# Developers

**URL:** https://discourse.canceridc.dev/c/developers/13.md

[Latest](https://discourse.canceridc.dev/latest.md) · [Categories](https://discourse.canceridc.dev/categories.md) · [Tags](https://discourse.canceridc.dev/tags.md)

---

## [About the Developers category](https://discourse.canceridc.dev/t/about-the-developers-category/87)

<div class="topic-metadata">

**Author:** [@fedorov](https://discourse.canceridc.dev/u/fedorov)\
**Replies:** 0

</div>

Open discussions among the developers and maintainers of IDC

---

## [Clinical data in/linked to from IDC](https://discourse.canceridc.dev/t/clinical-data-in-linked-to-from-idc/138)

<div class="topic-metadata">

**Author:** [@dclunie](https://discourse.canceridc.dev/u/dclunie)\
**Replies:** 1\
**Last updated:** [November 15, 2024, 4:16pm UTC](https://discourse.canceridc.dev/t/clinical-data-in-linked-to-from-idc/138 "2024-11-15T16:16:38Z")

</div>

Since the subject of clinical data copies and/or links came up again, I was reminded of this old paper from the TCIA folks: and a related poster: https://www.researchgate.net/publication/325768230\_Ontologies\_Enabling…

---

## [Storing definitions of data collections as DICOM entities](https://discourse.canceridc.dev/t/storing-definitions-of-data-collections-as-dicom-entities/286)

<div class="topic-metadata">

**Author:** [@hmeine](https://discourse.canceridc.dev/u/hmeine)\
**Replies:** 21\
**Last updated:** [May 10, 2024, 6:09pm UTC](https://discourse.canceridc.dev/t/storing-definitions-of-data-collections-as-dicom-entities/286 "2024-05-10T18:09:05Z")

</div>

Hi, I am not directly affiliated with IDC (yet), but I have a discussion topic that should be relevant to this audience and was pointed here by @fedorov: I have been wondering whether and how it makes sense to represent…

---

## [Extracting custom patch sizes for lung CT nodule segmentation demo](https://discourse.canceridc.dev/t/extracting-custom-patch-sizes-for-lung-ct-nodule-segmentation-demo/289)

<div class="topic-metadata">

**Author:** [@adwaykanhere](https://discourse.canceridc.dev/u/adwaykanhere)\
**Replies:** 15\
**Last updated:** [August 15, 2022, 8:55pm UTC](https://discourse.canceridc.dev/t/extracting-custom-patch-sizes-for-lung-ct-nodule-segmentation-demo/289 "2022-08-15T20:55:36Z")

</div>

Hi, I’m following the ai-notebooks tutorial and working through the updated notebook. My question is not related to a technical difficulty but rather a design choice - The paper that is referenced in the notebook ta…

---

## [Manifest file size](https://discourse.canceridc.dev/t/manifest-file-size/98)

<div class="topic-metadata">

**Author:** [@ulrike](https://discourse.canceridc.dev/u/ulrike)\
**Replies:** 2\
**Last updated:** [October 21, 2020, 1:19am UTC](https://discourse.canceridc.dev/t/manifest-file-size/98 "2020-10-21T01:19:04Z")

</div>

I generated a manifest file that contained 65,000 instances. The file size is 33 MB. How big will a manifest file get when users add all collections to the manifest? This test can only be done after the 65,000 instance…

---

## [Manifest size limitation](https://discourse.canceridc.dev/t/manifest-size-limitation/97)

<div class="topic-metadata">

**Author:** [@ulrike](https://discourse.canceridc.dev/u/ulrike)\
**Replies:** 1\
**Last updated:** [October 15, 2020, 6:50pm UTC](https://discourse.canceridc.dev/t/manifest-size-limitation/97 "2020-10-15T18:50:22Z")

</div>

I created a manifest file and wanted to include ISPY1 and QIN\_Headneck collections. The manifest file contained only the ISPY1 instances. Andrey explained that there is a 65,000 instance limit in place. This is too low. …

---

## [Cohort manifest content](https://discourse.canceridc.dev/t/cohort-manifest-content/37)

<div class="topic-metadata">

**Author:** [@fedorov](https://discourse.canceridc.dev/u/fedorov)\
**Replies:** 25\
**Last updated:** [September 9, 2020, 8:17pm UTC](https://discourse.canceridc.dev/t/cohort-manifest-content/37 "2020-09-09T20:17:59Z")

</div>

We currently do not have the “Save cohort”, or a feature to download the manifest corresponding to the selected cohort, but we will need to define what will go into that manifest (or at least the initial version of it) b…