TriFormer: A Multi-modal Transformer Framework For Mild Cognitive Impairment Conversion Prediction

Article Biochemical Research Methods

A transformer-based multi-features fusion model for prediction of conversion in mild cognitive impairment

Guowei Zheng et al.

Summary: The study found that the deep learning model showed promising performance in predicting MCI conversion, and specific brain regions and cortical features contributed significantly to the classification task.

METHODS (2022)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

M3T: three-dimensional Medical image classifier using Multi-plane and Multi-slice Transformer

Jinseong Jang et al.

Summary: This study proposes a three-dimensional Medical image classifier called M3T network that uses Multi-plane and Multi-slice Transformer to classify Alzheimer's disease in 3D MRI images. The network combines 3D CNN, 2D CNN, and Transformer to accurately classify AD. The CNN networks with inductive bias efficiently extract the locality information for AD-related abnormalities in the local brain, while the Transformer network learns the abnormalities distributed over a wider region in the brain without inductive bias. The proposed M3T achieved the highest performance in multi-institutional validation databases, demonstrating the feasibility of combining CNN and Transformer for 3D medical images.

2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022) (2022)

Add to Collection

Article Geriatrics & Gerontology

Mild Cognitive Impairment: the Manchester consensus

Ross A. Dunne et al.

Summary: MCI is a clinically heterogeneous syndrome that can lead to further cognitive decline, but may also remain stable or resolve over time. Diagnostic criteria, including cognitive testing, neuroimaging, and fluid biomarkers, can improve accuracy of diagnosis and potentially guide prognosis. Large-scale studies are needed to determine the practical usefulness of combining multiple markers for clinical use.

AGE AND AGEING (2021)

Add to Collection

Article Neurosciences

MRI-based Alzheimer's disease prediction via distilling the knowledge in multi-modal data

Hao Guan et al.

Summary: The study proposes a multi-modal multi-instance distillation scheme for MCI conversion prediction using MRI, achieving superior results compared to existing methods.

NEUROIMAGE (2021)

Add to Collection

Article Clinical Neurology