Machine learning in neuroscience
Machine learning and image processing algorithms have become important tools in brain imaging and computational neuroscience. A large amount of measured neural signals provides a means to decode and characterize task-related brain states and features to distinguish them from non-informative brain signals. There is no doubt that this mechanism contributes to successful and novel biological insights. To prevent the misinterpretation of signals, ideal machine learning techniques should work for any non-expert. The objective is to provide a convenient and simple way to use machine learning to extract features in neuroscience.
Big data on agricultural applications
Big data of telemetry satellite and radar images allow for a direct monitoring of environmental changes and agricultural coverage over a wide range. Such information is an important indicator of the land usage. Crop acreage and distribution have a great impact on the price of agricultural products. Machine/deep learning methods can be applied to the existing agricultural geographic information system (GIS) to analyze of aerial photographs of satellites, airplanes, and unmanned aerial vehicles to develop supervised classification.
Deep learning in medical image reconstruction and processing
Deep learning has been extensively applied to image processing and recognition. It can also be employed in image acquisition, reconstruction, and recovery for biomedical purposes.