• Augmented imaging data by illumination, rotation angle and scale. Extracted and imported data into deep learning platform. • Implemented deep convolutional neural network for detecting sinus, atrial, ventricular arrhythmia on 12-channel ECG data.• Participated Kaggle Tensorflow speech recognition challenge, 167 of 1315 (team AAA).• Performed cells segmentation of Neutrophils and lymphocyte by applying deep segmentation neural network.• Applied Deep Residual Network for retinal… Show more • Augmented imaging data by illumination, rotation angle and scale. Extracted and imported data into deep learning platform. • Implemented deep convolutional neural network for detecting sinus, atrial, ventricular arrhythmia on 12-channel ECG data.• Participated Kaggle Tensorflow speech recognition challenge, 167 of 1315 (team AAA).• Performed cells segmentation of Neutrophils and lymphocyte by applying deep segmentation neural network.• Applied Deep Residual Network for retinal diseases classification (Hole macular, Cataract, pre-retinal membrane, Neuroepithelial detachment, cystoid macular edema). • Performed image segmentation on neck MRI. 3D reconstructed carotid artery with plaque (http://www.aaalgo.com/demos/plaque/).• Detected myeloma lesion on MRI, CT/PET images. Show less

• Developed image processing algorithms using MATLAB for colorectal cancer segmentation.• Conducted hypothesis testing of the imaging performance using R and MS Excel.• Validated clinical application of a multimodal endoscope with fluorescent peptide in detection of colorectal neoplasia with diagnostic sensitivity of 89% and specificity of 92% at T/B ratio of 1.16