TranSOP: Transformer-based Multimodal Classification for Stroke Treatment Outcome Prediction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Samak, Zeynel A., Clatworthy, Philip, Mirmehdi, Majid
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910729059172352
author Samak, Zeynel A.
Clatworthy, Philip
Mirmehdi, Majid
author_facet Samak, Zeynel A.
Clatworthy, Philip
Mirmehdi, Majid
contents Acute ischaemic stroke, caused by an interruption in blood flow to brain tissue, is a leading cause of disability and mortality worldwide. The selection of patients for the most optimal ischaemic stroke treatment is a crucial step for a successful outcome, as the effect of treatment highly depends on the time to treatment. We propose a transformer-based multimodal network (TranSOP) for a classification approach that employs clinical metadata and imaging information, acquired on hospital admission, to predict the functional outcome of stroke treatment based on the modified Rankin Scale (mRS). This includes a fusion module to efficiently combine 3D non-contrast computed tomography (NCCT) features and clinical information. In comparative experiments using unimodal and multimodal data on the MRCLEAN dataset, we achieve a state-of-the-art AUC score of 0.85.
format Preprint
id arxiv_https___arxiv_org_abs_2301_10829
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TranSOP: Transformer-based Multimodal Classification for Stroke Treatment Outcome Prediction
Samak, Zeynel A.
Clatworthy, Philip
Mirmehdi, Majid
Image and Video Processing
Computer Vision and Pattern Recognition
Acute ischaemic stroke, caused by an interruption in blood flow to brain tissue, is a leading cause of disability and mortality worldwide. The selection of patients for the most optimal ischaemic stroke treatment is a crucial step for a successful outcome, as the effect of treatment highly depends on the time to treatment. We propose a transformer-based multimodal network (TranSOP) for a classification approach that employs clinical metadata and imaging information, acquired on hospital admission, to predict the functional outcome of stroke treatment based on the modified Rankin Scale (mRS). This includes a fusion module to efficiently combine 3D non-contrast computed tomography (NCCT) features and clinical information. In comparative experiments using unimodal and multimodal data on the MRCLEAN dataset, we achieve a state-of-the-art AUC score of 0.85.
title TranSOP: Transformer-based Multimodal Classification for Stroke Treatment Outcome Prediction
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2301.10829