I-DISC launched the I-DISC Fellows Program at the beginning of 2021. I-DISC Fellows are advanced doctoral students with expertise in data science, computation, and machine learning. From bugs to stars and healthcare to linguistics, I-DISC Fellows have worked on a diverse range of projects with faculty from across the Lehigh community.
Project: What Does it Take to Love a Bug? The Role of Causal Knowledge in Caring
PI: Barbara Malt (Psychology)
This project conducts sentiment analysis on terms related to insects from public media such as newspapers, books, and films. The Fellow provided assistance with sentiment analysis software, wrote a script to scrape data from publicly available sources online, and helped set up the management of the data collected.
Project: International Trade Flows
PI: Mary Anne Madeira (International Relations)
This project required the collection and analysis of a very large amount of international trade data from the World Bank. The Fellow wrote code to download and post-process this data, and also helped train the PI on the relevant data-science concepts and implementation.
Project: Development of a Cloud-based Software for Hemolysis Prediction
PI: Yaling Liu (Mechanical Engineering and Mechanics / BioEngineering)
This project develops computational software for hemolysis evaluation in medical devices. The Fellow proposed ways to speed up the inference stage of the machine learning algorithm, proposed a virtual environment to make the hardware setup more flexible, and helped design the logic for image segmentation.
(Image credit: Y. Liu)
Project: Machine Learning for Automated Radiographic Scoring
PI: Hannah Dailey (Mechanical Engineering)
This project uses machine learning to score radiographic images—for example, to identify bone fractures in X-ray images. The Fellow helped to assess the feasibility of the use of deep learning models for such scoring. X-rays (as depicted in the image on the right) are commonly used to examine the healing progress of a broken tibia (shinbone) and fibula. Machine learning techniques are a promising tool for automating the evaluation of medical images like these.
(Image Credit: H. Daley)
Project: Social, Demographic, and Environmental Influence on Ebola Spillover
PI: Paolo Bocchini (Civil and Environmental Engineering)
This project uses social, demographic, and behavioral data to try to estimate a person's likelihood to engage in habits and practices that may lead to Ebola spillover. The PI and his team created preliminary regression and machine learning models, and the Fellow investigated the possibility of applying more sophisticated data-driven models.
(Image Credit: Paolo Bocchini)
Here are some other examples of the diverse range of projects our I-DISC Fellows have worked on in 2021:
● Use ML to classify variable stars in astronomical surveys (Physics)
● Scrape social media posts to identify consumer sentiment (Marketing)
● Create a searchable database of children’s spontaneous stories (Psychology)
● Write Python function to extract text from SEC forms (Accounting)