Hello,
As I am in the midst of submitting the first preprint for this project, I wanted to see and take note of other researchers who are already using VOC analysis for disease diagnosis in astronauts. I know there is an existing group exploring VOCs for plant monitoring and an e-nose group at AMEs, so I felt that gauging existing efforts/interest in this field could also be useful.
Our prototype is slated to test on RCCS machine-cultured Regulatory T-cells in the coming weeks, but I am happy to learn about any other efforts in this field.
@AIMLawg
4 Likes
As I am collecting data, I wanted to get advice on which AI models/procedures would be best for getting the best possible result out of my data collection.
As it stands, my data looks like a ~2000 data point table with data organized by time collected, and a range of volatiles compounds changing as time goes on. My idea was to chunk out times along the range of exposure to a “microgravity” environment and prune the logged time for simple categories (30 minute intervals). Would there be a better way to train the AI model to identify the time immune cells are exposed to microgravity, w/o chunking it?
My current plan was to run a relatively lightweight classifier, but I am afraid of losing the “shape” of immune cell response. I aim to have a roughly 10,000 data points if that influences the advice given.
1 Like