search

AIM-Harvard / pyradiomics

Open-source python package for the extraction of Radiomics features from 2D and 3D images and binary masks. Support: https://discourse.slicer.org/c/community/radiomics
http://pyradiomics.readthedocs.io/
BSD 3-Clause "New" or "Revised" License
1.11k stars 485 forks source link

Seeking help for knowing DICOM SEG file #796

Open devo-id opened 1 year ago

devo-id commented 1 year ago

I am new to medical imaging and pyradiomics too. I am working on my project where we have dicom files of patients and we want to calculate radiomics features for the dicom files.

I want to use pyradiomics-dcm script which takes dicom-dir and DICOM SEG file as input : Here I don't know how to create one DICOM SEG file for all dicom series. I wanted help regarding this.

fedorov commented 1 year ago

@devo-id if your input segmentation is not in DICOM SEG, it is not worth the effort to create one. Just use dcm2niix or plastimatch to convert your DICOM image into a format that pyradiomics can accept.

If you do want to learn how to create DICOM SEG, you can look into dcmqi or highdicom for this purpose.

devo-id commented 1 year ago

@fedorov thank you for the reply sir. I got to know about this now.
Actually, I want to create a radiomics feature-based classifier. I am continuously looking at the datasets available on the internet but can't find one. Most of the datasets are restricted or deleted. I tried this https://wiki.cancerimagingarchive.net/display/Public/NSCLC-Radiomics Dataset and tried to extract the features. This dataset has multiple segment files for each dicom series. As I am not aware of medical science, it is very confusing for me to select a segmentation from others. I don't know which mask ensures the validity of the lung tumor.

I am looking for labeled cancer and non-cancer CT scans (having Dicom series as well as the segmentation) dataset. Please help, anyone