Fellows Scholarship - 2020 Davidson Fellows

Biography

Jack Albright is a rising junior from Los Altos, California. As a student at The Nueva School with an extensive background in mathematics and computer science, Jack is keenly interested in developing computational tools to answer biological and medical questions. He is incredibly honored to be named a Davidson Fellow and is humbled to be included in such an esteemed and talented group of young people striving to make positive changes in the world. Jack is grateful to the Davidson Institute for generously supporting his education and for giving him the opportunity to share his work on machine learning and Alzheimer’s disease.

Alzheimer’s disease is the most common neurodegenerative disease in older people, and the sixth leading cause of death in the United States. The disease takes a devastating toll on patients’ daily lives, causing a progressive decline in their cognitive abilities, including memory, language, behavior, and problem solving. Alzheimer’s disease also has a profound effect on healthcare systems. It and related dementias affect almost 50 million people worldwide, and the cost for treating Alzheimer’s patients is currently estimated at $604 billion annually. But, for Jack, the disease is more than just these statistics. Many years ago, his great-grandmother died of Alzheimer’s, and three years ago, his grandmother was diagnosed with Alzheimer’s-related dementia. Currently, there is no cure, and none of the medications currently approved by the FDA to treat Alzheimer’s have been shown to delay or halt its progression. Progress on finding a cure has been slow, and 99.6% of clinical trials for Alzheimer’s drugs fail. It’s thought that this is because patients are being treated too late, when they’re already in advanced stages of the disease. What is really needed is a way to identify patients who are at risk for Alzheimer’s before they become symptomatic. By developing a machine learning model that can accurately predict the progression of Alzheimer’s disease on a month-by-month basis up to seven years into the future, Jack has taken an important step towards achieving this goal.

Jack’s machine learning project is the culmination of several years of work, and like most research, it has had its share of challenges. During the initial design of his models, many of Jack’s ideas were unsuccessful, and he struggled to remain optimistic in the midst of these setbacks. In retrospect, these early challenges were beneficial since they involved applying various techniques for analyzing time series data that others had previously developed. By exhausting these known methods for performing such an analysis, Jack was forced to create his own methods, including his novel preprocessing algorithm, which turned out to be the key step necessary for success. Throughout his project, Jack sought the advice of teachers and mentors and is incredibly grateful for their encouragement and support. He is especially thankful for the suggestions and advice from his computer science teacher Jen Selby, his science teacher Rachel Dragos, and his math teacher Ted Theodosopoulos, as well as the valuable feedback on his manuscript that he received from Daphne Koller and Michael Weiner, the Principal Investigator of the Alzheimer’s Disease Neuroimaging Initiative.

The preprocessing algorithm and machine learning model that Jack developed have a number of practical uses that could benefit Alzheimer’s research and treatment. First, the model can be used for early detection of Alzheimer’s, even in patients who have not exhibited any cognitive decline or other symptoms. Second, the model can be used to identify patients who are at risk for developing Alzheimer’s and who may be good candidates for clinical trials for Alzheimer’s therapeutics. Since a drug could be tested on these patients before cognitive decline begins, it might be possible to identify treatments that are effective at preventing the onset of the disease, a potentially more successful approach than trying to reverse cognitive decline after it has taken hold. Third, by predicting the progression of the disease in individual patients, the model can also be used as a tool for monitoring patients in Alzheimer’s clinical trials. For example, a drug candidate could be evaluated based on its ability to prevent a patient from following the course of the disease predicted by the model. Having access to this sort of tool would allow for a more in-depth evaluation of drug candidates and could identify beneficial effects on patients that would otherwise go unnoticed. On a broader scale, by demonstrating that the use of machine learning to predict the progression of disease is feasible, Jack’s work also lays the groundwork for further research into the modeling of disease progression, both in Alzheimer’s and other disease areas.

Jack’s machine learning research would not have been possible without the exceptional education he has received at The Nueva School, which he has attended since kindergarten. Nueva has enabled Jack to explore his passions for math and computer science by offering him an accelerated curriculum from a young age. This has given him the opportunity to take an incredibly rich set of classes, including Multivariable Calculus, Linear Algebra, Graph Theory, Differential Equations, Mathematical Modeling, Advanced Machine Learning, Computational Biology, and Quantum Information and Computation. In addition, he has attended a number of rigorous math summer camps, including Canada/USA Mathcamp and the Stanford University Mathematics Camp (SUMaC), which have exposed him to diverse math disciplines such as algebraic topology, group theory, and non-Euclidean geometry and allowed him to build friendships with students from around the world who share his love of math.

Jack has been repeatedly recognized for his excellence in math and computer science. His machine learning research was awarded the Robert Wood Johnson Foundation Award for Health Advancement at Broadcom MASTERS and garnered him an invitation to the White House. Moreover, the MIT Lincoln Laboratory named Minor Planet 34328 as “Jackalbright” in recognition of his work. Jack has presented at the Alzheimer’s Association International Conference (AAIC) and has been published in a peer-reviewed journal. A frequent competitor in international and national math competitions, he is a two-time USAMO qualifier and three-time USAJMO qualifier, as well as a member of an all-star team organized by AlphaStar Academy that has won first place at the Stanford Math Tournament for the last two years. At Nueva, he is Co-Lead of the Math Club, and he is a member of Nueva’s AI Task Force, a group of industry leaders and educators formed to develop the emerging AI curriculum at Nueva. Outside of math, Jack has worked with the Alzheimer’s Association to advocate for bipartisan legislation to advance Alzheimer’s research and patient care. He plays forward on the Nueva basketball team, is an avid skier, and enjoys camping and traveling with his family. This summer, Jack has continued to work on Alzheimer’s disease with Dr. Michael Weiner’s lab at UCSF as part of the Brain Health Registry and is currently an intern at the Chan Zuckerberg Biohub conducting research on COVID-19.