AI-Rad Companion Prostate MR lifted readers' accuracy slightly

In the maker's trial that China's CMDE reviewed, doctors reading prostate MRI with AI-Rad Companion Prostate MR reached an AUROC of 0.7010, against 0.6758 without it.

  • Not rated an organization

    Doctors read stored images in the maker's pre-market trial, the gain in AUROC was small and no patient's care depended on the software.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Intended benefit.
Evidence

foundstudiedreportedclaimed

An independent regulatory document supports the benefit.

Compared with

Doctors reading the same cases without the software. Measured.

Without the system
AUROC 0.6758
With it
AUROC 0.7010
The system’s part
The trial compared doctors reading with and without the software, so the small gain in AUROC is attributed to the software. The trial was the maker's own. The report says that the product uses deep learning.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
AI-Rad Companion Prostate MR (前列腺癌磁共振图像辅助检测软件)
Made by
Siemens Healthineers, with 上海西门子医疗器械有限公司 (Siemens Shanghai Medical Equipment Ltd.) as the applicant in China
Used by
上海西门子医疗器械有限公司 (Siemens Shanghai Medical Equipment Ltd.), the applicant (retrospective reader trial with 12 doctors and 340 subjects)
Where
Trial sites that the report does not name, for a registration in China
Stage
Tested in a laboratory
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Identifying Objects, Actions, and Events
Collections
Scrutiny and findings, Cross-border records
Last revised
1 November 2026

Sources

  1. Regulatory document, by authors with no tie to the system.

    Center for Medical Device Evaluation (CMDE), National Medical Products Administration, China. 医疗器械产品注册技术审评报告：前列腺癌磁共振图像辅助检测软件 (technical review report for the registration of prostate cancer MR image detection-aid software), acceptance no. CQZ2401376, applicant 上海西门子医疗器械有限公司. 29 October 2025. Section 三、临床评价概述 (clinical evaluation).

    主要有效性评价病例水平前列腺病变诊断的 ROC 曲线下面积 AUROC：试验组为 0.7010，对照组为 0.6758，两组之差为0.0252，95%置信区间为（0.0011，0.0493），95%置信区间下限大于优效界值
    经综合评价，在目前认知水平上，认为该产品上市带来的受益大于风险，综合剩余风险可接受。

  2. Other, by the makers. Context or contrary evidence.

    Siemens Healthineers. AI-Rad Companion: AI-powered radiology workflow solutions (product page), section AI-Rad Companion Prostate MR. Read on 1 November 2026.

    AI‑Rad Companion Prostate MR delivers AI‑powered analysis of multiparametric prostate MR images with automated lesion detection, segmentation, and PI‑RADS scoring.

Details

AI system

Reported configuration

Release VA11 (full version VA11B), algorithm release VA1 (full version VA10A)

Workflow

The software analyzed prostate MR images with deep learning and marked suspected adenocarcinoma lesions. Twelve doctors read 340 retrospective cases with and without the software.

Follow-up

Conditions for replication

Patients are untreated adults aged 40 or over. Trained doctors make the diagnosis and do not base a clinical decision on the software alone.

Entry details

Sector

Healthcare.

Purpose

Siemens Healthineers built AI-Rad Companion Prostate MR to help trained doctors detect suspected prostate adenocarcinoma on MR images of untreated adults aged 40 and over.

29 more questions not yet checked

model, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). AI-Rad Companion Prostate MR lifted readers' accuracy slightly (AIBD-0959). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-0959.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)

BibTeX Data Field definitions