Dose+ (1.0)

Dose+ is a software-only medical device that uses machine learning algorithms to automatically generate 3D radiation dose distributions for prostate cancer patients undergoing external beam radiation therapy. It is intended for use by qualified radiation therapy professionals (medical physicists, oncologists, dosimetrists) to provide personalized organ-at-risk dose optimization during treatment planning, and outputs must be reviewed in a treatment planning system before clinical use.

Dose+ uses locked machine learning models to predict complete 3D dose distributions from patient CT images and dose prescriptions, outputting DICOM RT Dose objects. The predicate (Oncospace) instead predicts achievable dosimetric objectives for organs-at-risk. Both are cloud or local client-server software with DICOM-RT compliance, neither replaces a full treatment planning system, and both use machine learning on patient-specific anatomy.

Not stated in this summary.

Both devices share the same product code (MUJ), same device classification, same intended users, and provide dosimetric guidance for external beam radiotherapy treatment planning to the same patient population. Although Dose+ outputs complete 3D dose distributions while the predicate outputs OAR objectives, the underlying processing fundamentals are identical (locked ML models trained on patient anatomy). Performance verification on independent US datasets demonstrated non-inferiority in OAR mean dose predictions and target coverage metrics, with clinical validation at 4 US institutions showing equivalent plan quality and significant reduction in optimization iterations, with no safety hazards identified.