Identifying Suitable Anatomical Planes in 3D MRI for AD Classification by Employing ViT Pipeline

This work aims to identify the suitable anatomical planes in 3D MRI for discriminating different types of Alzheimer's disease (AD) - cognitively normal (CN), mild cognitive impairment (MCI), and final stage Alzheimer's (AD). Such classification tasks in deep learning (DL) framework displayed noticeable results. Existing works consider global feature selection-based techniques in convolutional neural networks (CNN) and most of them consider all 3D images as input, which creates insignificant use of data that may not be suitable. Therefore, this work aims to identify suitable images from different anatomical planes of 3D MRI and emphasize on local-level features by involving the self-attention mechanism in Vision Transformer (ViT). For validation, MRI data from benchmark ADNI database has been taken into account. Aiming to find suitable MRI scans, 10–12 slices from each of three different anatomical planes - axial <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathrm{A}_{\mathrm{x}})$</tex>, coronal <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathrm{C}_{\mathrm{r}})$</tex>, and sagittal <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathrm{S}_{\mathrm{g}})$</tex> are segregated. Further, the images are fed into ViT by framing three cases (a) individual planes <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{A}_{\mathrm{x}}, \mathrm{C}_{\mathrm{r}},\mathrm{S}_{\mathrm{g}}, (\mathrm{b})$</tex> planes in pair <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\left(\mathbf{A}_{x_{-}} \mathbf{C}_{\mathbf{r}}\right),\left(\mathbf{A}_{\mathbf{x}_{-}} \mathbf{S}_{\mathrm{g}}\right)$</tex>, and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathrm{C}_{\mathrm{r}_{-}}\mathrm{S}_{\mathrm{g}})$</tex> and (c) all planes together <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\left(\mathbf{A}_{\mathbf{x}_{-}} \mathbf{C}_{\mathrm{r}_{-}} \mathrm{S}_{\mathrm{g}}\right)$</tex>. The analysis has been performed by measuring accuracy <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(A_{c})$</tex>, sensitivity <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(S_{e})$</tex>, specificity <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(S_{p})$</tex>, and Fl-score for a different set of inputs; while CNN model as a baseline. The results show that the ViT outperforms the CNN model, while the input pair <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{A}_{\mathrm{x}-}\mathrm{C}_{\mathrm{I}}$</tex> achieved accuracy as high as 93.3%. Thus, the analysis shows that choosing pair of planes could be suitable for MRI-based AD while involving DL-based techniques.