Skip to content

Bulk Raw Data Transfer, Direct iPad-to-PC Backup, and Recovery of Missing Data #48

Open

Description

@jhuang-jhh

We have a few questions regarding the most efficient way to transfer and back up our raw data.

Is there a way to connect the iPad directly to a PC and transfer the raw data by simply dragging and dropping the files, rather than downloading the data from the cloud each time?

Alternatively, is there an option on the website to download all raw data in bulk at once, preferably with recognizable file names that allow us to easily identify and organize the data?

We collect data every day, so we are trying to streamline our workflow and make the data transfer and backup process as efficient and reliable as possible. Ideally, we would like to connect the iPad to a PC and transfer all of the raw data at once for local backup and archiving.

We also encountered an issue last week where the cloud backup appeared to fail, even though the app indicated that the data had been successfully uploaded. We can no longer find the data on the iPad, possibly because it was automatically removed after the upload. Is there any way to recover or restore this missing data?

Our goal is to ensure that the raw data is successfully backed up every time and that we have a simple, reliable workflow for transferring and archiving our daily data.

Could you please let us know:

  1. Whether direct iPad-to-PC transfer of the raw data is possible.
  2. Whether bulk download/export of all raw data is available through the website.
  3. Whether the exported files can retain recognizable file names or identifiers.
  4. Whether the missing data from last week can be recovered or restored.
  5. If there is a recommended workflow for reliably backing up raw data on a regular basis.

Thank you for your help!

Image

Activity

  1. clariusk commented on Aug 20, 2026

    @clariusk
    Contributor

    Thanks for the detailed write-up — answering all five.

    1. Direct iPad-to-PC transfer — yes, and it's the workflow I'd recommend. The App exposes its Documents folder to iOS file sharing:

    • iPad: Files app → On My iPad → Clarius
    • Windows: connect over USB, then the Apple Devices app → Files → Clarius
    • Mac: connect over USB, then Finder → select the iPad → Files tab → Clarius

    Drag the whole folder across in one go. Note that raw data always exports to this Documents location (as do PDF and DICOM) — the Export Location setting only affects images and movies.

    2. Bulk download from the website — not available today; raw data downloads one capture at a time. I've logged the request.

    3. Recognizable filenames on cloud download — raw data is stored keyed by capture UUID, so a downloaded file doesn't carry a patient/date/study identifier in its name. Also logged. For now this is a further reason to prefer the USB route in (1), where the on-device naming is preserved.

    4. The missing data — your instinct is right, and the cause is a setting you control. There's an Exam Auto Purge setting in the App (Settings → Exam Auto Purge): "the number of days before an exam is automatically deleted after successful upload to the Clarius Cloud or to a DICOM Storage server." Options are Never, 60, 30, 10, 2, 1 day, or Immediately. If yours is set short, exams are removed from the iPad that many days after they upload.

    Two things worth knowing:

    • It only triggers after the App has recorded a successful upload, so the data should exist in the cloud. If it isn't visible there, tell us and we'll check storage directly — there's a known case where a file uploads fine but our record of its arrival doesn't register, which would make it invisible in the web UI while still being safely stored.
    • An institution-level policy overrides the local setting. If your institution has a purge policy configured, changing it on the iPad won't take effect — that has to be changed at the institution level.

    For your workflow I'd set Exam Auto Purge to Never (or 30–60 days) and delete manually once you've confirmed your local backup.

    To chase the specific missing session, send us the approximate date/time, the institution/account, and the probe serial.

    5. Recommended workflow — make the USB copy primary, cloud secondary. At the end of each collection day: export raw for the day's exams, connect over USB, copy the Clarius folder across, verify it, and only then let anything be removed from the iPad. With Auto Purge set to Never, nothing disappears on its own, and given (2) and (3) I wouldn't rely on cloud download as the archival route for daily research collection.

  2. jhuang-jhh commented on Sep 16, 2026

    @jhuang-jhh
    Author

    Hi Clarius Team,

    Thank you for your response. I’d like to get some more details about the first question, particularly since you suggested using USB copy as the primary method.

    When I open the iPad Files app, I only see the Clarius Export folder. My understanding is that the data will not appear in this folder automatically unless I explicitly export it to the iPad’s local storage.

    This leads to my first question:

    When the data is exported, I see four different file formats/options: DICOM, JPEG, PNG, and BMP. Which of these should I generate/use if my goal is to match the original raw data that is stored in the Clarius Cloud?

    My second question is about the temporary data:

    Immediately after recording an ultrasound video, the data is temporarily stored on the iPad before it is uploaded to the Clarius Cloud. Where exactly is this temporary data stored on the iPad? Is there a way to access or locate these temporary files directly on the iPad?

    I’m particularly interested in understanding whether the raw ultrasound video/data exists locally on the iPad before or during the cloud upload, and whether we can access that data without going through the export process.

    Thank you for your clarification.

    Best regards

  3. clariusk commented on Sep 16, 2026

    @clariusk
    Contributor

    Both answers come down to one thing: raw data isn't one of the four export formats — the App copies it out for you automatically.

    1. Which format? DICOM/JPEG/PNG/BMP are rendered screen images only. Raw data is a separate .tar per capture, included in every local export regardless of the format you pick, for captures made with Buffer on. It's byte-for-byte the package the App uploads to the cloud.

    2. Pre-upload data isn't temporary — it lives in the App's private storage, which iOS doesn't expose to Files or USB. Export is the only route, and it's a straight copy, so nothing is lost.

    Getting the raw data out (step by step)
    • Before scanning, turn on Buffer ("Buffer Raw Data on Scanner") in the Research column of the scan menu. This needs the Raw Data feature on your account — if you don't see the column, let us know.
    • Captures that carry raw data show an RF badge in the review screen.
    • Export the exam locally with any format (JPEG is fine). The raw packages appear alongside the images as raw_<n>_<m>.tar under Clarius Export/<patient>/<date>/, and always land in Files regardless of the Export Location setting.
    • Inside each .tar are the per-capture .raw.lzo data blocks (RF / IQ / envelope — decompress with lzop -d) and a .yml with the acquisition parameters. Format docs and Python/MATLAB readers: https://github.com/clariusdev/raw
    • For a whole day at once, multi-select the exams in the exam list and export them in one go.
    Why you only see "Clarius Export"

    The App keeps every exam (images, cines and raw packages) in its own private storage on the iPad until you delete it, or Auto Purge does. iOS file sharing only exposes the App's Documents folder, and the exam store is deliberately not in there — Clarius Export is the only thing the App ever writes to Documents, and only as the destination of an export. There's no way to reach the exam store directly on an unmodified iPad.

  4. jhuang-jhh commented on Sep 28, 2026

    @jhuang-jhh
    Author

    Hi Clarius Team,

    Thank you for your previous response. The information you provided was very helpful. It could help us back up data more efficient.

    We have noticed that some data we previously stored in Clarius is now missing, and we lost a significant amount of data just this week. We would like to better understand what may have happened and whether there is a way to investigate the account activity.

    Could you please help us determine:

    Does data stored in Clarius cloud expire or get automatically deleted after a certain period of time?
    Is there an activity or audit log that would allow us to see whether any data was deleted or removed?
    Is there a way to determine whether one of our team members changed permissions or removed access to the data?
    Can you identify when the missing data was deleted or otherwise removed, and whether it can be recovered?

    If this issue needs to be handled by a specific department or support team, could you please let me know whom I should contact? I can provide our account number, ultrasound device number, and any other information you need to investigate the issue.

    We are particularly concerned because a large amount of data appears to have disappeared within the past week.

    Thank you for your assistance.

  5. clariusk commented on Sep 28, 2026

    @clariusk
    Contributor

    Short answer: Clarius Cloud never deletes exam data on its own, and nobody in your institution can permanently delete an exam. Data that seems to have vanished has almost always been archived (hidden), or someone has lost permission to see it. Both are reversible.

    1. Does cloud data expire or get auto-deleted? No. There is no retention period, storage quota or scheduled job that removes exams, images, cines or raw data. Letting a licence or trial lapse doesn't delete anything either. It can stop new uploads, or block raw-data downloads until the Research licence is renewed.

    2. Is there an activity or audit log? Yes. Institution admins (and IT admins) can open it from the institution settings, under Audit Logs, and export it as a CSV. It covers the last 6 months and shows the date, event type, who did it, which patient was affected, and details. Events include exams being archived and restored, users being removed, admin rights being granted or revoked, and user role edits.

    3. Can you tell whether a team member changed permissions or removed access? Mostly yes, from the same log: look for Removed user(s), Edited user(s), Granted/Revoked admin status. How much someone can see depends on their role:

    • Admins and observers see every exam in the institution.
    • Clinicians see only their own exams.
    • IT admins see none.

    So if an admin was changed to clinician, or removed from the institution, a large set of exams would disappear from their view overnight. The exams would still be there.

    4. When was it removed, and can it be recovered? "Delete" in the web portal is really archive, which hides the exam from every list but keeps it. Archived exams are listed at https://cloud.clarius.com/<institution id>/exams/archive/ and can be restored from there. Check for Exam Archived entries in the audit log to see when an exam was archived and by whom.

    Other reasons data can seem to disappear
    • Wrong institution: if your account belongs to more than one institution, exams go to whichever one was selected in the App at upload. Check the institution switcher.
    • Role change: as in (3), confirm the affected user's current role in the institution's user list.
    • Filters: the portal remembers table filters and sorting per session, so clear them before concluding something is gone.
    • Upload never completed: an exam can be recorded in the App but not finished on the cloud side. This is another reason for the daily USB backup from our earlier reply.

    Who to contact: if none of the above explains it, email support@clarius.com. Include your institution name/account number, the probe serial(s), the date range, and a couple of example exams you expect to see.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions