BAAD (Brain Anatomical Analysis using Diffeomorphic deformation)

BAAD (Brain Anatomical Analysis using Diffeomorphic deformation) is a software developed by research institute in Shiga University of Medical Science ^[1] to support diagnosis of various diseases from brain magnetic resonance images (MRI) using artificial intelligence (AI) technology composed of the voxel-based morphometry (VBM) and machine learning technologies. It is introduced as one of extensions of SPM, and is freely available. VBM is a neuroimaging technique that investigates focal differences in brain anatomy. It uses statistical methods to calculate the local brain volume or concentration, atrophy or hypertrophy, in voxel units from MRI. In Alzheimer's disease (AD), brain atrophy appears from the medial temporal lobe, including the hippocampus, to the temporoparietal cortex, and the BAAD-AI evaluates the atrophy pattern through multiple regions of interests (ROIs) and expresses the probability of an AD diagnosis as an AD likelihood score (ADLS). ADLS represents the distance to the hyperplane and is obtained from the posterior probability function, , the probability is the class is the class given that the input variable is . The probability is transformed by a sigmoid function to compress the value within the range of [0, 1]; the larger the value, the more likely is the diagnosis of AD.

In 200 MR images extracted from the North American ADNI database, BAAD outperformed radiologists for diagnosis^[2]. The machine learning was conducted using the North American Alzheimer's Disease Neuroimaging Initiative (ADNI) database, and the usefulness of the BAAD-AI in a Japanese population has also been reported.