About Me

👋 Hello! I’m Zhongbo Yan, a passionate software engineer with interest on cloud computing, machine learning, algorithms, information security, full-stack development, visual designs, and explaining cool stuffs.

I’m currently pursuing my M.S. in Mobile and IoT Engineering at Carnegie Mellon University. I hold a B.S. in Computing from Macao Polytechnic University, where I focused on enterprise information systems.

I have had the privilege of working as a full-stack SDE Intern in Suzhou and Macao. Additionally, I have contributed to academic research, most notably in the field of adversarial data augmentation on knee MRI datasets. My work led to a noticeable improvement in the generalizability of AlexNet on medical MRI datasets, surpassing state-of-the-art models from Stanford University. These findings were published in the IEEE 14th ICSESS.

Enhancing Classification Performance in Knee Magnetic Resonance Imaging Using Adversarial Data Augmentation
The utilization of adversarial data augmentation has demonstrated the potential capability to enhance the classification performance in training deep neural networks to perform computer vision tasks. In this paper, we investigate the effectiveness of this approach as a strategy for enlarging a knee Magnetic Resonance Imaging (MRI) dataset by adding adversarial perturbation. Specifically, we use the Fast Gradient Sign Method (FGSM) to perturb a subset of the training dataset as extra training images to re-train a baseline model that was trained under the same configuration as the top-ranked model on the MRNet leaderboard. Particularly, unlike most of the current work, we investigate the impact of two hyperparameters (attack magnitude and the proportion of data to be re-trained) on the performance of area under the ROC curve (AUC), accuracy, sensitivity, and specificity. Additionally, our results show that adversarial data augmentation can further improve the well-trained baseline model’s AUC by 0.26%, as well as provide a slight improvement in specificity at the same classification threshold. These findings underscore the potential advantages of adversarial data augmentation as a technique for optimizing the decision boundaries of deep learning models. The code of this work will be available on GitHub after the paper is published.

Feel free to reach out to me through my GitHub and LinkedIn. You can also email me at me at aspires dot cc.

Tech-stacks I used are:


Thank you for visiting my blog! I look forward to connecting with like-minded individuals and exploring new opportunities in the field of software engineering and beyond.