A closeup image of the Lyme disease parasite Borrelia. (PHOTO CREDIT: CDC/Claudia Molina) Despite its presence in over 80 countries, Lyme disease remains one of the least understood illnesses. In the United States alone, an estimated 476,000 people are diagnosed each year. This bacterial infection, spread through tick bites, poses a unique challenge due to the wide range of symptoms it produces. Known as “The Great Imitator,” Lyme disease often mimics other conditions, complicating diagnosis and treatment. Because misdiagnoses are common, actual case numbers may far exceed reported figures. Adding to these challenges, climate change has further exacerbated the issue by expanding tick habitats.
A new study funded by the Department of Defense (DoD) aims to tackle these challenges in a remarkably efficient way. With just $750,000—relatively small compared to many multi-million-dollar research grants—the project leverages existing patient data, advanced machine learning, and a dedicated team of experts to develop better diagnostic tools and treatment strategies. Led by UCLA mathematics professor Deanna Needell, the three-year initiative is a perfect example of how smaller investments, when combined with innovative approaches, can have an outsized impact.
“This project is a perfect example of how the federal government and research universities can work together to improve society here and now.” —Professor Deanna Needell
Professor Needell joined forces with Professor Jana Gevertz of The College of New Jersey and longtime collaborator Lorraine Johnson, CEO of LymeDisease.org. Drawing on their combined expertise in mathematics, data-driven research, and patient advocacy, the team is confident in their ability to shed new light on this perplexing disease.
Professor Needell has collaborated with LymeDisease.org for over a decade. When Johnson launched the patient registry MyLymeData ten years ago, Needell saw an opportunity to use its data for new types of research. Today, the registry includes over 19,000 patients who report on their diagnoses, symptom severity, and treatment outcomes. Over the years, Needell and Johnson have co-authored five papers, blending statistical and machine learning approaches to analyze this data. Their earlier studies have explored antibiotic responses in chronic Lyme disease and key factors influencing treatment success.
Building on earlier successes, the current project focuses on the persistent neurological symptoms of Lyme disease, which affect the nervous system. Patients with these symptoms can experience issues such as neuropathy, twitching, memory loss, cognitive impairment, sleep impairment, and psychiatric symptoms, which have been linked to functional and structural changes in the brain. These symptoms can disrupt daily life. By using topic modeling, a machine learning technique, Needell, Johnson, and Gevertz aim to uncover trends and patterns in patient responses to antibiotics. These insights could directly inform treatment strategies. Recent advancements in AI, combined with new machine learning methodology allow the team to design better algorithms than ever before to define and understand neurological-based Lyme disease.
“Simply put, government grants are essential for solving healthcare challenges,” said Needell, “This project is a perfect example of how the federal government and research universities can work together to improve society here and now.” Programs like the Congressionally Directed Medical Research Programs (CDMRP) enable universities to partner with experts, patient advocates, and organizations to turn research into meaningful change.
Their hope is that by understanding the nature of the persistent neurological manifestations of the disease, they can then study how clinicians can better diagnose and treat the illness, ultimately providing patients with better care to help them live quality lives. In doing so, they will also highlight the potential for precision medicine in Lyme disease and show that data-driven research is crucial for this type of patient care. These results also inform future projects focusing on other symptom clusters, such as musculoskeletal symptoms, that further our understanding of why certain symptoms predominate in some patients to help us develop more individualized treatments.