MIT Lincoln Laboratory has unveiled NeuroTrALE, an innovative open-source software tool designed to streamline the processing of large-scale brain imaging data, potentially accelerating research into neurological disorders like Alzheimer's disease.
Lars Gjesteby, technical staff member at MIT Lincoln Laboratory's Human Health and Performance Systems Group, led the development of NeuroTrALE (Neuron Tracing and Active Learning Environment).
The tool employs machine learning and supercomputing to automate much of the data processing involved in creating detailed brain maps. It features an active learning approach, allowing researchers to manually correct errors and improve the algorithm's performance over time.
The team demonstrated a 90 percent decrease in computing time needed to process 32 gigabytes of data compared to conventional AI methods. Remarkably, a 10,000 percent increase in dataset size resulted in only a 9 to 22 percent increase in total processing time.
NeuroTrALE has already yielded significant results. In a study published in Science, researchers used the tool to quantify prefrontal cortex cell density in relation to Alzheimer's disease, finding lower cell density in certain regions of affected brains.
The project, a collaboration between Lincoln Laboratory and Professor Kwanghun Chung's laboratory at MIT, aims to make NeuroTrALE fully open-source. "It's a grassroots effort by the community where data and algorithms are meant to be shared and accessed by all," Gjesteby stated.
As neurological disorders affect one-eighth of the world's population, tools like NeuroTrALE could prove crucial in advancing our understanding and treatment of these conditions.