UC Santa Cruz neuroscientists are using artificial intelligence to improve their understanding of how brain connectivity influences perception, thoughts, and behavior, speeds up brain mapping for neuroscience.
By integrating AI, they have significantly streamlined the process of aligning thin slices of mouse brain tissue with a reference atlas, making it easier to identify key details such as brain region origins.
A student-driven innovation
This cutting-edge technology was developed by UC Santa Cruz student Alec Soronow. He began the project as an undergraduate in the lab of Euiseok Kim, an assistant professor of molecular, cell, and developmental biology at UC Santa Cruz. Soronow continued working with Kim until the digital tool, Bell Jar, was fully developed and ready for use.
The name “Bell Jar” was inspired by Sylvia Plath’s semi-autobiographical novel about mental breakdown and recovery, which Soronow was reading while working on the initial software version. He and Kim introduced Bell Jar to the scientific community through an article published last month in the open-access journal eNeuro.
Tackling a persistent challenge
Bell Jar makes neuroanatomy analysis more accessible and efficient. For years, researchers at UC Santa Cruz and beyond struggled with outdated tools, many of which were built using MATLAB or other specialized software that eventually became obsolete due to updates.
Previous methods also had difficulties integrating machine learning (ML), limiting their flexibility and accuracy. Bell Jar stands out by leveraging ML techniques to improve precision and efficiency.
Unlike older tools, it is designed to be user-friendly and open-access. By sharing the code on Soronow’s GitHub, the team has ensured that researchers worldwide can customize and refine the tool for their own projects.
Reducing the burden of traditional brain mapping
Bell Jar was developed out of necessity. In the UC Santa Cruz Kim Neuroscience Lab, manually analyzing an entire mouse brain was often tedious.
The process of histology—slicing the brain into ultra-thin sections—introduced human errors such as sections being too thick, too thin, or damaged. Researchers then had to align these slices with a reference atlas, a process that often involved subjective decisions.
Bell Jar solves these challenges by using ML to detect and match neurons across brain sections. “In the past, we relied heavily on human judgment, which introduced subjectivity,” Kim said. “Now, Bell Jar can make these determinations more efficiently and with greater accuracy.”
Traditional brain mapping is a painstaking process. A single research project leading to publication requires analyzing over 100 thin sections from one brain. Each section contains multiple brain regions that need to be identified and analyzed, making the process incredibly time-consuming.
Training new students to perform these analyses is also challenging, as it requires a deep understanding of neuroanatomy. Bell Jar helps ease this burden.
“It enables far more efficient analysis of large experimental datasets that would otherwise need to be reviewed manually,” Soronow explained. “Reducing the time to results allows labs to assess experiments more quickly and make necessary adjustments—enabling higher-throughput science.”
To emphasize its impact, Soronow added, “For a standard experiment in our lab, it saves us about three weeks of manual alignment and counting time per brain.”
A deep connection to brain networks
Soronow’s interest in brain function stems from personal experience. His grandmother suffered from dementia and often cared for him as a child, sparking his curiosity about how brain circuits malfunction.
This interest deepened in college when he took his first developmental biology class and later attended a seminar by Kim that aligned with his passion.
In 2022, Soronow won a Dean’s Award from the Baskin School of Engineering at UC Santa Cruz for the Bell Jar project. He graduated that year with a B.S. in biomolecular engineering and bioinformatics, followed by an M.S. in molecular, cell, and developmental biology in 2024.
He has since founded PlusTen Intelligence, a company that integrates AI and engineered neuron-like cells to develop powerful molecular sensors in handheld devices.
“I’m able to use the unique skill set cultivated in the Kim lab,” Soronow said, “combining wet lab, genetic engineering, innovative software design, and hardware design.”
Kim emphasized that Bell Jar aligns perfectly with his lab’s broader research goals. By improving efficiency and accuracy, the tool allows researchers to focus on the bigger picture: understanding brain connectivity and neural function.
“This tool enables us to conduct research more quickly and effectively,” Kim said. “That is the main point here.”
