FDA product code · Radiology
QIHAutomated Radiological Image Processing Software
Product code QIH covers Automated Radiological Image Processing Software, a Class II device regulated under 21 CFR 892.2050. FDA has cleared 314 devices in this code, 93 of them in the last 12 months. Median review time across the 314 most recent decisions is 140 days. openFDA lists 2 recall records against this code.
Updated · ingest status · JSON
- Class
- Class II
- Regulation
- 21 CFR 892.2050
- Review panel
- Radiology
- Clearances
- 314
What is happening in QIH right now?
QIH (Automated Radiological Image Processing Software) has three findings, most consequential first. 1 recall opened in the last 12 months and not yet terminated (Class II) (Z-2767-2026). 40 first-time filers in the last 12 months, including ROPCA ApS and Velmeni, Inc (K260192, K254096, K260206, K253628, K260316). The code is crowded, scoring 98 of 100 — 175 companies hold a clearance here and 93 arrived in the last 12 months (QIH:crowding).
| Finding | How much it matters | FDA records |
|---|---|---|
| Recall | Class II recall | Z-2767-2026 |
| New entrant | Worth knowing | K260192, K254096, K260206, K253628, K260316 |
| Crowding | Context | QIH:crowding |
Sources checked: predicate ancestry, recalls, adverse events, new entrants, review time, crowding. Composed . Every finding carries the record ids it was derived from; nothing here is an estimate.
What does product code QIH cover?
QIH is FDA's product code for Automated Radiological Image Processing Software, a Class II device regulated under 21 CFR 892.2050.
To provide automated radiological image processing and analysis tools. Software implementing artificial intelligence including nonadaptive machine learning algorithms trained with clinical and/or artificial data.
Source: openFDA device/classification · record QIH · retrieved
How many devices has FDA cleared in QIH?
FDA has cleared 314 devices under product code QIH since the 510(k) programme began in 1976, 93 of them in the last 12 months.
Source: openFDA device/510k · retrieved
| 510(k) | Device | Applicant | Decision |
|---|---|---|---|
| K261349 | SwiftMR | Airs Medical, Inc. | |
| K260192 | DIANA | ROPCA ApS | |
| K254096 | VELMENI for DENTISTS (V4D) 3D | Velmeni, Inc. | |
| K261535 | Hi-D Imaging 4TAVR | Hi-D Imaging AG | |
| K260206 | SkeletonPlanner | UPlanner B.V. - Presurgeo | |
| K260973 | InferOperate Suite | Beijing Infervision Healthcare Medical Technology Co., Ltd. | |
| K253969 | ADAS 3D | Adas3D Medical S.L. | |
| K253628 | Auto-Seg (SO-0012), Spine Auto-Seg (SO-0012) | Agada Medical, Ltd. | |
| K253786 | syngo.CT Coronary Cockpit | Siemens Medical Solutions USA, Inc. | |
| K260316 | CMN Capillary Function with Oncology Virtual Expert | Cercare Medical A/S |
How long does FDA review take for QIH?
The median FDA review time for QIH is 140 days, measured across the 314 most recent decisions with both a received date and a decision date.
Review time here is decision date minus date received, as published by FDA. It is elapsed calendar time, not FDA review days — the two differ whenever the agency stops the clock for an additional information request.
Source: openFDA device/510k · retrieved
Are there recalls in QIH?
openFDA lists 2 recall records against product code QIH. A recall record covers one firm's corrective action and may span many device lots.
Z-2767-2026Open, ClassifiedPIE Medical Imaging B.V.
User measurement value reported by diagnostic bioimaging software intended to measure/visualize cardiovascular structures may be larger (0.8% to 34.4%) than the actual anatomical dimension, if all the following occur: affected software is used, and the user has access only to 3D Echo OR Mitral 3D Echo, and the dataset being analyzed is a Philips 3D echo dataset containing Color Doppler information
Z-2111-2021TerminatedCaption Health
The firm is aware of a issue with ultrasound software that results in unintended video frames being included. This could lead to an incorrect automated ejection fraction (AutoEF) and image quality score (IQS) estimates.
Recall class comes from openFDA's enforcement endpoint, which does not carry every recall — where it is missing, none is shown rather than guessed. Status is FDA's own wording; a terminated recall is a closed one, and the total spans the code's whole history, not just what is open today.
Source: openFDA device/recall · retrieved
How many adverse events are reported for QIH?
MAUDE holds 70 adverse-event reports naming product code QIH, 64 of them received in the last 12 months.
MAUDE reports are submitted by manufacturers, importers, user facilities and the public. They are unverified, and a report is not proof that the device caused the event.
Source: openFDA device/event · retrieved
Is adverse-event reporting in QIH unusual?
QIH does not yet have twelve months of adverse-event history in our data, so there is no expected range to compare against. That is a gap in what we hold, not a statement about the device.
| Month | Reports | Expected | Verdict |
|---|---|---|---|
| 1 | Insufficient history | No expected range — too little history | |
| 1 | Insufficient history | No expected range — too little history | |
| 1 | Insufficient history | No expected range — too little history | |
| 1 | Insufficient history | No expected range — too little history | |
| 2 | Insufficient history | No expected range — too little history | |
| 1 | Insufficient history | No expected range — too little history | |
| 4 | Insufficient history | No expected range — too little history | |
| 5 | Insufficient history | No expected range — too little history | |
| 3 | Insufficient history | No expected range — too little history | |
| 6 | Insufficient history | No expected range — too little history | |
| 4 | Insufficient history | No expected range — too little history | |
| 6 | Insufficient history | No expected range — too little history | |
| 4 | Insufficient history | No expected range — too little history |
13 months do not have a full, unbroken year of history before them, so no expected range is shown for those. Insufficient history is reported as insufficient rather than as zero: an empty baseline would turn a gap in our data into a spike.
Who has entered QIH in the last year?
8 companies filed in QIH for the first time in the last 12 months.
| Company | First cleared | K-number |
|---|---|---|
| ROPCA ApS | K260192 | |
| Velmeni, Inc. | K254096 | |
| UPlanner B.V. - Presurgeo | K260206 | |
| Agada Medical, Ltd. | K253628 | |
| Cercare Medical A/S | K260316 | |
| Anumana, Inc. | K260300 | |
| Brainomix Limited | K260406 | |
| Keya Medical Technology Co., Ltd. | K260497 |
"First time" means their first clearance in this product code, not their first ever — an established manufacturer entering a new category is exactly the movement worth knowing about.
Source: openFDA device/510k · retrieved
What are the predicate devices in QIH?
FDA publishes no machine-readable predicate field, so no predicate chain is recorded for this code yet.
A predicate chain matters because a device inherits the safety history of what it was compared to. FDA publishes the chain only inside each 510(k) summary PDF, never as data — so where we have edges, they are read from those documents, and where we do not, we say so instead of showing an empty list.
Who files 510(k)s in QIH?
Across the 314 most recent decisions, these applicants filed most often in QIH.
| Applicant | Recent clearances |
|---|---|
| Siemens Medical Solutions USA, Inc. | 12 |
| Dia Imaging Analysis, Ltd. | 10 |
| Shanghai United Imaging Healthcare Co., Ltd. | 6 |
| Clarius Mobile Health Corp. | 5 |
| Ischemaview, Inc. | 5 |
Applicant names above are exactly as FDA published them, so one company can appear under several spellings.
Source: openFDA device/510k · retrieved
Watch QIH
Get an email the morning anything changes in QIH — a new clearance, a recall, or an adverse-event spike above this code's own baseline.
Related product codes in Radiology
Data on this page comes from FDA's public openFDA API and is US public domain (CC0). DeviceSignal is an independent commercial service and is not affiliated with the U.S. Food and Drug Administration. Information only, not regulatory advice. Verify anything you rely on against open.fda.gov.