Nalvera FAQ — Medical Imaging AI Inference

Question 1

How do I submit my first inference AI job?

Accepted Answer

After registering and verifying your email, open Submit Job, upload a compressed NIfTI file (.nii.gz), pick an AI model, and submit it to the GPU cluster. You can track progress on the Job Queue page; most jobs finish in under ten minutes, and results appear in My Results with an in-browser viewer and a download link.

Question 2

What file format do I need to upload?

Accepted Answer

Nalvera accepts compressed NIfTI files (.nii.gz). DICOM data must be converted to NIfTI before upload, and the file must stay within the platform's file-size limit.

Question 3

How long does a job typically take to process?

Accepted Answer

Most jobs complete in under ten minutes. Processing time depends on the image volume (more megavoxels take longer) and the AI model selected (more sophisticated models need more compute). Job status can be monitored in real time on the Job Queue page.

Question 4

What is Study Design and when should I use it?

Accepted Answer

Study Design organises your imaging into studies, patients and scans instead of a flat list of uploads. You tag each scan's modality and date, store it on the platform, and can then dispatch a single scan, a whole patient, or an entire cohort to any AI model in one action — useful for running the same models across many patients while keeping results grouped.

Question 5

How is the credit cost of a job calculated?

Accepted Answer

Each job has a flat processing fee (fixed per job) plus a volume-based model cost that scales with the image's megavoxel count and the selected model's rate. The combined cost is converted to credits; monthly credits are drawn first and top-up credits cover any remainder.

Question 6

Who can see the medical images I upload?

Accepted Answer

Only you and one platform administrator can access your uploaded data. Other users — including members of your own institution or lab — cannot see your jobs, input files, or results. Uploads are stored on access-controlled EU infrastructure, every transfer is TLS-encrypted, and you can delete a job and its files at any time from My Results.

Question 7

Should I anonymize my images before uploading?

Accepted Answer

Yes. For NIfTI uploads Nalvera strongly recommends stripping all header metadata before upload, removing any field that could link an image to an individual. DICOM folders are converted to NIfTI on upload and DICOM header attributes are discarded in the process. The platform also applies its own header sanitisation as a secondary safeguard. Tools such as dcm2niix and nibabel can help inspect and clean NIfTI headers.

Question 8

How long are my files stored on the platform?

Accepted Answer

By default, files remain on the platform indefinitely and are only removed when you delete them from My Results. You can also enable auto-delete (Settings → Extras) with a retention window between 7 and 365 days, after which older completed jobs are removed automatically. Deletion is permanent and removes the input file, the output file, and all associated job information.

Question 9

What is a Lab, and how does it differ from a personal account?

Accepted Answer

A Lab is a shared workspace for teams and organisations — research groups, companies, or CROs. A lab manager can invite members, top up a shared credit pool, and set per-member spending limits from a central dashboard. Each member still submits and owns their jobs under their personal account, and job data and results stay private per user.

Question 10

Can I view my results without downloading?

Accepted Answer

Yes. When a job completes, open My Results and click Preview to use the in-browser viewer. Segmentation results are shown as a colour overlay on the original scan in axial, coronal and sagittal views, with toggleable label classes and adjustable opacity; generated images are shown side by side with the input. Built-in tools let you review and adjust outputs before exporting.

Frequently Asked Questions