AI for Kidney Cancer Prognosis
Unmute
Mentor: Keith Callenberg, Associate Director at Thrive Earlier Detection Corp.
Team members: Poojitha Arangam, Kris Knapp, Jonathan Quintero Ramos
Environment/Techstack: CLAM, PyTorch, SHAM, AWS S3, Scikit-Learn, Jupyter Notebook/Google Colab
This open-source software project focused on computer vision-related machine learning (ML) technologies to predict a forecast of renal cancer from whole slide images.
Team members: Poojitha Arangam, Kris Knapp, Jonathan Quintero Ramos
Environment/Techstack: CLAM, PyTorch, SHAM, AWS S3, Scikit-Learn, Jupyter Notebook/Google Colab
This open-source software project focused on computer vision-related machine learning (ML) technologies to predict a forecast of renal cancer from whole slide images.
How much experience does your group have? Does the project use anything (art, music, starter kits) you didn't create?
CodeDay Labs team in the advanced track.
What challenges did you encounter?
Some of the challenges found in the project dealt with the size of the data being used. As the total data size was too massive to comfortably store locally. It was also formatted in a very particular way and form giving a lot preprocessing overhead by way of tiling.
We also had a big learning curve as we used new libraries none of us had used before. This meant that there was a bit of trial and error that slowed us down along the way. For example, figuring out when the model was over fit was a challenge.
We also had a big learning curve as we used new libraries none of us had used before. This meant that there was a bit of trial and error that slowed us down along the way. For example, figuring out when the model was over fit was a challenge.
Made with 💖 by robots running our open-source software and GraphQL APIs.