FDA Grants 510(k) Clearance to DeepHealth's AI-Powered Breast Ultrasound Analysis Platform

DeepHealth, a subsidiary of RadNet, has secured FDA 510(k) clearance for its AI-powered Breast Ultrasound platform. The software automates lesion detection and reporting, achieving over 98% localization accuracy and boosting cancer detection sensitivity by 8% in clinical studies. It also reduces radiologist interpretation times by 37%. RadNet plans to deploy the platform across its national outpatient network by the end of 2026, targeting over 700,000 annual exams under an existing Category III CPT reimbursement code.

FDA 510(k) clearance for DeepHealth

• ▪The FDA-cleared DeepHealth Breast Ultrasound platform is designed to automate lesion detection, characterize findings according to ACR BI-RADS standards, and generate structured draft reports.
• ▪RadNet, Inc. acquired the technology underlying the DeepHealth Breast Ultrasound platform when it purchased the AI ultrasound firm See-Mode Technologies in 2025.
• ▪DeepHealth, Inc., a wholly owned subsidiary of RadNet, Inc., received FDA 510(k) clearance for its AI-powered DeepHealth Breast Ultrasound platform.

AI platform clinical performance metrics

• ▪The clinical validation of the DeepHealth Breast Ultrasound platform was supported by a multi-reader, multi-case study involving 16 board-certified U.S. radiologists.
• ▪The DeepHealth Breast Ultrasound platform demonstrated an 8 percent increase in overall cancer detection sensitivity during clinical validation.
• ▪Clinical validation of the DeepHealth Breast Ultrasound platform demonstrated a lesion localization accuracy exceeding 98 percent.

Radiologist workflow efficiency gains

• ▪The DeepHealth Breast Ultrasound platform reduced radiologist interpretation time by 37 percent during clinical validation studies.
• ▪The DeepHealth Breast Ultrasound platform automatically populates measurements and findings into draft reports to eliminate manual data entry for sonographers.

RadNet national deployment plans

• ▪RadNet, Inc. plans to deploy the DeepHealth Breast Ultrasound platform across its national outpatient imaging network by the end of 2026.
• ▪The national deployment of the DeepHealth Breast Ultrasound platform is expected to support more than 700,000 annual breast ultrasound examinations across RadNet's network.

CPT code reimbursement pathway

• ▪Category III CPT codes are temporary codes that can be difficult to convert into consistent, widespread payer coverage from insurance companies.
• ▪U.S. healthcare providers can seek reimbursement for using the DeepHealth Breast Ultrasound platform under an existing Category III CPT code for quantitative ultrasound tissue characterization.

DeepHealth OS modular breast suite

• ▪The DeepHealth Breast Ultrasound platform integrates into DeepHealth OS, a cloud-native operating environment that includes AI tools for mammography detection and breast density assessment.
• ▪DeepHealth and HOPPR entered a partnership in September 2024 to commercialize a medical-grade generalized foundation model to support fine-tuned AI models for cancer detection.
• ▪DeepHealth's broader AI portfolio includes tools for breast arterial calcification assessment, breast cancer risk prediction, and operational analytics.