Deep learning and machine learning techniques have been recently revolutionizing not only the scientific community but the entire society. In line with the traditional techniques and algorithms, we have been developing advanced algorithms and methdologies that can lead to an effective and efficient knowledge represenation and decision making. Our recent developments include, not limited to, multi-scale, multi-modal, and multi-task learning methods as well as semi-supervised and unsupervised learning approaches.
Digital & computational pathology is an emerging practice of computerized image processing, analysis, and interpretation of digitized tissue specimen images at microscopic scales. The practice entails 1) preprocessing/normalization of images, 2) quantification of structural/biochemical/functional tissue characteristics, 3) discovery of useful knowledge, 4) decision making/support. We have been working on developing segmentation methods for tissues/cells/nuclei as well as machine learning methods for accurate and robust cancer detection, diagnosis, and prognosis. We are also dedicated to developing computational methods to analyze and understand tumor microstructure and heterogeneity.
Magnetic resonance imaging (MRI) permits non-invasive visualization of suspicious lesions. Digital pathology enables reproducible and quantitative measures of tissue microstructures. These quantitative measures are not only useful in improving cancer diagnosis and prognosis but also provide an index of tumor heterogeneity. We investigate the relationship between MRI and digital pathology analysis in a systematic fashion, including advanced image processing and registration, 3-D printing, machine learning techniques. Our goal is to facilitate the detailed histological and prognostic characterization of cancers on MRI.
Diagnosis of stroke has been primarily dependent on manual interpretation and assessment of several CT/MR images. The manual process is slow and inaccurate, leading to complications and poor outcomes. We develop computational tools that are tailored for brain/stroke imaging. The tools can 1) automatically detect infarcts, generate and interpret collateral imaging for tissue outcome analysis, and 3) predict patients' outcome in regard to treatment options.
Drones have recently gained much attention for surveillance and monitoring purposes. Drones, equipped with high-resolution camera, computer vision, object recognition, and tracking techniques, enable collecting imaging data from a distance or altitude. We develop computerized methods to process and analyze the data/information obtained from drones. Our aim is to develop an agricultural surveillance and monitoring system that 1) detects plant pest and disease at early stage, 2) predicts the spread of the pest and disease.