Crowd-Sourced Deep Learning for Intracranial Hemorrhage Identification: Wisdom of Crowds or Laissez-Faire

Article Computer Science, Interdisciplinary Applications

3D Deep Neural Network Segmentation of Intracerebral Hemorrhage: Development and Validation for Clinical Trials

Matthew F. Sharrock et al.

Summary: Deep neural networks can quickly and accurately estimate the volume of intracranial hemorrhage (ICH) with high agreement, demonstrating better performance when trained with diverse pathology. 3D networks outperformed 2D and random forest models when trained with the appropriate anatomic context.

NEUROINFORMATICS (2021)

添加到收藏夹

Article Geriatrics & Gerontology

Artificial Intelligence Can Effectively Predict Early Hematoma Expansion of Intracerebral Hemorrhage Analyzing Noncontrast Computed Tomography Image

Linyang Teng et al.

Summary: This study developed and validated an artificial intelligence model based on deep learning to predict early hematoma enlargement in patients with intracerebral hemorrhage. Through the analysis and validation of NCCT images, the AI model showed higher specificity and sensitivity in predicting early hematoma enlargement compared to existing methods.

FRONTIERS IN AGING NEUROSCIENCE (2021)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Efficiency of a deep learning-based artificial intelligence diagnostic system in spontaneous intracerebral hemorrhage volume measurement

Tao Wang et al.

Summary: This study evaluated a deep learning-based automated segmentation algorithm for quantifying spontaneous intracerebral hemorrhage, showing its high efficiency and accuracy compared to manual segmentation techniques, and suggesting its promise for precise ICH quantification in practice.

BMC MEDICAL IMAGING (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Code-free deep learning for multi-modality medical image classification

Edward Korot et al.

Summary: The analysis compares the performance of six 'code-free deep learning' platforms in creating medical image classification models and finds that they demonstrate higher classification performance with optical coherence tomography modality. Potential use cases include research dataset curation, mobile 'edge models' for regions without internet access, and baseline models for comparing and iterating bespoke deep learning approaches.

NATURE MACHINE INTELLIGENCE (2021)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging