Benjamin K
- Research Program Mentor
PhD candidate at Johns Hopkins University
Expertise
Computer science, artificial intelligence, computer vision, medical image analysis, reinforcement learning, simulation, scientific visualization, Python package development
Bio
A PhD Student at Johns Hopkins University, I am interested in intelligent surgical systems based on explorative computer vision and deep reinforcement learning (RL) that directly improve patient outcomes. I have (first/co)-authored peer-reviewed journal articles and conference papers on topics ranging from robotic manipulation with RL to robust 3D feature descriptors for point cloud correspondence. My current projects focus on automating fluoroscopic guidance in percutaneous surgical procedures. Before coming to JHU, I studied at the University of Chicago, where I received a BA in Computer Science with Honors and a Minor in Physics. I have been fortunate enough to intern with industry leaders in AI and health technology, including Intuitive Surgical, Epic Systems, and IBM Research. In my spare time, I enjoy running, cycling, and bouldering, and I have recently taken up learning to paint. Over the years, my hobbies have included archery, sailing, fishing, wrestling, and the cello. I am the author of a modest blog on Medium, in which I reflect on some of these experiences, and am currently pursuing publication for a science fiction novel, which draws on my expertise in AI and health. As a mentor, I love connecting with students over understanding and concepts, as well as digging into the weeds of why some line of code isn't doing what it should be. In my students, I value and try to instill strong Googling skills. This is probably a better indicator of success in computer science than anything else.Project ideas
Identifying Bias in User-facing Machine Learning Models
Machine learning models are ubiquitous in our digital lives, but their presence is not alway obvious. Recently, Twitter users discovered that the automatic cropping used to display preview images was biased toward white, male faces. This is an example of statistical bias arising out of our human bias; because the facial detection algorithm was trained on a dataset of predominantly white, male faces, it predicts similar faces with high confidence. This project challenges students to identify similar instances of bias in interactive digital media resulting from machine learning and AI. Students will perform controlled experiments to understand the extent of these biases and explore their consequences for end users. Potential deliverables range from a blog series to a research paper.
2D/3D Registration of Lego Bricks
Registration is the process of aligning two different measurements of the same object. For example, in computer vision, it is often desirable to stitch two photos together to create a panorama, thereby registering one photo with the overlapping portion of the other. 2D/3D registration is when an existing 3D model, like a CAD surface, is aligned with an image of the same object, either simulated or in the real world. This has applications in medical imaging, such as aligning a 2D X-ray image with a 3D CT scan. In this projects, students will curate a dataset for 2D/3D registration of LEGO bricks and train a keypoint detection algorithm to perform the registration. LEGOs are well suited to this task because 3D models and simulated images are readily available. Minimum deliverables include a dataset hosted on Kaggle and an arXiv paper describing it. The maximum deliverables includes a trained registration algorithm and accompanying paper.