Sorry, you need to enable JavaScript to visit this website.

In this study, we investigate how different types of masks affect automatic emotion classification in different channels of audio, visual, and multimodal. We train emotion classification models for each modality with the original data without mask and the re-generated data with mask respectively, and investigate how muffled speech and occluded facial expressions change the prediction of emotions.

Ultrasound imaging (US) often suffers from distinct image artifacts from various sources. Classic approaches for solving these problems are usually model-based iterative approaches that have been developed specifically for each type of artifact, which are often computationally intensive. Recently, deep learning approaches have been proposed as computationally efficient and high performance alternatives. Unfortunately, in the current deep learning approaches, a dedicated neural network should be trained with matched training data for each artifact type.

The study of vascular morphometry requires segmenting vessels with high precision. Of particular clinical interest is the morphometric analysis of arterial bifurcations in Adaptive Optics Ophthalmoscopy (AOO) images of eye fundus. In this paper, we extend our previous approach for segmenting retinal vessel branches to the segmentation of bifurcations. This enables us to recover the microvascular tree and extract biomarkers that charactarize the blood flow.

Adaptive subspace detectors (ASD) generalize matched subspace detectors (MSD) by accounting for possible correlation. Both ASD and MSD are derived using the generalized likelihood ratio test (GLRT). While MSD assumes there is no correlation between observations, ASD estimates a sample covariance matrix of possibly correlated samples using signal-free observations. In this paper, we address the performance of the ASD when the number of secondary data is insufficient and the observed signal lies in higher dimensional space.

In this work, we study the problem of tracking multiple frequency components in a noisy signal using a spectrogram-based method. Previous approaches such as image processing based or hidden Markov model-based methods may not be capable of tracking multiple frequency components, may require extensive training, and may be time-consuming. To address these issues, we propose an accurate and efficient method named Adaptive Multi-Trace Carving (AMTC) for tracking multiple frequency traces by iterative forward and backward dynamic programming and adaptive trace compensation.

Monitoring sleep quality and status is important to learn health condition for improvement and prevent sleep apnea. A bed-mounted seismometer system is proposed to monitor the heart and respiratory rates, and body movement and posture, during the sleep. To effectively monitor sleep status, an innovative local maxima statistics based approach and an instantaneous property based method are developed to estimate heart and respiratory rates, respectively. These methods are more robust and stable compared to previous works.

Time-lapse microscopy provides 4D imaging data for monitoring and studying down to single-cell, the stochastic processes involved as bacterial colonies grow and interact under different stress conditions. Two main factors prevent high throughput analysis: a) cell segmentation and tracking are very time-consuming and error-prone and b) analytics tools are lacking to interpret the plethora of features extracted from a complex “cell-movie.” To address both limitations, we have recently developed a multi-resolution Bio-image Analysis & Single-Cell Analytics framework, called BaSCA.