FDA 510(k) clearance explained

Interactive 510(k) Summary Metadata Inspector

Click on any field on the simulated clearance document below to inspect its official definition, plain-language breakdown, and regulatory implications.

Filter Field:

FDA 510(k) Summary Sheet — K260667

Public Document (21 CFR 807.92)

Select a Field to Inspect

Click any entry on the left to unpack its regulatory meaning, CFR citations, and specific context for the Alturion filing.

Reference Document

Philips Alturion K260667 Summary

FDA Database

How to Dissect a 510(k) Clearance Like an FDA Reviewer

A step-by-step masterclass analyzing K260667 across predicate selection, hardware comparison, AI validation, PCCP strategy, and post-market duties.

1 Predicate Selection Strategy

Everything Starts with the Predicate

The entire 510(k) pathway hinges on proving that a new device is **Substantially Equivalent (SE)** to a legally marketed predicate device. In submission K260667, Philips selected their own previously cleared **Affiniti Series Diagnostic Ultrasound System (K243794)** as the primary predicate.

Pro Tip: Own Predicate vs. Competitor Predicate

Citing your own prior cleared device (as Philips did) generally leads to a smoother review process because the manufacturer possesses all underlying design histories, raw safety test reports, and transducer calibration files. Citing a competitor device is permitted, but requires proving equivalence through public specs or bench testing without access to the competitor's proprietary design history file (DHF).

Role of Reference Devices in K260667

Philips also cited three **reference devices**. Reference devices are not predicates used to claim SE directly, but serve as supporting evidence that specific sub-technologies are already cleared and standard in the market:

K231190 Philips Epiq Series

Establishes baseline ultrasound console architecture.

K253595 Philips Epiq (Auto Measure)

Establishes prior acceptance of Auto Measure Abdomen algorithm.

K231966 GE LOGIC E10

Competitor reference confirming industry standard for transducer design.

Device Lineage Tree

Submission Lineage Map

Primary Predicate

Affiniti Series (K243794)

Core acoustic engine & console platform

Reference Features

Epiq Series (K253595)

Shared AI Auto-Measure codebase

NEW SUBMISSION

Alturion Series (K260667)

Console refresh + eV14-2v probe + Auto Measure AI

Substantial equivalence does not require identity — only equivalent safety & effectiveness profile.

2 Side-by-Side Scrutiny

Read the Comparison Table Like an FDA Reviewer

The heart of any 510(k) summary is the comparative analysis. In K260667, Philips compared the new **eV14-2v transducer** against the predicate **3D9-3V transducer** across nearly 20 design parameters.

"Same as Predicate" Rows

Parameters like intended use, acoustic output display indices, and materials in patient contact matched directly. These require minimal reviewer intervention.

"Different Than Predicate" Rows

This is where reviewer scrutiny focuses! Philips proactively framed every design change as either an anatomical benefit or ergonomic improvement without introducing new risks.

How Differences Were Defended in K260667:

•
436 elements vs. 128 elements: Framed as improved spatial resolution, contrast, and clutter reduction.
•
180° FOV vs. 155.5° FOV: Framed as clinical enhancement allowing visualization of larger anatomical structures in a single view.
•
Smaller tip dimensions & shorter length: Framed as patient comfort and operator tip maneuverability optimization.

3 AI Performance Validation Strategy

Validating AI Software Without a Clinical Trial

A major highlights of K260667 is how Philips cleared the **Auto Measure Abdomen** feature. They did NOT conduct a prospective randomized clinical trial. Instead, they performed a rigorous **non-clinical algorithm performance study**.

150 Subjects & 3 Sites

Retrospective dataset collected across 3 independent clinical sites to ensure demographic and operator diversity.

Expert Ground Truth

3 ARDMS-certified sonographers with 10+ years experience established manual measurement baselines independently.

Train/Test Independence

Validation images were completely isolated from the training set, sourced from Epiq and Affiniti consoles.

Statistical Methodology

Bland-Altman Limits of Agreement (LoA) Simulation

Select Parameter:

Bland-Altman plots quantify agreement between AI automated measurements and human expert ground truth by plotting the difference against the mean of the two measurements. All 4 organ parameters met pre-specified limits of agreement.

MEAN BIAS +0.12 mm

UPPER LIMIT (+1.96 SD) +2.45 mm

LOWER LIMIT (-1.96 SD) -2.21 mm

5 Post-Market Compliance

Never Skip the "Standard" Boilerplate Text

Junior regulatory professionals often skip the standardized closing paragraphs of an FDA clearance letter. However, these paragraphs represent binding legal duties triggered upon market entry.

21 CFR Part 820 / ISO 13485

QMSR & Design Controls

Clearance is granted under the condition that design controls (7.3), nonconforming product procedures (8.3), and CAPA systems (8.5) are maintained during manufacturing.

21 CFR Part 803

Medical Device Reporting (MDR)

Mandatory reporting obligations kick in immediately upon clearance. Any device malfunction that could cause death or serious injury must be reported to FDA.

21 CFR Part 830

Unique Device Identification (UDI)

Proper labeling and electronic submission of device identifiers to FDA’s Global UDI Database (GUDID) prior to commercial distribution.

21 CFR Part 807 Subpart E

Annual Registration & Listing

Philips Ultrasound, LLC must maintain active establishment registration and list the Alturion platform under product code IYN.

Interactive Technical Comparison Matrix

Comparing New eV14-2v Transducer vs. Predicate 3D9-3V Transducer (K260667 Filing Data)

Toggle Mode:

Parameter	New Device: eV14-2v	Predicate: 3D9-3V	Comparison Status	FDA Reviewer Justification Analysis

Philips Alturion Series Diagnostic Ultrasound System