children with cerebral palsy, enabling these students to interact with computers.
Leung’s passion for medical devices extends to her academic research. As a member of the robotics lab of Cynthia Sung, Gabel Family Term Assistant Professor in Mechanical Engineering and Applied Mechanics, Computer and Information Science, and Electrical and Systems Engineering, Leung characterizes origami patterns for energy-saving applications in the heart and in facial reconstruction. Leung has also served as Vice President External for the Penn Lions and Vice President of Member Engagement for the Wharton Undergraduate Healthcare Club, and belongs to the Phi Gamma Nu professional business fraternity.
ADAPT members working on medical devices.
For the Cornell Hackathon, Leung’s team developed a prototype for Current Care, a closed-loop device to prevent pressure ulcers through electrical muscle stimulation. Pressure ulcers, often called bed sores, result from prolonged pressure, which often occurs during extended hospitalization or in patients who are bedridden. This condition is exacerbated by understaffing and strained resources, and can create an extra burden on hospitals, patients and healthcare workers. The U.S. Department of Health and Human Services estimates that pressure ulcers cost the U.S. healthcare system approximately $9.1 billion to $11.6 billion per year.
Current Care is designed to deliver electrical stimulation, which increases blood flow to affected body parts. Conceptualizing and designing complex devices on short notice is the nature of a hackathon, so the team focused their efforts on creating proof-of-concept prototypes for all the different sensors required for the device, as well as providing the judges with on-screen read-outs to demonstrate the logic and hypothetical inputs for the device.
For their design, the team was awarded the $3,000 Grand Prize in the Cornell Hackathon. In addition to Leung, the team consisted of Johnson Liu (Cornell ECE & MSE’26); Antranig Baghdassarian (Cornell BME’27); Andrew Lee (Weill Cornell M.D.’25); Leah Lackey (Cornell ECE Ph.D.’28); and Justin Liu (Northeastern CS’27).
In choosing a project, Leung was inspired by her late grandmother’s experiences. “My role on the team largely consisted of coordinating and leading aspects of its development as needed. I also ultimately presented our idea to the judges,” she says. “This was actually all of my teammates’ first hackathon, so it was really exciting to serve a new role (considering it was actually only my second hackathon!). I had a lot of fun working with them, and we have actually been meeting regularly since the event to continue to work on the project. We had a range of expertise and experience on our team, and I deeply appreciate their hard work and enthusiasm for a project that means so much to me.”
Having found success at the Cornell hackathon, the team is discussing next steps for Current Care. “Our team is still very motivated to continue working on the project, and we’ve been speaking with professors across all of our schools to discuss feasibility and design plans moving forward,” says Leung.
Several other projects developed by Penn ADAPT members were recognized in the Cornell Hackathon:
ADAPT members and Hackathon participants, left to right: Brianna Leung, Rebecca Wang, Claire Zhang, Amy Luo, Mariam Rizvi, Natey Kim, Joe Kojima. Also in attendance but not pictured: Suhani Patel, Harita Trivedi, Dwight Koyner.
Claire Zhang, a sophomore studying Bioengineering and Biology in the VIPER program, was a member and presenter for team CEDAR (winner of Most Innovative/2nd Place), a portable ultrasound imaging device used to monitor carotid artery stenosis development in rural areas.
Natey Kim, a sophomore in Bioengineering, was a member and presenter for team HMSS (finalist), a low-cost digital solution for forecasting infections in hospitals.
Rebecca Wang, a sophomore in Bioengineering and Social Chair of Penn ADAPT, was a member of Team Femnostics (winner of Most Market Ready/4th Place) which developed QuickSense, an all-in-one diagnostic tool that streamlines testing for a handful of the most common vaginal disease infections simultaneously.
Mariam Rizvi, a sophomore in Computational Biology, was a member of team IPVision (winner of Most Potential Impact/5th Place), an application programming interface (or API) that integrates into electronic health records such as Epic, leveraging AI to detect intimate partner violence cases and provide personalized treatment in acute-care settings.
Suhani Patel and Dwight Koyner worked with team RealAIs, which developed a full-stack multi-platform application using React Native and Vertex AI on the Google Cloud Platform (GCP). Patel, a sophomore double majoring in Bioengineering and Computer and Information Science in Penn Engineering, serves as ADAPT’s treasurer, while Koyner is a first-year M&T student studying Business and Systems Engineering in Penn Engineering and Wharton.
Team Femnostics from left to right: Antonia Li, Bhavishya Agarwal, Urmila Sehrawat, Rebecca Wang, Justin Xiang, and Edward Kim.Team IPVision from left to right: Elizabeth Madamidola, Keshika Gopinathan, Emily L. Leventhal, Mariam Rizvi, Raquel Castromonte, and Lina ChihoubTeam Cedar from left to right: Claire Zhang, Ethan Tse, Danny Nguyen, Allen Dinh, and Reem UllayTeam RealAIs from left to right: Rishi Shah, Suhani Patel, Saanvi Mehra, Dwight Koyner, Chimdi Ejiogu, and Garrett Breslin
Learn more about Penn ADAPT here and follow their Instagram.
Read more about the 2024 Cornell Tech Hackathon in the Cornell Chronicle.
Lipid nanoparticles present one of the most advanced drug delivery platforms to shuttle promising therapeutics such as mRNA but are limited by the time it takes to synthesize cationic lipids, a key component. Now, Michael Mitchell and his team at the School of Engineering and Applied Science have developed a faster way to make cationic lipids that are also more versatile, able to carry different kinds of treatments to target specific organs. (Image: iStock / Dr_Microbe)
Imagine a scenario where a skilled hacker must upload critical software to update a central server and thwart a potentially lethal virus from wreaking havoc across a vast computer network. The programmer, armed with the lifesaving code, must navigate through treacherous territory teeming with adversaries, and success hinges on promptly getting a safe, stealthy delivery vehicle that can place the hacker exactly where they need to be.
In the context of modern medicine, messenger RNA (mRNA) serves as the hacker, carrying genetic instructions to produce specific proteins within cells that can induce desired immune responses or sequester maladaptive cellular elements. Lipid nanoparticles (LNPs) are the stealthy delivery vehicles that transport these fragile mRNA molecules through the bloodstream to their target cells, overcoming the body’s defenses to deliver their payload safely and efficiently.
However, much like building an advanced stealth vehicle, the synthesis of cationic lipids—a type of lipid molecule that’s positively charged and a key component of LNPs—is often a time-consuming process, involving multiple steps of chemical synthesis and purification.
Now, Michael Mitchell and a team at the University of Pennsylvania have addressed this challenge with a novel approach that leverages a compound library fabrication technique known as “click-like chemistry” to create LNPs in a single, simple step. Their findings, published in the journal Nature Chemistry, show that this method not only speeds up the synthesis process but also presents a way to equip these delivery vehicles with a “GPS” to better target specific organs such as the liver, lungs, and spleen, potentially opening new avenues for treating a range of diseases that arise in these organs.
“We’ve developed what we call an amidine-incorporated degradable (AID) lipid, a uniquely structured biodegradable molecule,” Mitchell says. “Think of it as an easy-to-build custom mRNA vehicle with a body kit that informs its navigation system. By adjusting its shape and degradability, we can enhance mRNA delivery into cells in a safe manner. By adjusting the amount of the AID lipid that we incorporate into the LNP, we can also guide it to different organs in the body, much like programming different destinations into a GPS.”
First author Xuexiang Han, a former postdoctoral researcher in the Mitchell Lab, explains that their new approach allows the rapid creation of diverse lipid structures in just an hour, compared to the weekslong process traditionally required.
Congratulations to Alison Pouch, Assistant Professor in Bioengineering in the School of Engineering and Applied Science, and in Radiology in the Perelman School of Medicine, on winning a 2024 Cardiac Center Innovation Award for scientific research from the Children’s Hospital of Philadelphia (CHOP)’s Philly Spin-In. Pouch’s study, titled “Systemic Semilunar Valve Mechanics and Simulated Repair in Congenital Heart Disease,” is a collaboration with Matthew Jolley, Assistant Professor of Anesthesiology and Critical Care at CHOP:
“Through biomechanical assessment, Drs. Matthew Jolley and Alison Pouch are leading an interdisciplinary CHOP-Penn team that plans to determine why current approaches to systemic semilunar valve (SSV) repair fail. They will also investigate methods to design improved repairs before going to the operating room by using computational simulation to iteratively optimize repair.
‘We believe that understanding biomechanics of abnormal SSVs and explorations of simulated repair will markedly improve our ability to characterize, risk stratify, and surgically treat SSV dysfunction, thereby improving long-term outcomes and quality of life in patients with SSV dysfunction,’ Dr. Jolley said.”
Pouch’s lab focuses on 3D/4D segmentation and modeling of heart valves in echocardiographic images with applications to surgical treatment of valvular regurgitation as part of the Penn Image Computing and Science Laboratory.
The instrument imaging team, from left: Philadelphia Orchestra bassist Duane Rosengard; Peter Noël, PhD, director of CT Research at the Perelman School of Medicine; luthier Zachary S. Martin; Leening Liu, a PhD student in Noël’s Laboratory of Advanced Computed Tomography Imaging; and Mark Kindig.
When you’re an expert in medical CT imaging, two things are bound to happen, says Peter Noël, PhD, associate professor of Radiology and director of CT Research at the Perelman School of Medicine. One: You develop an insatiable curiosity about the inner workings of all kinds of objects, including those unrelated to your research. And two: Both colleagues and complete strangers will ask for your help in imaging a wide variety of unexpected items.
Over the course of his career, in between managing his own research projects, Noël has imaged diverse objects ranging from animal skulls to tree samples from a German forest, all in the name of furthering scientific knowledge. But none has intrigued him as much as his current extracurricular project: the first known attempt to perform CT imaging of some of the world’s finest string basses.
The goal is to crack the code on what makes a world-class instrument. This knowledge could both increase the ability to better care for masterworks built between the 17th and 19th centuries, as well as providing insights into refining the building of new ones, including possibly shifting from older, scarcer European wood to the use of sustainably harvested U.S. wood.
That’s why Noël and Leening Liu, a PhD student in Noël’s Laboratory of Advanced Computed Tomography Imaging, have found themselves volunteering to run the basses through a Penn CT scanner occasionally, when they’re not developing next-generation CT technology.
“We always learn something out of projects like this … the more appealing part is that medical research can also be applied to non-medical things,” Noël said. “We have the opportunity to take what we learn in medicine and use it for something else—in this case, moving the arts forward.”
Leening Liu is a Ph.D. student in Bioengineering. She is a member of the Laboratory for Advanced Tomography Imaging (LACTI) with research interests including clinical applications of spectral CT and spectral CT thermometry.
Congratulations to the 2024 Bioengineering student recipients of the annual Penn Engineering Graduate Student Awards! The awardees were honored in a ceremony on May 15, 2024, hosted by Dean Vijay Kumar and graduate program faculty leadership.
Master’s Student Awards: Elizabeth Brown – Outstanding Service Tianyu Cai – Outstanding Research Ekta Singh – Outstanding Service
PhD Student Awards: Dimitris Boufidis – Outstanding Service Katherine Mossburg – Outstanding Service Kelsey Swingle – Outstanding Teaching
Cesar de la Fuente (left), Fangping Wan (center), and Marcelo der Torossian Torres (right). Fangping holds a 3D model of a unique ATP synthase fragment, identified by their lab’s deep learning model, APEX, as having potent antibiotic properties.
“Make sure you finish your antibiotics course, even if you start feeling better’ is a medical mantra many hear but ignore,” says Cesar de la Fuente of the University of Pennsylvania.
He explains that this phrase is, however, crucial as noncompliance could hamper the efficacy of a key 20th century discovery, antibiotics. “And in recent decades, this has led to the rise of drug-resistant bacteria, a growing global health crisis causing approximately 4.95 million deaths per year and threatens to make even common infections deadly,” he says.
De la Fuente, a Presidential Assistant Professor, and a team of interdisciplinary researchers have been working on biomedical innovations tackling this looming threat. In a new study, published in Nature Biomedical Engineering, they developed an artificial intelligence tool to mine the vast and largely unexplored biological data—more than 10 million molecules of both modern and extinct organisms— to discover new candidates for antibiotics.
“With traditional methods, it takes around six years to develop new preclinical drug candidates to treat infections and the process is incredibly painstaking and expensive,” de la Fuente says. “Our deep learning approach can dramatically reduce that time, driving down costs as we identified thousands of candidates in just a few hours, and many of them have preclinical potential, as tested in our animal models, signaling a new era in antibiotic discovery.” César de la Fuente holds a 3D model of a unique ATP synthase fragment, identified by his lab’s deep learning model, APEX, as having potent antibiotic properties. This molecular structure, resurrected from ancient genetic data, represents a promising lead in the fight against antibiotic-resistant bacteria.
These latest findings build on methods de la Fuente has been working on since his arrival at Penn in 2019. The team asked a fundamental question: Can machines be used to accelerate antibiotic discovery by mining the world’s biological information? He explains that this idea is based on the notion that biology, at its most basic level, is an information source, which could theoretically be explored with AI to find new useful molecules.
The growing threat of antimicrobial resistance demands innovative solutions in drug discovery. Scientists are turning to artificial intelligence (AI) and machine learning (ML) to accelerate the discovery and development of antimicrobial peptides (AMPs). These short strings of amino acids are promising for combating bacterial infections, yet transitioning them into clinical use has been challenging. Leveraging novel AI-driven models, researchers aim to overcome these obstacles, heralding a new era in antimicrobial therapy.
A new article in Nature Reviews Bioengineering illuminates the promises and challenges of using AI for antibiotic discovery. Cesar de la Fuente, Presidential Assistant Professor in Microbiology and Psychiatry in the Perelman School of Medicine, in Bioengineering and Chemical and Biomolecular Engineering in the School of Engineering and Applied Science, and Adjunct Assistant Professor in Chemistry in the School of Arts and Sciences, collaborated with James J. Collins, Termeer Professor of Medical Engineering and Science at MIT, to provide an introduction to this emerging field, outlining both its current limitations and its massive potential.
In the past five years, groundbreaking work in the de la Fuente Lab has dramatically accelerated the discovery of new antibiotics, reducing the timeline from years to mere hours. AI-driven approaches employed in his laboratory have already yielded numerous preclinical candidates, showcasing the transformative potential of AI in antimicrobial research and offering new potential solutions against currently untreatable infections.
Recent advancements in AI and ML are revolutionizing drug discovery by enabling the precise prediction of biomolecular properties and structures. By training ML models on high-quality datasets, researchers can accurately forecast the efficacy, toxicity and other crucial attributes of novel peptides. This predictive power expedites the screening process, identifying promising candidates for further evaluation in a fraction of the time required by conventional methods.
Traditional approaches to AMP development have encountered hurdles such as toxicity and poor stability. AI models help overcome these challenges by designing peptides with enhanced properties, improving stability, efficacy and safety profiles, and fast-tracking the peptides’ clinical application.
While AI-driven drug discovery has made significant strides, challenges remain. The availability of high-quality data is a critical bottleneck, necessitating collaborative efforts to curate comprehensive datasets to train ML models. Furthermore, ensuring the interpretability and transparency of AI-generated results is essential for fostering trust and wider adoption in clinical settings. However, the future is promising, with AI set to revolutionize antimicrobial therapy development and address drug resistance.
Integrating AI and ML into antimicrobial peptide development marks a paradigm shift in drug discovery. By harnessing these cutting-edge technologies, researchers can address longstanding challenges and accelerate the discovery of novel antimicrobial therapies. Continuous innovation in AI-driven approaches is likely to spearhead a new era of precision medicine, augmenting our arsenal against infectious diseases.
The de la Fuente Lab uses use the power of machines to accelerate discoveries in biology and medicine. The lab’s current projects include using AI for antibiotic discovery, molecular de-extinction, reprogramming venom-derived peptides to discover new antibiotics, and developing low-cost diagnostics for bacterial and viral infections. Read more posts featuring de la Fuente’s work in the BE Blog.
