A Sequence Agnostic Multimodal Preprocessing for Clogged Blood Vessel Detection in Alzheimer's Diagnosis

TL;DR

This paper introduces a novel multimodal preprocessing approach that enhances machine learning-based detection of blood vessel blockages in brain images, aiding Alzheimer's diagnosis.

Contribution

It presents a sequence-agnostic preprocessing scheme combining 3D point cloud extraction and feature fusion, improving vessel classification accuracy.

Findings

Outperforms existing preprocessing methods on Clog Loss dataset.

Achieves higher accuracy in classifying stalled and non-stalled vessels.

Demonstrates the effectiveness of sequence-order invariance in imaging analysis.

Abstract

Successful identification of blood vessel blockage is a crucial step for Alzheimer's disease diagnosis. These blocks can be identified from the spatial and time-depth variable Two-Photon Excitation Microscopy (TPEF) images of the brain blood vessels using machine learning methods. In this study, we propose several preprocessing schemes to improve the performance of these methods. Our method includes 3D-point cloud data extraction from image modality and their feature-space fusion to leverage complementary information inherent in different modalities. We also enforce the learned representation to be sequence-order invariant by utilizing bi-direction dataflow. Experimental results on The Clog Loss dataset show that our proposed method consistently outperforms the state-of-the-art preprocessing methods in stalled and non-stalled vessel classification.