nnUNet semi-supervised (trained w/ external public dataset [ProstateX])


Logo for nnUNet semi-supervised (trained w/ external public dataset [ProstateX])

About

Creator:
Contact email:
Image Version:
a4d009a4-e0a3-4421-9141-a0437b12e677
Last updated:
Oct. 15, 2022, 8:05 a.m.
Associated publications:
Cuocolo R, Comelli A, Stefano A, et al.. Deep Learning Whole‐Gland and Zonal Prostate Segmentation on a Public MRI Dataset. Magnetic Resonance Imaging. 2021;54(2):452-459.
Armato SG, Huisman H, Drukker K, et al.. PROSTATEx Challenges for computerized classification of prostate lesions from multiparametric magnetic resonance images. J Med Imag. 2018;5(04):1.
Inputs:
  • Coronal T2 Prostate MRI  (Coronal T2 MRI of the Prostate)
  • Transverse T2 Prostate MRI  (Transverse T2 MRI of the Prostate)
  • Sagittal T2 Prostate MRI  (Sagittal T2 MRI of the Prostate)
  • Transverse HBV Prostate MRI  (Transverse High B-Value Prostate MRI)
  • Transverse ADC Prostate MRI  (Transverse Apparent Diffusion Coefficient Prostate MRI)
  • Clinical Information Prostate MRI  (Clinical information to support clinically significant prostate cancer detection in prostate MRI. Provided information: patient age in years at the time of examination (patient_age), PSA level in ng/mL as reported (PSA_report or PSA), PSA density in ng/mL^2 as reported (PSAD_report), prostate volume as reported (prostate_volume_report), prostate volume derived from automatic whole-gland segmentation (prostate_volume_automatic), scanner manufacturer (scanner_manufacturer), scanner model name (scanner_model_name), diffusion b-value of (calculated) high b-value diffusion map (diffusion_high_bvalue), Malignant Neoplasm Histotype (histology_type), Prostate Imaging-Reporting and Data System (PIRADS), Neural invasion (neural_invasion, yes/no), Vascular invasion (vascular_invasion, yes/no), Lymphatic invasion (lymphatic_invasion, yes/no). Values acquired from radiology reports will be missing, if not reported.)
Outputs:
  • Case-level Cancer Likelihood Prostate MRI  (Case-level likelihood of harboring clinically significant prostate cancer, in range [0,1].)
  • Transverse Cancer Detection Map Prostate MRI  (Single-class, detection map of clinically significant prostate cancer lesions in 3D, where each voxel represents a floating point in range [0,1].)

Challenge Performance

Date Challenge Phase Rank
Oct. 15, 2022 PI-CAI Open Development Phase - Tuning 35

Model Facts

Summary

Additional dataset was used (ProstateX dataset).

Mechanism

Additional dataset was used (ProstateX dataset).

Validation and Performance

Uses and Directions

This algorithm was developed for research purposes only.

Warnings

Common Error Messages

Information on this algorithm has been provided by the Algorithm Editors, following the Model Facts labels guidelines from Sendak, M.P., Gao, M., Brajer, N. et al. Presenting machine learning model information to clinical end users with model facts labels. npj Digit. Med. 3, 41 (2020). 10.1038/s41746-020-0253-3