Skip to content

System Requirements

Mohamed E. Masoud edited this page Jul 9, 2023 · 13 revisions

GPU

GPU card is highly recommended on the client side. For example, GeForce GTX 1050 Ti could run all the models.

Internal or built-in graphics chips (e.g., HD, UHD) can run mostly light models such as: "Full Brain GWM (light)", "Compute Brain Mask (FAST)", or "Extract the Brain (FAST)", but they can't run relatively large models such as "Full Brain GWM (large)", "FS aparc+aseg Atlas 104 (failsafe)" or "Cortical Atlas 50".

Memory

System memory of 16 GB or higher is recommended on the client side.

Browser

Brainchop needs a browser that supports the JS ES6 version and WebGL2. For the JS ES6 requirement, simply any browser version released on or after 2018 should satisfy this requirement. For a list of all compatible browsers, please refer to this link.

For browsers compatible with WebGL2, please refer to this link.

Verified H/W and S/W

Brainchop models have been tested on different platforms, a list of verified GPUs and software can be found here.

Users can check their browser resources window by click on Screenshot 2023-03-08 at 05-52-15 Brainchop icon in the UI taskbar.

Screenshot from 2023-03-10 16-52-47

Recommended Texture Size for segmentation models:

Full Brain GWM (light): At least texture size of 9159.

Full Brain GWM (large): At least texture size of 13585.

Extract the Brain (FAST) : At least texture size of 9159.

Extract the Brain (High Acc) : At least texture size of 13585.

Compute Brain Mask (FAST) : At least texture size of 9159.

Compute Brain Mask (High Acc): At least texture size of 13585.

Cortical Atlas 50: Texture size of 16384 is recommended.

FS aparc+aseg Atlas 104: At least texture size of 18121. Recommended 32768.

FS aparc+aseg Atlas 104 (failsafe): At least texture size of 16384.

Clone this wiki locally