Research Questions: Given the lung CT scans, demographics, and FVC of a patient at the current time, can we predict the prognosis of Idiopathic pulmonary fibrosis (IPF) in the future? Can we formulate an intuitive metric for the prognosis that the multi-modal model will predict instead of noisy time-series modeling?
Research Questions: Can pre-trained DNNs be good feature extractors for downstream biomedical tasks? How can we make biomedical segmentation more robust by using regularization?
Research Questions: Can we extend the Random Subspace Method with tunable variable feature length in a neural-network learning setup? How can we improve baseline approaches by utilizing RSE-Net for distinguishing motor imagery EEG signals?
Research Questions: Can we train a rotation-invariant convolutional neural network by designing a multi-head model? How can we formulate a robust learning scheme by proposing a novel augmentation algorithm that scales to emulate a large set of image degradation and occlusion policies?
Research Questions: Can we train a rotation-invariant convolutional neural network by designing a multi-head model? How can we formulate a robust learning scheme by proposing a novel augmentation algorithm that scales to emulate a large set of image degradation and occlusion policies?
Research Questions: Can we propose a conditional variational autoencoder (CVAE) to generate anomalous ECG signals as an augmentation tool? To minimize the number of parameters in the network, multiple dot layers are designed. These dot layers can be considered as non-learnable similarity gates, which can mix features from two branches in the network.