Spectral AI (NASDAQ:MDAI) executives said fiscal 2025 marked a “pivotal year” as the company advanced its DeepView burn diagnostic system toward potential FDA clearance and strengthened its funding profile, highlighted by a new award under its longstanding BARDA partnership.
DeepView FDA path and clinical data
Chief Executive Officer Vincent Capone said the company’s June 2025 De Novo submission to the U.S. Food and Drug Administration was a key milestone. Following the submission, Spectral AI received an FDA “additional information” notification letter and responded “timely and completely” earlier this month, Capone said. Management said it remains in active dialogue with the agency and that FDA feedback has been consistent with expectations. Capone added the company is hopeful for a positive FDA response before the end of the second quarter of 2026.
If the device receives FDA clearance, Spectral AI plans to initiate an outcomes study aimed at measuring DeepView’s real-world impact in hospitals, including effects on patient journeys and clinician workflows.
BARDA contract expansion and planned enhancements
Spectral AI emphasized its relationship with the Biomedical Advanced Research and Development Authority, which dates back to 2013. Capone pointed to the September 2023 Project BioShield contract with a value “up to $150 million,” including $55 million awarded at signing as part of the base phase. The company said BARDA support has funded product development as well as U.S.-based studies validating DeepView and its AI algorithm in emergency, trauma, and burn settings.
Capone said BARDA recently awarded Spectral AI $31.7 million in advanced funding to accelerate and support additional feature development for DeepView. In connection with that award, Spectral AI committed to contribute an additional $9.7 million toward the overall development costs of those advancements.
Management said accelerating the second phase of the BARDA contract is expected to speed development work, including:
- Total body surface area (TBSA) measurement tools
- Electronic health record (EHR) integration
During the Q&A session, Capone added the company expects to pursue label expansion work and described feature refinements informed by user feedback, including improving the battery and user interface. He also said the company intends to begin economic outcomes studies “sooner rather than later.”
Capone noted that, upon FDA approval, BARDA provisions would allow the agency to subsidize initial sale and distribution of up to 30 DeepView systems in burn centers in key U.S. regions, with an additional subsidy mechanism for up to 140 more systems in burn centers and Level 1 trauma emergency departments.
Asked how BARDA deployments might translate into commercial activity, Capone said health systems will determine whether they treat access to the device as a purchase or lease, which would affect revenue recognition. He said Spectral AI envisions an opportunity to bundle multi-year arrangements that include software improvements, software licensing, and device maintenance, but added he does not expect the initial placements to meaningfully increase 2026 revenue, with greater impact anticipated in 2027 and 2028 as installations “stack.”
On contract timing, Capone said the BARDA contract extends until 2030, with the current acceleration running mostly through 2028.
International plans and U.K. authorization
Capone said systems already placed in the United Kingdom have generated positive user feedback. He reiterated the company obtained UKCA authorization for the burn indication in 2024. Following a positive FDA determination on the De Novo submission, Spectral AI plans to update its U.K. authorization in 2026 to reflect the improved system submitted to FDA.
Capone said the company anticipates initial sales in the U.K., Australia, or Gulf Cooperation Council nations could begin in late 2026 after the expanded UKCA authorization, while noting that international revenue models may differ due to how various health systems procure devices, including in nationalized systems.
Handheld device development under DoD-related contract
Capone also discussed development of a handheld device under a Department of Defense contract through the MTEC consortium. He said the company signed a no-cost extension of the current Phase II contract through June 30, 2026, and anticipates delivering a fully functioning prototype by the end of the second quarter of 2026. After that, management will assess whether additional scope items can be completed before the end of the third quarter of 2026.
Capone said Phase II is scheduled to be reviewed and awarded by the end of the fourth quarter of 2026, and the company is hopeful it will be asked to bid on the next work. He added Spectral AI expects to leverage any FDA approvals for the cart-based DeepView system as a predicate for a future 510(k) submission for the handheld version.
Financial results and 2026 revenue outlook
Corporate Controller Thomas Spieth reported fourth-quarter 2025 research and development revenue of $3.8 million, down from $7.6 million in the prior-year quarter, reflecting what management described as an anticipated reduction in reimbursed costs as the company neared completion of the base phase of the BARDA Project BioShield contract. Fourth-quarter gross margin was 39.8% versus 44.0% a year earlier, which Spieth attributed primarily to a lower percentage of reimbursed direct labor in the revenue mix.
Fourth-quarter general and administrative expense was $4.0 million, down from $4.5 million, driven by lower spending on third-party accounting and legal providers. Net income for the quarter was $0.6 million, or $0.02 per diluted share, compared with a net loss of $7.7 million, or $(0.41) per diluted share, in Q4 2024. Spieth said Q4 2025 results included a $4.0 million gain from changes in the fair value of the company’s warrant liability.
For the full year, Spectral AI reported R&D revenue of $19.7 million, down from $29.6 million, which Spieth linked to reduced reimbursed costs under the BARDA contract following the De Novo submission. Full-year gross margin was 45.4%, compared with 44.9% in 2024. G&A expense declined to $17.5 million from $19.9 million, which Spieth said reflected continued operating efficiencies.
The company’s net loss for 2025 was $7.6 million, or $(0.29) per diluted share, compared with a net loss of $15.3 million, or $(0.85) per diluted share, in 2024. Spieth cited changes in the warrant liability valuation, reduced borrowing-related costs, net amortization of debt discount, and improved operating efficiencies as primary drivers of the year-over-year improvement.
As of December 31, 2025, the company reported $15.4 million in cash, up from $5.2 million at the end of 2024, reflecting debt and equity financings and warrant and stock option exercises. Total debt was $8.5 million, and shares outstanding were approximately 30.7 million. Capone said the company entered 2026 in its strongest liquidity position in recent history.
Looking ahead, Capone provided 2026 revenue guidance of approximately $18.5 million, which he said includes the effect of the new BARDA funding. He added the forecast does not include any significant contributions from sales of the VPU system.
On commercialization preparations, Capone said Spectral AI has begun searching for a chief commercial officer, engaged Deloitte Consulting to support commercialization strategy in the U.S. and overseas, and plans to expand its sales team. He also said the company intends to expand its biomedical engineering staff to support training for new installations, though he did not provide a specific headcount target.
About Spectral AI (NASDAQ:MDAI)
Spectral AI, Inc (NASDAQ: MDAI) is a technology company focused on delivering advanced analytics and insights through the application of machine learning to multi-spectral and hyperspectral data. Its core platform ingests imagery from satellites, aerial drones and ground-based sensors, applying proprietary algorithms to identify patterns and anomalies invisible to the naked eye. The company’s solutions are designed to help clients make more informed decisions in areas such as agriculture, environmental monitoring, infrastructure inspection and resource exploration.
The company offers a cloud-native software-as-a-service (SaaS) platform that enables users to visualize and analyze large volumes of spectral data via customizable dashboards.
