Skip to content

Add application: Prostate Cancer Risk Predictor #6

@asalic

Description

@asalic

prostate-predictor (Prostate Cancer Risk Predictor)

Description

An application developed during the Open Challenge organized by the CHAIMELEON project for the training of AI models to resolve a specific clinical question related to Prostate cancer: Low vs. High risk patients prediction (in order to avoid unnecessary aggressive procedures in low-grade patients and to provide the most advanced therapy for high-grade).

Usage

jobman submit -i prostate-predictor:latest -r large-gpu -- <INPUT_DIR> <OUTPUT_CSV> <THRESHOLD>

The application accepts three parameters:

  • The input path of a dataset directory (string)
  • The full path of the output including the file name (the resulting file is a CSV) (string)
  • The threshold for the cut-off between the Low and High risk patients (<1.0 real number, e.g. 0.41)

Example:

jobman submit -i prostate-predictor:latest -r large-gpu -- ~/datasets/b875a59a-32f0-4e47-9da6-1e635a85b370/ ~/persistent-home/results_b875a59a-32f0-4e47-9da6-1e635a85b370.csv 0.41

Note the output directory path should be any within the persistent-home, which is shared between all desktops and jobs created by the user, otherwise the results will be lost after the end of the job.

This algorithm is designed to run on GPUs, please specify a resources flavor with GPU when submitting a job via Jobman.

Authors

CHAIMELEON Consortium

Contact info

info.chaimeleon-eu@i3m.upv.es

URL

Available internally on the CHAIMELEON platform.

License

CHAIMELEON Consortium

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions