Philips Wins FDA Clearance for Verida, the First Detector-Based Spectral CT With AI Baked Into the Reconstruction Chain
The Dutch medtech giant cleared a 510(k) for a spectral CT that reconstructs 145 images per second using a deep learning engine, bringing always-on spectral imaging to US hospitals for the first time.
Overview
Royal Philips has received US Food and Drug Administration 510(k) clearance for the Philips Spectral CT Verida, the company’s next-generation computed tomography platform that integrates deep learning reconstruction directly into the imaging chain. The clearance, announced on April 16, 2026, opens the US market to a system Philips had previewed at RSNA 2025 and shipped only in CE-marked regions since late 2025.
Philips describes Verida as the world’s first detector-based spectral CT to combine always-on spectral acquisition with an AI reconstruction engine across the full imaging pipeline. The regulatory nod lands as the three legacy imaging incumbents — GE HealthCare, Philips, and Siemens Healthineers — compete aggressively on AI-integrated scanners, which dominated the 2025 RSNA showcase.
What We Know
The hardware. Verida is built around what Philips calls its third-generation Nano-panel Precise dual-layer detector, a design with intrinsic noise reduction that captures both high- and low-energy x-ray data in a single acquisition. According to Philips’ FDA announcement, the system reconstructs 145 images per second and produces full exams in under 30 seconds — twice as fast as the previous generation — supporting throughput of up to 270 exams per day on the premium configuration.
The AI layer. Verida incorporates a feature Philips calls Spectral Precise Image (SPI), an AI-based deep learning reconstruction engine designed to reduce noise and enhance image quality. Per the April 16 press release, clinicians can customize image de-noising to their preferences, and the feature was clinically evaluated in adults aged 22 and older — pediatric use of SPI has not yet been validated. Philips stresses that SPI provides supplemental data and does not replace professional clinical judgment.
Efficiency claims. At the November 2025 RSNA debut, Philips said Verida’s AI features can enable dramatic dose reduction without compromising image quality and cut energy consumption by up to 45 percent, according to the launch announcement. The company also framed Verida as an extension of its broader spectral CT platform, which it said has more than 800 global installations and is supported by over 800 peer-reviewed publications, per the same launch release.
Clinical indications. The FDA cleared Verida for diagnostic imaging in radiology, interventional radiology, and cardiology, plus oncology applications covering treatment preparation and radiation therapy planning. Indications include head, whole-body, cardiac, and vascular CT across all patient ages, as well as low-dose CT lung cancer screening within established protocols, per the April 16 release. Extended field-of-view imaging and respiratory-correlated 4DCT are limited to treatment preparation and radiation therapy planning.
Executive framing. Dan Xu, business leader of CT at Philips, said in the company statement that combining always-on spectral imaging with AI-powered reconstruction enables clinicians to see more information on the first scan, reducing the need for repeat acquisitions.
What We Don’t Know
- Philips has not disclosed pricing for Verida in the US market, the timeline for first hospital installations, or specific customer commitments.
- The April 16 clearance materials do not cite head-to-head clinical study data comparing Verida to competing dual-source or photon-counting CT systems from GE HealthCare or Siemens Healthineers, whose own AI-forward imaging platforms shared the RSNA 2025 stage with Verida.
- The 45 percent energy reduction claim is framed by Philips in its November 2025 launch release as being based on axial body scans with 80 percent dose reduction, excluding system preparation energy, per the original launch documentation — the real-world energy profile across mixed clinical workloads has not been independently verified.
- The FDA clearance letter and Philips’ announcement do not indicate whether the submission included a Predetermined Change Control Plan that would allow Philips to update the Spectral Precise Image algorithm post-market without filing new 510(k)s.
Context
Spectral CT — which captures tissue response at two different x-ray energy levels simultaneously — has been commercially available for more than a decade but historically required either dual-source hardware or manual selection of spectral protocols. Philips’ pitch with Verida is that the detector-based design delivers spectral data on every scan by default, and that AI reconstruction closes the noise and dose gap that has limited spectral adoption.
The clearance caps a months-long US rollout effort following the November 2025 debut. Philips had been selling Verida in CE-marked markets since early 2026, with 510(k) clearance listed as pending in the United States at the time of launch, per the RSNA 2025 announcement. The FDA decision now clears the path for commercial sales in Philips’ largest imaging market, where it competes against Siemens’ photon-counting Naeotom Alpha family and GE HealthCare’s Revolution Apex and Revolution Ascend platforms — all of which featured prominently in RSNA 2025 AI imaging displays.