This data is supplementary to article:
Harri Merisaari, Pekka Taimen, Rakesh Shiradkar, Otto Ettala, Marko Pesola, Jani Saunavaara, Peter J.Boström, Anant Madabhushi, Hannu J.Aronen, IvanJambor
Repeatability of radiomics and machine learning for DWI: Short-term repeatability study of 112 patients with prostate cancer.
Magnetic Resonance in Medicide

harri.merisaari@utu.fi

For more in detail description of the features, please see the article and 
Github page https://github.com/haanme/ProstateFeatures.

Glossary:
- ADC/ADCk: Apparent Diffusion Coefficient map of kurtosis function fitted to Diffusion Weighted Imaging data
- K: Kurtosis parameter of kurtosis function
- ROI: Region of Interest
- LS: Lesion ROI
- PM: Prostate mask ROI

Data contents:
 +-scan_1_QC: Quality control file for 1st repetition
 |  +- 1_L1_ADCLS.png : ROI overlaid on ADC parameter map
 |  +- 1_L1_ADCLS_noROI.png : ADC parameter map without ROI
 |  +- 1_L1_ADCPM.png : Prostate mask overlaid on ADC parameter map
 |  +- 1_L1_ADCPM_noROI.png : ADC parameter map without prostate mask
 |  .
 |  \- N_L1_ADCPM_noROI.png
 +-scan_2_QC: Quality control file for 2nd repetition
 +-Features_ADCk_N78_rep1.txt: Radiomic features for most relevant features of 1st repetition of ADCk in training/validation set
 +-Features_ADCk_N78_rep2.txt: Radiomic features for most relevant features of 2nd repetition of ADCk in training/validation set
 +-Features_K_N78_rep1.txt: Radiomic features for most relevant features of 1st repetition of K in training/validation set
 \-Features_K_N78_rep2.txt: Radiomic features for most relevant features of 2nd repetition of K in training/validation set