Machine learning has witnessed substantial developments recently and is proving to be crucial for Artificial Intelligence (AI). Deep neural networks are considered to be state-of-the-art machine learning models to be used in various topics for computer vision, image processing, and biomedical sensors. It has also become a popular culture and gained tremendous attention from industry and academia. Healthcare and biomedical sciences have become data-intensive fields that require sophisticated data mining techniques to extract knowledge from accessible data. With the recent developments in machine learning, it gives potential insights that AI techniques can extract from complex medical data.
Although the current study in medical imaging technology, medical data analysis, biomedical sensor data processing, medical diagnostics, and the healthcare sector have produced highly supportive results, there is still a need to investigate novel feature selection approaches to increase predictive performance as well as interpretation and to investigate large-scale data. In large-scale AI algorithms and medical imaging data sets, image analysis and model training processes are critical and can result in high dimensionality, class imbalance, and small sample size. Using Deep Transfer Learning, an approach in deep learning where knowledge is transferred from one model to another can overcome these data analysis issues. Creating such innovative deep transfer learning algorithms which are customized to medical data is a challenging task but has potential and can be considered as the need for the present.
This Research Topic provides an excellent platform for process mining and informatics strategies in smart healthcare systems. We encourage contributions on both theoretical and practical aspects of the application of IoT in biomedical sensors.
Potential topics include but are not limited to the following:
• Analysis of biomedical signals and images
• Big data analytics for healthcare sensors in use
• Data-driven simulation and optimization of healthcare processes
• Real-time decision-making, prognosis, and diagnosis based on biomedical sensors
• Statistical analysis, machine learning, and deep learning for biomedical signals
• Deep transfer learning for medical image analysis
• Machine learning methods applied to biomedical data
• Deep transfer networks for biomedical data augmentation and processing
• Deep transfer learning-based multi-modal computing for medical imaging
• Deep learning-based features extraction for medical images
• Visualization of deep learning features for medical images
Machine learning has witnessed substantial developments recently and is proving to be crucial for Artificial Intelligence (AI). Deep neural networks are considered to be state-of-the-art machine learning models to be used in various topics for computer vision, image processing, and biomedical sensors. It has also become a popular culture and gained tremendous attention from industry and academia. Healthcare and biomedical sciences have become data-intensive fields that require sophisticated data mining techniques to extract knowledge from accessible data. With the recent developments in machine learning, it gives potential insights that AI techniques can extract from complex medical data.
Although the current study in medical imaging technology, medical data analysis, biomedical sensor data processing, medical diagnostics, and the healthcare sector have produced highly supportive results, there is still a need to investigate novel feature selection approaches to increase predictive performance as well as interpretation and to investigate large-scale data. In large-scale AI algorithms and medical imaging data sets, image analysis and model training processes are critical and can result in high dimensionality, class imbalance, and small sample size. Using Deep Transfer Learning, an approach in deep learning where knowledge is transferred from one model to another can overcome these data analysis issues. Creating such innovative deep transfer learning algorithms which are customized to medical data is a challenging task but has potential and can be considered as the need for the present.
This Research Topic provides an excellent platform for process mining and informatics strategies in smart healthcare systems. We encourage contributions on both theoretical and practical aspects of the application of IoT in biomedical sensors.
Potential topics include but are not limited to the following:
• Analysis of biomedical signals and images
• Big data analytics for healthcare sensors in use
• Data-driven simulation and optimization of healthcare processes
• Real-time decision-making, prognosis, and diagnosis based on biomedical sensors
• Statistical analysis, machine learning, and deep learning for biomedical signals
• Deep transfer learning for medical image analysis
• Machine learning methods applied to biomedical data
• Deep transfer networks for biomedical data augmentation and processing
• Deep transfer learning-based multi-modal computing for medical imaging
• Deep learning-based features extraction for medical images
• Visualization of deep learning features for medical images