MorphoCloud On-Demand Instances: Run 3D Slicer and SlicerMorph inside a web browser.

Powerful virtual machines (compute instances) that provide interactive remote desktop interface with 3D Slicer and SlicerMorph and its related extensions preloaded for 3D digital morphology and morphometrics data processing.

In a nutshell, you:

1. Have a research and/or teaching focus centered on 3D biological structure (e.g., 3D morphometrics, or visualization);
2. Need occasional access to powerful computers with large amounts of RAM and GPU (up to 40GB GPU)
3. You have a github account;
4. You have an ORCID with a public profile.

NOTE: MorphoCloud and JetStream2 are designed for open science and public data; it is NOT compliant with high-security federal privacy regulations such as HIPAA or FERPA. Consequently, users are strictly prohibited from uploading, processing, or storing any data containing Personally Identifiable Information or Protected Health Information. See the detailed acceptable use policies here

Available Instance Types

Flavor	RAM	Cores	GPU	Storage	Use Cases
g3.l*	60GB	16	A100 (20GB)	100GB	General purpose morphology and morphometrics
g4.xl	125GB	32	L40S	100GB	Photogrammetry, NNInteractive, AI applications
m3.x	250GB	64	None	100GB	Computationally intense tasks that don’t require GPU; eg., Image registration with ANTsPy
r3.l	500GB	64	None	100GB	Image registration with larger dataset
r3.xl	1000GB	128	None	100GB	Image registration with larger dataset

*g3.l is the default instance and is usually more available than other instance flavors.

Click to see the approximate (delayed 5 minutes) count of available resources on JS2

Installed Software

Software/Tool	Description
3D Slicer	v5.10
SlicerMorph	ImageStacks, GPA, ALPACA, and other tools
DeCA	Morphometrics via dense correspondence analysis
Photogrammetry	Generate textured 3D models from photographs
MorphoDepot	Collaborative segmentation and data sharing
MEMOs	AI based organ segmentation for E15 mouse embryos
NNInteractive	AI assisted interactive segmentation
PyTorch	GPU accelerated tensor library for AI tools
R/Rstudio	Provided by JetStream2
Python3	Provided by JetStream2

Ready to give a try?

• Individual users: Go to https://github.com/MorphoCloud/MorphoCloudInstances/issues/new?template=instance-request.yml, fill in the necessary information and choose your instance flavor. Your request will be approved within 24h (often faster). Then, you can create and start using your instance.

• Workshop or Short-Course organizers: If you are planning an event (5 days or less) in which participants would need access to MorphoCloud throughout the event, you can fill in the Workshop Request. Please do that at least 2 weeks prior to your planned event. We will get back to you with the necessary instructions on how to setup your environment.

MorphoCloud Commands

Once your request is approved you can use these commands to manage your instance.

Command	Description	Who can run
/create	Create the instance (and the storage volume if not present already).	Issue creator, Admin
/shelve	Shelve (turn off) a running instance.	Issue creator, Admin
/unshelve	Unshelve (turn on) a stopped instance.	Issue creator, Admin
/delete_instance	Delete the instance (e.g. when it is unresponsive, then you can use the /create to recreate the instance afresh.	Issue creator, Admin
/delete_volume	Delete the storage volume. (Beware you will loose all the existing data)	Issue creator, Admin
/encode_email	Update issue description obfuscating emails.	Issue creator, Admin
/decode_email	Update issue description deobfuscating emails.	Issue creator, Admin
/email	Resend email to Issue creator with connection URL (only possible if instance is active)	Issue creator, Admin
/renew	Extend the instance lifespan one-time.	Issue creator, Admin

MorphoCloud Desktop Interface

• A: Side toolbar that gets activated by pressing the CTRL (or CMD) + ALT + SHIFT keys. It allows copy/paste into the remote session, browse and download files on the remote drive and adjust screen zoom levels (cut from the screenshot).

• B: Shortcuts to commonly used applications and to MyData storage volume.

• C: Displays list of available applications (searchable)

• D: Right mouse clicking anywhere on desktop brings this menu, including changing screen resolution (Display settings).

• E: Click on this icon anytime to extend your session for additional 4 hours.

MorphoCloud Instances FAQ for common issues

• After I login to the instance, only a black background and the mouse pointer is visible. This is due to a known race condition. Connect to your instance from the command line using the SSH instructions and type these two commands to fix it:

sudo systemctl disable gdm
sudo reboot now

then wait a couple minutes for your instance to reboot and become online, and then retry connecting from GUI.

• Cursor is laggy, and there is high latency in UI (not just Slicer, all windows are slow to refresh): This is due to limited bandwidth of your network connection. If you can try, switching to a different network. Also for better performance, we suggest using the dedicated TurboVNC client, which is freely available at https://github.com/TurboVNC/turbovnc/releases. TurboVNC allows you to reduce the quality of the desktop image so that less bandwidth is used when refreshing the screens. Once the TurboVNC is installed, you can connect to your running instance using the convention: W.X.Y.Z:1, where W.X.Y.Z is the IP address of the instance provided in the email. You will use the same passphrase as the connection password. You can also check the email sent by the MorphoCloudPortal for more instructions on using TurboVNC.

• Instance GUI doesn’t scale correctly (i.e., fonts too small/large): Try using TurboVNC client for connecting to the instance. TurboVNC handles the high-resolution DPI scaling and resizing the application window much better than the web browser connections.

• 3D rendering performance of Slicer appears slow:, You need to make sure: (1) 3D Slicer is actually using the GPU, and (2) GPU on the instance is functional. For the first issue, make sure you are always launching the Slicer from the shortcut on the desktop. For the latter, open a terminal window and type the command nvidia-smi. This should list the GPU in your system, its driver and CUDA version, along with the applications using it (e.g., if Slicer is running, it should be listed). If there is any error, please copy and paste the full screenshot and notify us. Sometimes rebooting the system from the command line via the command sudo reboot, helps clear the issue.

• I checked, but 3D rendering performance of Slicer still very slow, application window is unresponsive: Try to stop volume rendering (turn it off). See if any dimension of the your volume is equal to or greater than 4096 voxels. If so, use CropVolume module in Slicer to downsample those dimensions to 4095 or less. Note that this appears to be a specific problem with the drivers provided on the MorphoCloud host (JetStream2), and does not happen in other systems.

Citation

If you use MorphoCloud on your teaching and research, please cite this publication:

Maga AM, Fillion-Robin J-C (2026) MorphoCloud: Democratizing Access to High-Performance Computing for Morphological Data Analysis. https://doi.org/10.12688/f1000research.176328.1

Funding & Acknowledgement

MorphoCloud services, including MorphoCloud On-Demand Instances, are supported by funding from the National Science Foundation (DBI/2301405) and National Institutes of Health (NICHD/HD104435). MorphoCloud runs on cyberinfrastructure that is made available by current and previous funding from the National Science Foundation (JetStream2: OAC/2005506, Exosphere: TI/2229642). Initial development of SlicerMorph was previously supported by National Science Foundation (DBI/1759883).

If you use any of the MorphoCloud services for your project, please acknowledge our funders with this statement:

This study relied on cyberinfrastructure supported by grants from the National Science Foundation (MorphoCloud: DBI/2301405; JetStream2: OAC/2005506; Exosphere: TI/2229642) and the National Institutes of Health (MorphoCloud: NICHD HD104435).