QIH: Automated radiological image processing software
What it covers
Software that provides automated radiological image processing and analysis tools using artificial intelligence, including non-adaptive machine learning algorithms trained with clinical or artificial data. FDA’s product code definition notes that the training data typically affects device performance, and that adaptive AI algorithms are outside this code. It falls under the medical image management and processing regulation, 21 CFR 892.2050.
Recent clearances
| 510(k) | Device | Applicant | Decision | Type |
|---|---|---|---|---|
| K262147 | CINA-Aneurysm+ | Avicenna.Ai | Oct 19, 2026 | Traditional |
| K261733 | Ligence Heart | UAB Ligence | Oct 7, 2026 | Traditional |
| K262134 | NeuroDR | Deep Reasoning Ai, Inc. | Oct 7, 2026 | Traditional |
| K260523 | QGuideCT (1.0) | Medis Medical Imaging Systems BV | Oct 5, 2026 | Traditional |
| K260055 | syngo.CT Brain Quantification | Siemens Medical Solutions USA, Inc. | Sep 25, 2026 | Traditional |
| K262390 | BoneXpert | Visiana | Sep 22, 2026 | Traditional |
| K261368 | Rapid Platform (R6.7) | Ischemaview, Inc. | Sep 21, 2026 | Traditional |
| K261623 | EchoPAC Software Only / EchoPAC Plug-in (v211) | GE Medical Systems Ultrasound and Primary Care Diagnostics | Sep 16, 2026 | Traditional |
| K261677 | Artificial Intelligence Renal Assessment (AIRA) Version 1 | Aramis Global AI Consultancies, LLC | Sep 15, 2026 | Traditional |
| K260879 | 4DM | INVIA, LLC | Sep 10, 2026 | Traditional |
Clearances by year
337 QIH 510(k)s were cleared from 2011 to Oct 29, 2026. Days to decision run from the day FDA received the 510(k) to its decision, including time on hold.
| Year | Clearances | Median days to decision |
|---|---|---|
| 2026 (so far) | 77 | 132 |
| 2025 | 94 | 150 |
| 2024 | 64 | 141 |
| 2023 | 45 | 151 |
| 2022 | 25 | 134 |
| 2021 | 19 | 125 |
| 2020 | 11 | 145 |
| 2019 | 2 | 190 |
Who submits
194 applicants cleared a QIH device since 2011. 40% of the clearances went to applicants in United States, then South Korea (7%), Canada (6%) and Israel (6%). By type: Traditional 88%, Special 12% and Abbreviated 0%.
| Applicant | Clearances since 2011 |
|---|---|
| Siemens Medical Solutions USA, Inc. | 13 |
| Dia Imaging Analysis, Ltd. | 10 |
| Ischemaview, Inc. | 6 |
| Shanghai United Imaging Healthcare Co., Ltd. | 6 |
| Clarius Mobile Health Corp. | 5 |
| Overjet, Inc. | 5 |
| Siemens Healthcare GmbH | 5 |
| Aidoc Medical , Ltd. | 4 |
What FDA’s draft AI guidance asks for
21 CFR 892.2050 sets Class II with special controls and names DICOM, JPEG and the SMPTE test pattern as voluntary standards, but lists no device-specific testing. These items come from FDA’s January 2025 draft guidance on AI-enabled device software functions, which is not final. Read the source.
- Performance validation on test data independent of the training data, for example from different clinical sites, and kept away from the model developers
- Performance across subgroups of the intended use population, to address AI bias
- A description of the model, its development and its data management
- Software and cybersecurity documentation, and labeling that describes the model and its performance
Standards
QIH is one of the product codes most common on FDA’s list of AI-enabled devices. See how FDA regulates AI-enabled medical devices.
Ask about QIH
Questions about QIH
What is FDA product code QIH?
Product code QIH covers automated radiological image processing software, Class II devices regulated under 21 CFR 892.2050 and reviewed by FDA's Radiology panel. FDA's 510(k) database lists 338 clearances under QIH.
How many 510(k) clearances are there under QIH?
FDA’s 510(k) database listed 338 clearances under QIH as of Nov 16, 2026, 337 of them decided since 2011; the most recent decision was Oct 19, 2026. The Predicate Finder searches the live database.
How long does FDA take to clear a QIH 510(k)?
For the 182 QIH clearances decided in the two years before Oct 29, 2026, the median time from FDA receiving the 510(k) to its decision was 141 days. That counts calendar days, including time the submission was on hold waiting for answers from the submitter.
Which companies have the most QIH clearances?
Since 2011: Siemens Medical Solutions USA, Inc. (13), Dia Imaging Analysis, Ltd. (10), Ischemaview, Inc. (6), Shanghai United Imaging Healthcare Co., Ltd. (6) and Clarius Mobile Health Corp. (5). 194 applicants have cleared a QIH device in that time.
What testing do QIH submissions include?
21 CFR 892.2050 sets Class II with special controls and names DICOM, JPEG and the SMPTE test pattern as voluntary standards, but lists no device-specific testing. These items come from FDA’s January 2025 draft guidance on AI-enabled device software functions, which is not final. In short: performance validation on test data independent of the training data, for example from different clinical sites, and kept away from the model developers; performance across subgroups of the intended use population, to address AI bias; a description of the model, its development and its data management; software and cybersecurity documentation, and labeling that describes the model and its performance.
Sources
- FDA product classification: QIH
- 21 CFR 892.2050: Medical image management and processing system
- openFDA 510(k) query for QIH
- Artificial Intelligence-Enabled Device Software Functions: Lifecycle Management and Marketing Submission Recommendations
- 510(k) Premarket Notification database
- List of Artificial Intelligence-Enabled Medical Devices