RIVANNA 的 BoneEnhance AI 模組獲 FDA 510(k) 許可，可用於 Accuro XV

RIVANNA announced on October 22, 2026, that the U.S. Food and Drug Administration granted 510(k) clearance for BoneEnhance, a machine learning software module for the company’s Accuro XV diagnostic ultrasound system. The module segments bone and surrounding tissues and supports visualization of skeletal anatomy in three dimensions during image review.

The Charlottesville, Virginia-based company, which develops AI-enabled clinical decision-support systems, described BoneEnhance as the first machine learning software module cleared for the Accuro XV platform. RIVANNA framed the clearance as moving the platform from acquisition toward interpretation: the Accuro XV performs the volumetric image acquisition, and BoneEnhance labels the anatomy captured within it. The underlying system was cleared in April 2026 for B-mode imaging in musculoskeletal conventional and superficial applications, according to the release; RIVANNA publicly announced that earlier clearance on May 7, 2026.

How BoneEnhance Works on the Accuro XV

Each automated Accuro XV acquisition produces a volumetric dataset for review. Within that volume, BoneEnhance identifies bone and surrounding tissues, then marks them with color-coded overlays on the two-dimensional B-mode image. The overlays leave the original anatomical data unchanged and visible underneath, and an on-screen three-dimensional indicator shows the scan’s orientation and where the displayed slice sits within the acquired volume.

Co-founder and CEO Will Mauldin said RIVANNA was founded to address a problem that cuts across care settings: decision-critical information sitting outside the point-of-care workflow. “BoneEnhance is where the stack begins to close that gap for musculoskeletal care: automated volumetric acquisition, now paired with cleared software-assisted review,” Mauldin said, calling the clearance the foundation for decision-support capabilities now in development.

The cart-based Accuro XV uses a conformable three-dimensional probe with motorized linear translation to acquire B-mode images across a 10-centimeter scan extent, automating large field-of-view acquisition after initial patient positioning. A compliant polyurethane stand-off medium forms a patient contact surface that conforms to irregular anatomies to improve acoustic coupling. The system includes a touchscreen interface, an integrated battery pack for point-of-care mobility, and DICOM-compatible image archival workflows.

The system’s earlier clearance, announced on May 7, 2026, authorized commercial use of the Accuro XV for musculoskeletal imaging by qualified and trained healthcare professionals in hospital and medical clinic environments. RIVANNA says the integrated workflow is backed by a worldwide intellectual-property portfolio of more than 100 issued and pending patents across more than 20 patent families.

Validation Across Eight Sites

BoneEnhance was trained and validated within RIVANNA’s nationwide eight-site clinical study. Three sites contributed the training data, and the other five provided an independent test set of 233 volumetric datasets comprising 69,900 images from 173 subjects. According to the release, none of the test data were used for training or tuning. Three board-certified radiologists established the reference labels, and RIVANNA reported that per-volume label prediction accuracy, a measure of anatomical labeling, ranged from 95.1% to 100.0%.

Separate from the validation supporting the BoneEnhance clearance, the company cited peer-reviewed evidence for the platform’s automated acquisition. In a study of 205 patients published in the Journal of Emergency Medicine, operators with one hour of hands-on training acquired wrist and ankle images rated adequate for diagnostic interpretation in more than 90% of scans, with image quality comparable to physician-acquired scans. RIVANNA’s earlier clearance announcement states that the study was conducted at UVA Health and UT Southwestern Medical Center and that it compared non-physician operators with board-certified emergency physicians.

Study Growth and the Injury Detection Module

The clinical study behind BoneEnhance has continued to grow. RIVANNA announced in April 2026 that the study had expanded from two sites to eight — Denver Health, Texas Tech Health El Paso, the University of California, San Francisco, the University of Utah, UVA Health, UT Southwestern Medical Center, Virginia Commonwealth University, and Yale University — after the Biomedical Advanced Research and Development Authority (BARDA) executed a funding option. At that stage, the program was training CADe/x, which RIVANNA describes as the first computer-aided detection algorithm for automated fracture identification derived from volumetric ultrasound imaging, and completing validation of BoneEnhance.

According to the new release, the study has since expanded from eight to ten sites and is building the datasets used to train an investigational musculoskeletal injury detection module for the Accuro XV. Where BoneEnhance labels anatomy, the investigational module is being developed to assist clinicians in evaluating potential extremity fractures within acquired ultrasound images. The release states that because the sites and data infrastructure were established once and are reused for each new module, every subsequent module builds on the program’s federal investment.

Delphine Le Roux, RIVANNA’s senior director of market access and strategic partnerships, credited the participating clinical sites and BARDA’s sustained support for the evidence program behind the clearance. “The infrastructure, partnerships, and regulatory playbook we established for BoneEnhance are designed to carry forward,” she said.

The project has been supported in whole or in part with federal funds from the U.S. Department of Health and Human Services, the Administration for Strategic Preparedness and Response, and BARDA, under contract number 75A50121C00035.