Predicting Clinical Outcome in Acute Ischemic Stroke Using Parallel Multi-Parametric Feature Embedded Siamese Network

Abstract

Stroke is the second leading cause of death and disability worldwide, with ischemic stroke as the most common type. The preferred diagnostic procedure at the acute stage is the acquisition of multi-parametric magnetic resonance imaging (MRI). This type of imaging not only detects and locates the stroke lesion, but also provides the blood flow dynamics that helps clinicians in assessing the risks and benefits of reperfusion therapies. However, evaluating the outcome of these risky therapies beforehand is a complicated task due to the variability of lesion location, size, shape, and cerebral hemodynamics involved. Though the fully automated model for predicting treatment outcomes using multi-parametric imaging would be highly valuable in clinical settings, MRI datasets acquired at the acute stage are mostly scarce and suffer high class imbalance. In this paper, parallel multi-parametric feature embedded siamese network (PMFE-SN) is proposed that can learn with few samples and can handle skewness in multi-parametric MRI data. Moreover, five suitable evaluation metrics that are insensitive to imbalance are defined for this problem. The results show that PMFE-SN not only outperforms other state-of-the-art techniques in all these metrics but also can predict the class with a small number of samples, as well as the class with high number of samples. An accuracy of 0.67 on leave one cross out testing has been achieved with only two samples (minority class) for training and accuracy of 0.61 with the highest number of samples (majority class). In comparison, state-of-the-art using hand crafted features has 0 accuracy for minority class and 0.33 accuracy for majority class.

A Web-Based Dynamic Nomogram to Predict the Risk of Methicillin-Resistant Staphylococcal Infection in Patients with Pneumonia

“…The model was made available as a web-based interactive tool to ease clinical use and enhance its accessibility. The number of MRS cases in our research was sufficient to build a model that responded to at least 200 observations [ 28 ] and complied with the rules of thumb (1:10) [ 29 ] or Claudia Beleites (5–25 independent samples per class) [ 30 ] and other predictive research [ 31 ]. Furthermore, the quality of the model is measured based on two factors, including accuracy and precision of estimation.…”

A comprehensive review for artificial intelligence on neuroimaging in rehabilitation of ischemic stroke

“…The overall AUC value increased from 0.50 for the random forest method to 0.81. The research conclusion explicitly stated that PMFE-SN demonstrated excellent performance in predicting categories with both few and numerous samples ( 84 ).…”

Improving Ischemic Stroke Care With MRI and Deep Learning Artificial Intelligence

“…[52][53][54][55][56] However, images may provide more information such as the spatial location of infarct and hemorrhage, and presence of brain atrophy. Osama et al 57 proposed a parallel multiparametric feature-embedded Siamese neural network to classify 3month mRS from 0 to 4 using the MRI perfusion maps and clinical data from the ISLES 2017 challenge. The Siamese neural network 58 is often applied in human face recognition and handwriting recognition, by identifying if the input images pairs belong to the same subject.…”