An Overview of Open Source Deep Learning-Based Libraries for Neuroscience

In recent years, deep learning has revolutionized machine and its applications, producing results comparable to human experts in several domains, including neuroscience. Each year, hundreds of scientific publications present applications neural networks for biomedical data analysis. Due the fast growth domain, it could be a complicated extremely time-consuming task worldwide researchers have clear perspective most advanced software libraries. This work contributes clarifying current situation outlining useful libraries that implement facilitate neuroscience, allowing scientists identify suitable options their research or clinical projects. paper summarizes main developments relevance neuroscience; then reviews neuroinformatic toolboxes collected from literature specific hubs projects oriented neuroscience research. The selected tools are presented tables detailing key features grouped by domain application (e.g., type, area, task), model engineering programming language, customization), technological aspect interface, code source). show that, among high number available tools, stand out terms functionalities applications. aggregation discussion this information can help community develop more efficiently quickly, both means readily knowing which modules may improved, connected, added.