Decision algorithms in acute stroke have continuously evolved over the last two decades to more precisely capture the heterogeneous distribution of this devastating disease. Clinical trials, as well as large scale retrospective studies, have demonstrated safe use of re-canalization therapy for prolonged time windows, refined perfusion–diffusion mismatch, core estimation, risk of hemorrhage, and better estimation of recovery. With the advent of genomics and personalized biomarkers, decision support algorithms in stroke will continue to be improved and increasingly rely on complex computational models.
In recent years, the greater accessibility to large imaging and clinical datasets and the increased computational power has led to an awakening in the use of machine learning algorithms (such as deep learning) in various medical diagnostic areas. These algorithms can efficiently aggregate millions of multi-modal data points to come up with a model that can often provide fast and accurate diagnostic support and predictions.
This Research Topic focuses on recent advances in the applications of pattern recognition and machine learning for clinical decision support in acute ischemic stroke. For this purpose, we solicit high quality, original research articles, or review articles related to advancing the design and clinical applications of feature extraction and/or classification of biomedical data including biomedical signals and images. In particular, we encourage the submission of clinically significant studies that undertake a multidisciplinary approach to detect or predict parameters related to stroke using signal processing and machine learning techniques.
Potential areas of interest include, but are not limited to:
• Infarct core, mismatch, and lesion segmentation in various imaging modalities including X-ray, NMR, CT, PET, MRI.
• Decision support algorithms
• Personalized neurorehabilitation
• Machine Learning of neuroimaging biomarkers
• Predictive modeling of treatment efficacy
• Multi-parametric real-time signal processing for the patients in the intensive care units
• Machine learning and Pattern recognition for telemedicine
• Sequence processing in genomic and proteomic data
• Automatic processing of physiological signals such as TCD, ECG, EMG, EEG, PPG, etc.
Decision algorithms in acute stroke have continuously evolved over the last two decades to more precisely capture the heterogeneous distribution of this devastating disease. Clinical trials, as well as large scale retrospective studies, have demonstrated safe use of re-canalization therapy for prolonged time windows, refined perfusion–diffusion mismatch, core estimation, risk of hemorrhage, and better estimation of recovery. With the advent of genomics and personalized biomarkers, decision support algorithms in stroke will continue to be improved and increasingly rely on complex computational models.
In recent years, the greater accessibility to large imaging and clinical datasets and the increased computational power has led to an awakening in the use of machine learning algorithms (such as deep learning) in various medical diagnostic areas. These algorithms can efficiently aggregate millions of multi-modal data points to come up with a model that can often provide fast and accurate diagnostic support and predictions.
This Research Topic focuses on recent advances in the applications of pattern recognition and machine learning for clinical decision support in acute ischemic stroke. For this purpose, we solicit high quality, original research articles, or review articles related to advancing the design and clinical applications of feature extraction and/or classification of biomedical data including biomedical signals and images. In particular, we encourage the submission of clinically significant studies that undertake a multidisciplinary approach to detect or predict parameters related to stroke using signal processing and machine learning techniques.
Potential areas of interest include, but are not limited to:
• Infarct core, mismatch, and lesion segmentation in various imaging modalities including X-ray, NMR, CT, PET, MRI.
• Decision support algorithms
• Personalized neurorehabilitation
• Machine Learning of neuroimaging biomarkers
• Predictive modeling of treatment efficacy
• Multi-parametric real-time signal processing for the patients in the intensive care units
• Machine learning and Pattern recognition for telemedicine
• Sequence processing in genomic and proteomic data
• Automatic processing of physiological signals such as TCD, ECG, EMG, EEG, PPG, etc.