Commit 9eb0dfbc authored by Kevin Bronik's avatar Kevin Bronik

Update README.md

parent f3ef2366
# performance_analysis
# Performance Analysis Tool
This software/script calculates all the following important performance metrics:
This software calculates all the following important performance metrics
(can be used to evaluate algorithmic performance against a manual segmentation):
- Dice Similarity Index (SI)
- Sensitivity
- Precision
......@@ -30,22 +31,22 @@ Use one of the following methods to run the program:
```sh
- chmod 755 RunMe.sh
- ./RunMe.sh <lesionmask> <threshold> <manualmask> <foldershortname> <modelname> <distancefunction> <factor/operation> <symmetry> <weight> <p_norm> <saveoutput>
- ./RunMe.sh <lesionmask> <threshold> <manualmask> <foldershortname> <modelname> <distancefunction> <factor/opt> <symmetry> <weight> <p_norm> <saveoutput>
or
- bash RunMe.sh <lesionmask> <threshold> <manualmask> <foldershortname> <modelname> <distancefunction> <factor/operation> <symmetry> <weight> <p_norm> <saveoutput>
- bash RunMe.sh <lesionmask> <threshold> <manualmask> <foldershortname> <modelname> <distancefunction> <factor/opt> <symmetry> <weight> <p_norm> <saveoutput>
```
where
```sh
- <lesionmask>: Segmented lesion mask using lesion prediction algorithm
- <threshold>: Probability threshold that will be applied to output before calculation of the overlap measures
- <manualmask>: The manual segmentation
- <foldershortname>
- <modelname>
- <distancefunction>
- <factor/operation>
- <symmetry>
- <weight>
- <p_norm>
- <saveoutput>
- <foldershortname>: The testing folders short name
- <modelname>: The Model Name that has been used for training data
- <distancefunction>: Euclidean, minkowski, cityblock, cosine, hamming, seuclidean, sqeuclidean, correlation, jaccard, chebyshev, canberra, braycurtis, mahalanobis, yule, matching, dice, kulsinski, rogerstanimoto, russellrao, sokalmichener, sokalsneath, wminkowski
- <factor/opt>: Mean, direct
- <symmetry>: Symmetric, nonsymmetric
- <weight>: Needed for computing the weighted Minkowski distance
- <p_norm>: The order of norm
- <saveoutput>: 0, 1
```
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment