Skip to content

Multi Stage Batch Testing with Dilution Effect Considerations

Notifications You must be signed in to change notification settings

Nyu10/Multi-Stage_Batch_Testing

Repository files navigation

Multi Stage Batch Testing with Dilution Effect Considerations

Link to Comprehensive Slide Presentation https://docs.google.com/presentation/d/1DUTcTp5IZ67zpCfv_3nZZue2eKuF44Oy3TqNYrNUFHU/edit?usp=sharing

Mult Stage Batch Testing without Dilution Effect Considerations

Regular_Batch_Testing.py is the main source code with all the functions and logic to produce results of the multi-stage batch testing simulation. Final BTK.ipynb will create a dataframe to display all the statistics

My results replicate Stony Brook University's 2020 Paper. https://arxiv.org/pdf/2006.16079.pdf

Batch Testing Results

Mult Stage Batch Testing with Dilution Effect Considerations

Dilution Effect Considerations are introduced with this paper (pages 6-8) https://arxiv.org/abs/2008.06642

gamma_eta_n_d.py

Generates dilution effect by creating a function, Gamma, that takes in the batch size, the number of positves in the batch and returns the false negative rate. These results are saved in F_N_d.npy

Dilultion_Effect.py

Uses the false negative rates generated by gamma_eta_n_d.py to improve our multi-stage batch testing simulation results.

About

Multi Stage Batch Testing with Dilution Effect Considerations

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published