Performance of Machine Learning for Tissue Outcome Prediction in Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

Article Engineering, Biomedical

C(2)MA-Net: Cross-Modal Cross-Attention Network for Acute Ischemic Stroke Lesion Segmentation Based on CT Perfusion Scans

Tianyu Shi et al.

Summary: This study proposes a network with a cross-modal and cross-attention mechanism for segmenting acute ischemic stroke lesions. The network achieves higher recall and F2 scores compared to other state-of-the-art models.

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING (2022)

Add to Collection

Article Clinical Neurology

U-net Models Based on Computed Tomography Perfusion Predict Tissue Outcome in Patients with Different Reperfusion Patterns

Yaode He et al.

Summary: Evaluation of cerebral perfusion is crucial in treatment selection for patients with acute large vessel occlusion (LVO). We developed a deep learning model named U-net, using computed tomography perfusion (CTP) images, to accurately assess ischemic core and tissue at risk. The model outperformed fixed-thresholding methods in estimating infarct core and tissue at risk for patients with different reperfusion patterns, providing individualized prediction of final lesion.

TRANSLATIONAL STROKE RESEARCH (2022)

Add to Collection

Article Computer Science, Interdisciplinary Applications

ISP-Net: Fusing features to predict ischemic stroke infarct core on CT perfusion maps

Haichen Zhu et al.

Summary: This study developed an encoder-decoder based semantic model to predict stroke infarct core after thrombolysis treatment on CT perfusion maps. The method achieved improved accuracy by using a multi-scale atrous convolution block and outperformed several advanced deep learning methods in the evaluation.

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE (2022)

Add to Collection

Review Clinical Neurology

Deep Learning Applications for Acute Stroke Management

Isha R. Chavva et al.

Summary: Brain imaging is crucial for the clinical care of stroke patients, but there are challenges in applying advanced neuroimaging techniques due to time delays, inter-clinician variability, and lack of systematic integration of clinical information. Recent advancements in deep machine learning provide new strategies for using computational medical image analysis to assist in acute stroke decision making. This article examines the current state of using deep learning models in stroke triage and evaluates their ability to automatically select clinical features, reduce inter-rater variability, and integrate neuroimaging with electronic medical record data.

ANNALS OF NEUROLOGY (2022)

Add to Collection

Article Engineering, Biomedical

Toward automated segmentation for acute ischemic stroke using non-contrast computed tomography

Shih-Yen Lin et al.

Summary: This paper proposes a novel model, R2U-RNet, for AIS lesion segmentation using NCCT. The proposed model outperforms other segmentation models and shows potential for improving AIS diagnosis and treatment.

INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY (2022)

Add to Collection

Article Endocrinology & Metabolism

Integrating regional perfusion CT information to improve prediction of infarction after stroke

Julian Klug et al.

Summary: The study showed that integrating perfusion information from adjacent tissues improves the prediction of infarction, with the best performance observed in models using Tmax and multiple perfusion parameters. This approach may help better identify ischemic core and penumbra thresholds, as well as improve patient selection for acute stroke treatment in future studies.

JOURNAL OF CEREBRAL BLOOD FLOW AND METABOLISM (2021)

Add to Collection

Article Biology

Improvement of automatic ischemic stroke lesion segmentation in CT perfusion maps using a learned deep neural network

Mohsen Soltanpour et al.

Summary: Acute ischemic stroke, caused by blood clot blocking brain artery, is a major global cause of death and disability. Current segmentation methods lack precision, but machine learning techniques show promise for improvement. MultiRes U-Net, a deep learning-based technique, presents better results for ischemic stroke lesion segmentation.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Add to Collection

Article Clinical Neurology

Computed Tomography Perfusion-Based Machine Learning Model Better Predicts Follow-Up Infarction in Patients With Acute Ischemic Stroke

Hulin Kuang et al.

Summary: In this study, a threshold-free computed tomography perfusion-based machine learning model was developed to predict follow-up infarction in patients with acute ischemic stroke. The ML model demonstrated a high correlation and small volumetric difference with actual follow-up infarct volume, outperforming current methods. The use of computed tomography perfusion data and time estimates in the ML model resulted in more accurate predictions.

STROKE (2021)

Add to Collection

Article Clinical Neurology

Tissue at Risk and Ischemic Core Estimation Using Deep Learning in Acute Stroke

Y. Yu et al.

Summary: Deep learning models with fine-tuning lead to better performance for predicting tissue at risk and ischemic core, outperforming conventional thresholding methods.

AMERICAN JOURNAL OF NEURORADIOLOGY (2021)

Add to Collection

Review Radiology, Nuclear Medicine & Medical Imaging

Deep learning for automatic target volume segmentation in radiation therapy: a review

Hui Lin et al.

Summary: Deep learning is a fast growing trend in medical imaging, particularly in radiation therapy, showing potential to improve efficiency and reduce variabilities. Automatic segmentation methods have also expanded into target volume segmentation successfully.

QUANTITATIVE IMAGING IN MEDICINE AND SURGERY (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Combining unsupervised and supervised learning for predicting the final stroke lesion

Adriano Pinto et al.

Summary: Predicting the final stroke lesion is crucial for treatment planning, but it is a complex task due to variability in lesion characteristics and cerebral hemodynamics.

MEDICAL IMAGE ANALYSIS (2021)

Add to Collection

Article Clinical Neurology

Predicting Infarct Core From Computed Tomography Perfusion in Acute Ischemia With Machine Learning Lessons From the ISLES Challenge

Arsany Hakim et al.

Summary: The ISLES challenge aims to develop advanced tools for stroke lesion analysis using machine learning. Based on diffusion-weighted imaging, the aim of ISLES-2018 was to segment infarcted tissue on CTP. Twenty-four teams participated in the challenge, with the top team's algorithm outperforming conventional methods, showing improved accuracy in predicting infarcted tissue from CTP.

STROKE (2021)

Add to Collection

Review Radiology, Nuclear Medicine & Medical Imaging

Controversies in Imaging of Patients With Acute Ischemic Stroke: AJR Expert Panel Narrative Review

Eduardo Portela de Oliveira et al.

Summary: The development of reperfusion therapies has significantly impacted stroke care through the use of IV thrombolytic treatment and EVT, with neuroimaging playing a crucial role in patient triage and treatment selection. This article examines the use of multimodality imaging for AIS patients, highlighting controversial topics and providing guidance for clinical practice, while also reviewing major randomized trials supporting EVT. Advanced imaging is expected to further expand patient selection and treatment possibilities as ongoing studies push the boundaries of stroke care.

AMERICAN JOURNAL OF ROENTGENOLOGY (2021)

Add to Collection

Article Biology

MCA-DN: Multi-path convolution leveraged attention deep network for salvageable tissue detection in ischemic stroke from multi-parametric MRI

Anusha Vupputuri et al.

Summary: The study proposed a deep network that combines multi-path convolution with attention mechanisms to assist manual diagnosis decisions in stroke treatment. After being evaluated on 139 patients, the network achieved encouraging success with high accuracy and sensitivity, making it suitable for stroke segmentation and patient-specific treatment planning.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Add to Collection

Article Endocrinology & Metabolism

Tissue outcome prediction in hyperacute ischemic stroke: Comparison of machine learning models

Joseph Benzakoun et al.

Summary: Machine Learning has been utilized for tissue fate prediction after acute ischemic stroke, with Gradient Boosting model showing superior predictive performance compared to other ML models and clinical methods. This suggests promising potential for future decision-making in stroke management.

JOURNAL OF CEREBRAL BLOOD FLOW AND METABOLISM (2021)

Add to Collection

Review Radiology, Nuclear Medicine & Medical Imaging

Deep learning for segmentation in radiation therapy planning: a review

Gihan Samarasinghe et al.

Summary: This paper provides a descriptive review of deep learning techniques for segmentation in radiation therapy planning, highlighting the prevalence of U-net architecture and the use of CT images from both inhouse sources and public datasets. The research in this area is rapidly expanding, emphasizing the importance of consistent metrics and independent validation for benchmarking proposed methods.

JOURNAL OF MEDICAL IMAGING AND RADIATION ONCOLOGY (2021)

Add to Collection

Article Clinical Neurology

Machine Learning-Based Prediction of Brain Tissue Infarction in Patients With Acute Ischemic Stroke Treated With Theophylline as an Add-On to Thrombolytic Therapy: A Randomized Clinical Trial Subgroup Analysis

Boris Modrau et al.

Summary: The study investigated the neuroprotective effect of theophylline in acute ischemic stroke patients undergoing thrombolytic therapy. Using predictive modeling, it was found that there was no significant difference in predicted lesion volumes between patients virtually treated with theophylline and those virtually treated with placebo. This suggests that theophylline did not show a neuroprotective effect as an add-on to thrombolytic therapy, contrary to preclinical stroke models.

FRONTIERS IN NEUROLOGY (2021)

Add to Collection

Article Clinical Neurology

Automated Prediction of Ischemic Brain Tissue Fate from Multiphase Computed Tomographic Angiography in Patients with Acute Ischemic Stroke Using Machine Learning

Wu Qiu et al.

Summary: A machine learning-based mCTA model can accurately predict brain tissue perfusion abnormality and follow-up infarction, comparable to CTP.

JOURNAL OF STROKE (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Computed tomography angiography-based deep learning method for treatment selection and infarct volume prediction in anterior cerebral circulation large vessel occlusion

Lasse Hokkinen et al.

Summary: The study demonstrated that a CTA-based CNN method has moderate correlation with final infarct volumes in patients successfully treated with EVT.

ACTA RADIOLOGICA OPEN (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Evaluation of a CTA-based convolutional neural network for infarct volume prediction in anterior cerebral circulation ischaemic stroke

Lasse Hokkinen et al.

Summary: In this study, a convolutional neural network (CNN) based on CTA was found to provide accurate estimations of infarct core volume as observed in follow-up imaging studies. The CNN-derived infarct volumes showed a strong correlation with CTP-RAPID ischemic core volumes.

EUROPEAN RADIOLOGY EXPERIMENTAL (2021)

Add to Collection

Article Computer Science, Information Systems

Prediction of Stroke Lesion at 90-Day Follow-Up by Fusing Raw DSC-MRI With Parametric Maps Using Deep Learning

Adriano Pinto et al.

Summary: Stroke is the second most common cause of death in developed countries, and rapid clinical assessment and intervention play a crucial role in improving patients' quality of life. Clinical interventions aim to restore perfusion deficits, and a deep learning method can automatically predict ischemic stroke tissue outcome.

IEEE ACCESS (2021)

Add to Collection

Article Neuroimaging

Impact of the reperfusion status for predicting the final stroke infarct using deep learning

Noelie Debs et al.

Summary: The study assessed the impact of integrating reperfusion status into deep learning models for predicting the final infarct in acute ischemic stroke patients, finding that CNN-based models outperformed clinically-used perfusion-diffusion mismatch models in terms of performance. Comparing predicted infarct in cases of successful vs failed reperfusion may aid in estimating treatment effect and guiding therapeutic decisions for selected patients.

NEUROIMAGE-CLINICAL (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Prediction of final infarct volume from native CT perfusion and treatment parameters using deep learning

David Robben et al.

MEDICAL IMAGE ANALYSIS (2020)

Add to Collection

Article Multidisciplinary Sciences

Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features

Malte Grosser et al.

PLOS ONE (2020)

Add to Collection

Editorial Material Radiology, Nuclear Medicine & Medical Imaging

Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers-From the Radiology Editorial Board

David A. Bluemke et al.

RADIOLOGY (2020)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Machine Learning for Detecting Early Infarction in Acute Stroke with Non-Contrast-enhanced CT

Wu Qiu et al.

RADIOLOGY (2020)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Brain SegNet: 3D local refinement network for brain lesion segmentation

Xiaojun Hu et al.

BMC MEDICAL IMAGING (2020)

Add to Collection

Article Medicine, General & Internal

Novel Estimation of Penumbra Zone Based on Infarct Growth Using Machine Learning Techniques in Acute Ischemic Stroke

Yoon-Chul Kim et al.

JOURNAL OF CLINICAL MEDICINE (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

CSNet: A new DeepNet framework for ischemic stroke lesion segmentation

Amish Kumar et al.

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE (2020)

Add to Collection

Article Computer Science, Artificial Intelligence

Automatic ischemic stroke lesion segmentation from computed tomography perfusion images by image synthesis and attention-based deep neural networks

Guotai Wang et al.

MEDICAL IMAGE ANALYSIS (2020)

Add to Collection

Article Multidisciplinary Sciences

Localized prediction of tissue outcome in acute ischemic stroke patients using diffusion- and perfusion-weighted MRI datasets

Malte Grosser et al.

PLOS ONE (2020)

Add to Collection

Article Medicine, General & Internal

Use of Deep Learning to Predict Final Ischemic Stroke Lesions From Initial Magnetic Resonance Imaging

Yannan Yu et al.

JAMA NETWORK OPEN (2020)

Add to Collection

Editorial Material Medicine, General & Internal

Reporting of artificial intelligence prediction models

Gary S. Collins et al.

LANCET (2019)

Add to Collection

Article Mathematical & Computational Biology

Evaluation of Enhanced Learning Techniques for Segmenting Ischaemic Stroke Lesions in Brain Magnetic Resonance Perfusion Images Using a Convolutional Neura Network Scheme

Carlos Uziel Perez Malla et al.

FRONTIERS IN NEUROINFORMATICS (2019)

Add to Collection

Article Clinical Neurology

Artificial Neural Network Computer Tomography Perfusion Prediction of Ischemic Core

Aimen S. Kasasbeh et al.

STROKE (2019)

Add to Collection

Article Multidisciplinary Sciences

Technical considerations of multi-parametric tissue outcome prediction methods in acute ischemic stroke patients

Anthon Jy Winder et al.

SCIENTIFIC REPORTS (2019)

Add to Collection

Article Biology

Acute ischemic stroke lesion core segmentation in CT perfusion images using fully convolutional neural networks

Albert Clerigues et al.

COMPUTERS IN BIOLOGY AND MEDICINE (2019)

Add to Collection

Article Medicine, General & Internal

Thrombectomy for Stroke at 6 to 16 Hours with Selection by Perfusion Imaging

G. W. Albers et al.

NEW ENGLAND JOURNAL OF MEDICINE (2018)

Add to Collection

Article Medicine, General & Internal

Thrombectomy 6 to 24 Hours after Stroke with a Mismatch between Deficit and Infarct

R. G. Nogueira et al.

NEW ENGLAND JOURNAL OF MEDICINE (2018)

Add to Collection

Review Radiology, Nuclear Medicine & Medical Imaging

Current Applications and Future Impact of Machine Learning in Radiology

Garry Choy et al.

RADIOLOGY (2018)

Add to Collection

Article Clinical Neurology

Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning

Anne Nielsen et al.

STROKE (2018)

Add to Collection

Article Clinical Neurology

2018 Guidelines for the Early Management of Patients With Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association

William J. Powers et al.

STROKE (2018)

Add to Collection

Article Clinical Neurology

ISLES 2016 and 2017-Benchmarking Ischemic Stroke Lesion Outcome Prediction Based on Multispectral MRI

Stefan Winzeck et al.

FRONTIERS IN NEUROLOGY (2018)

Add to Collection

Article Medicine, General & Internal

Global, Regional, and Country-Specific Lifetime Risks of Stroke, 1990 and 2016

Valery L. Feigin et al.

NEW ENGLAND JOURNAL OF MEDICINE (2018)

Add to Collection

Article Clinical Neurology

Stroke Lesion Outcome Prediction Based on MRI Imaging Combined With Clinical Information

Adriano Pinto et al.

FRONTIERS IN NEUROLOGY (2018)

Add to Collection

Article Endocrinology & Metabolism

Fully automated stroke tissue estimation using random forest classifiers (FASTER)

Richard McKinley et al.

JOURNAL OF CEREBRAL BLOOD FLOW AND METABOLISM (2017)

Add to Collection

Article Medicine, General & Internal

A Randomized Trial of Intraarterial Treatment for Acute Ischemic Stroke

O. A. Berkhemer et al.

NEW ENGLAND JOURNAL OF MEDICINE (2015)

Add to Collection

Article Health Care Sciences & Services

Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range

Xiang Wan et al.

BMC MEDICAL RESEARCH METHODOLOGY (2014)

Add to Collection

Review Medicine, General & Internal

The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration

Alessandro Liberati et al.

BMJ-BRITISH MEDICAL JOURNAL (2009)

Add to Collection

Article Medicine, General & Internal

Thrombolysis with alteplase 3-4•5 h after acute ischaemic stroke (SITS-ISTR):: an observational study

Nils Wahlgren et al.

LANCET (2008)

Add to Collection

Article Medicine, General & Internal

Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke

Werner Hacke et al.

NEW ENGLAND JOURNAL OF MEDICINE (2008)

Add to Collection

Article Computer Science, Artificial Intelligence

Applying instance-based techniques to prediction of final outcome in acute stroke

C Gottrup et al.

ARTIFICIAL INTELLIGENCE IN MEDICINE (2005)

Add to Collection

Article Medicine, General & Internal

Measuring inconsistency in meta-analyses

JPT Higgins et al.

BMJ-BRITISH MEDICAL JOURNAL (2003)

Add to Collection

Performance of Machine Learning for Tissue Outcome Prediction in Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

Related references

Export Citation

Share Paper