HeroofAge

About

Creators:

NataliaAlves

briandzt

Contact email:

ztdeng@ucla.edu

Image Version:

24d19846-6992-401f-8180-5be62fab7089

Last updated:

Dec. 16, 2024, 6:08 p.m.

Interfaces

This algorithm implements all of the following input-output combinations:

Inputs Outputs

	Inputs	Outputs
1	Venous phase CT scan Description A contrast-enhanced CT scan in the venous phase and axial reconstruction Kind Image Read from `/input/images/venous-ct/<uuid>.mha` or `/input/images/venous-ct/<uuid>.tif` Venous phase CT scan Clinical Information Pancreatic CT Description Clinical information in JSON format, includes "age", "sex", "study_date" and "scanner". In formats, respectively, "042Y", "M", "YYYY-MM-DD HH:MM:SS", and "TOSHIBA". All values can be null. Kind Anything Read from `/input/clinical-information-pancreatic-ct.json` Example { "key": "value", "None": null } Clinical Information Pancreatic CT	PDAC Likelihood Description This indicates the likelihood of pancreatic ductal adenocarcinoma (PDAC) occurring, presented as a value between 0.0 and 1.0. Kind Float Write to `/output/pdac-likelihood.json` Example 42.0 PDAC Likelihood PDAC Detection Map Description Single-class, detection map of pancreatic ductal adenocarcinoma (PDAC) in 3D, where each voxel represents a floating point in range [0,1] Kind Heat Map Write to `/output/images/pdac-detection-map/<uuid>.mha` or `/output/images/pdac-detection-map/<uuid>.tif` PDAC Detection Map

Challenge Performance

Date	Challenge	Phase	Rank
Oct. 28, 2024	PANORAMA	Development Phase - Sanity Check	1
Jan. 23, 2025	PANORAMA	Development Phase - Tuning	35
Feb. 25, 2025	PANORAMA	Testing Phase	2

Model Facts

Summary

This algorithm is designed to perform PDAC lesion detection on contrast-enhanced CT(CECT) scans. The algorithm requires abdominal CECT image volume in Portal Venous phase and outputs: 1. A json file with the patient-level likelihood of lesion presence. 2.A .mha file containing the segmentation of detected lesions with each lesion volume represented by a probability value. This algorithm is based on nnUNet and utilizes pyradiomics for radiomics feature extraction.

Mechanism

Algorithm: The algorithm is based on two nnUNet model working at different resolutions. The first nnUNet extracts the ROI using the subsampled low resolution images. The LR ROI is then used to crop the original HR images. In addition, 4 voxel-based radiomics feature maps are generated by pyradiomics using the high resolution images and the LR ROI segmentation. The second nnUNet then performs PDAC detection using both the HR images and the matched radiomics feature maps.

Input:

Datatype: Abdominal CECT volume in Portal Venous phase.
File format: 4D DICOM or NIFTI image
Target group: Patients older than 18 and younger than 99

Output:

A json file containing the likelihood of patient having PDAC in percentage
A mha file containing the segmentation of the detected PDAC lesion, with each lesion mask represented by the maximum voxel predicted probability within each mask

Validation and Performance

	AUROC	AP
Validation	96.7%	67.1%
Test	92.4%	63.5%

Uses and Directions

This algorithm was developed for research purposes only.

Warnings

Left empty by the Algorithm Editors

Common Error Messages

Left empty by the Algorithm Editors

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