B.R.A.I.N.
Brain tumor Recognition and Identification Network
What is B.R.A.I.N.?
B.R.A.I.N. is my brainchild from high school. The culmination of over two years of work, brain is an advanced artificial intelligence and machine learning based program aimed to identify and diagnose brain tumors. Officially titled “A Comparative Analysis of the Efficacy of Convolutional Neural Networks and Vision Transformers in the Diagnosis of Glioblastomas and Astrocytomas”, B.R.A.I.N. was my dream project to help reduce brain tumor misdiagnosis and recognition failure.
How does B.R.A.I.N. work?
B.R.A.I.N. is built from both convolutional neural networks and vision transformers. B.R.A.I.N. intakes T1W files from MRIs and outputs a simple “yes” or “no” if there is a tumor or growth in the file.
How was B.R.A.I.N. trained?
B.R.A.I.N. was subjected to rigorous testing, by evaluating over 10,000 brain scans for tumors. Trained with over 1000 scans, B.R.A.I.N. peaked at a 98.6% accuracy in identifying brain tumors.
What are Convolutional Neural Networks and Vision Transformers?
As a note, an in-depth, simplistic guide can be found in my article, linked here. You will want to look for the section, “A Perfunctory Explanation of CNNs and VTs”.
Put simply, CNNs, also known as ConvNets and convolutional neural networks, are a type of deep learning model commonly used for image recognition and computer vision tasks. In non-computer science terms, that means it’s an advanced program that can be used to identify parts of specifics.
Vision Transformers, or VTs for short, are a type of deep learning model that is used for image recognition tasks. At a high level, a VT works by dividing an input image into a grid of smaller patches and then processing each patch separately using a transformer network. In non-computer terms, a VT works by slicing and dicing an image to identify parts of images.
Where can I learn more?
My article, linked here, discusses the research behind B.R.A.I.N.. While the name B.R.A.I.N. was not yet in use, the research is still the same!
Have additional questions?
Let’s talk.