TAIMedImg DeepMets

by Taiwan Medical Imaging Co.  · Based in Taiwan →Trustworthy AI for Healthcare, Powered by Federated Technology.
Neurology Oncology Radiology

Contact for pricing.
Regulatory Status Disclosed

Overview

TAIMedImg DeepMets is an AI-aided brain metastasis supportive detection system, specifically DeepMets®-Plus, designed to assist trained medical professionals. It provides initial object contours on axial T1-weighted contrast-enhanced (T1WI+C) brain magnetic resonance (MR) images to accelerate the workflow for radiation therapy treatment planning. The software is intended for patients with known (imaging diagnosed) brain metastases (BM) and is specifically for lesions with a diameter of ≥ 10 mm. Utilizing deep learning neural networks, DeepMets generates contours and annotations for diagnosed brain metastases and converts them into standard DICOM-RTSTRUCT files, ensuring compatibility with radiotherapy treatment planning systems. This innovation significantly reduces the time physicians require to review medical images from approximately 10 minutes to just 30 seconds and can shorten the patient’s overall medical service process from two weeks to half a day. Furthermore, the platform incorporates federated learning and validation mechanisms to protect patient privacy while enabling collaborative AI model training across multiple institutions.

Reviewed by Pouyan Golshani, MD — Interventional Radiologist

Key Features

  • AI-aided brain metastasis detection and contouring on MRI
  • Automated segmentation for Gross Tumor Volume (GTV) contours
  • Accelerates workflow for radiation therapy treatment planning
  • Generates reports in PDF and DICOM-RTSTRUCT (RTSS) format
  • Federated Learning for privacy-preserving AI model training across institutions
  • High sensitivity (85% for 4mm tumors, 97% for 10mm tumors)
  • Reduces image review time from 10 minutes to 30 seconds
  • Supports adult patients with known brain metastases (lesions u2265 10mm)
  • Explainable, Traceable, Verifiable, and Auditable AI
  • AI Agents with generative capabilities for report drafting and clinical decision assistance

Use Cases

  • Assisting radiologists and clinicians in detecting brain metastases on MRI
  • Accelerating radiation therapy treatment planning
  • Improving workflow efficiency in radiosurgery procedures
  • Generating diagnostic reports and assisting with clinical decisions
  • Prognosis prediction and treatment tracking
  • Collaborative development and validation of cross-center AI models while safeguarding privacy

What Physicians Need to Know

Key Capabilities
Automated segmentation of Gross Tumor Volume (GTV) contours for brain metastases on axial T1-weighted contrast-enhanced (T1WI+C) brain MR images. Accelerates radiation therapy treatment planning workflow. Identifies and contours brain metastases (4mm to 40mm) within minutes, generating PDF and RTSS reports. Offers AI Agents for automatic diagnostic report generation and clinical decision assistance.
Clinical Utility
Assists trained medical professionals in radiation therapy planning, reducing physician burden and enhancing care quality. Improves workflow efficiency, leading to faster, more cost-effective, and accurate treatment decisions for patients with known brain metastases (u226510mm lesions). Shortens detection time for brain tumors, providing results in approximately five minutes.
Integration Options
Comprises an AI inference module and a DICOM Radiotherapy Structure Sets (RTSS) converter module. Requires external platforms supporting DICOM-RT viewing/editing for contour finalization. Deployed globally via Microsoft Azure and integrates with MRI systems. Utilizes Federated Learning (FL) and Federated Validation (FV) for secure, on-premise data utilization and collaborative AI model training across institutions.
Compliance Status
Passed safety and effectiveness requirements, demonstrating substantial equivalence to predicate devices (VBrain K203235). Regulated under 21CFR 892.2050, Classification II, Product Code QKB. Certified as a Software as a Medical Device (SaMD) by Taiwan's TFDA in 2022, and has received official approval in Vietnam. Compliant with ISO 13485:2016 and certified by international partners including the EU's MDR and the US CDC. Federated learning ensures GDPR compliance.
Pricing Model
Specific pricing details are not publicly available, but the mention of 'seamless subscription and deployment' and 'increased cost-effectiveness' suggests a commercial, likely subscription-based, model.
User Experience
Designed for trained medical professionals, requiring thorough review and finalization of AI-generated contours. Aims to reduce procedural time and stress for clinicians. The underlying federated platform suggests a user-friendly interface for monitoring and reporting.
Support Quality
Taiwan Medical Imaging Co., Ltd. is dedicated to providing medical teams with validated AI medical imaging detection and predictive models, offering a complete usage system. Specific details on support channels or service level agreements are not publicly available.
Implementation Complexity
Involves a distributed federated learning approach, allowing data to remain on-premise within hospitals, which helps with privacy and data sovereignty. Requires integration with existing DICOM-RT compatible systems for contour review and modification.
Evidence Base
Performance testing confirmed safety and effectiveness, demonstrating substantial equivalence to predicate devices. Clinical testing involved 158 cases from 16 US MRI sources. The AI model was trained on a retrospective dataset of 1,029 brain metastasis patients from a major Taiwanese medical center, with further tuning using 559 patients from a nationwide database. Achieves high sensitivity (85% for 4mm tumors, 97% for 10mm tumors). Jointly developed by Taiwan AI Labs and Taipei Veterans General Hospital.
Physician Tip

TAIMedImg DeepMets is an assistive AI tool designed to streamline radiation therapy planning for brain metastases. Physicians must always thoroughly review and finalize the AI-generated contours on T1WI+C brain MR images, especially for lesions u226510mm. This tool can significantly reduce diagnostic turnaround time and enhance workflow efficiency, allowing for faster treatment decisions. Remember, the physician retains ultimate responsibility for diagnosis and treatment.

DeepMets outputs in DICOM RTSS format, ensuring compatibility with standard image visualization software and treatment planning systems. Its federated learning architecture allows for secure, on-premise data processing and collaborative model improvement across institutions without sharing raw patient data, aligning with strict privacy regulations like GDPR. Deployment via Microsoft Azure further enhances its accessibility and integration potential within existing healthcare IT infrastructures.

Details

Category Oncology AI
Pricing Contact for pricing.
DeploymentGlobally via Microsoft Azure; Federated platform supporting on-premise data processing with shared models.
Compliance
BAA AvailableUnknown AI-estimated
HIPAA Compliant Yes AI-estimated
FDA Status Yes AI-estimated

TAIMedImg DeepMets has received TFDA (Taiwan FDA) approval as a Software as a Medical Device (SaMD) since September 2022. It also obtained Philippine FDA certification in November 2024. The device has a US FDA 510(k) Premarket Notification (K250427) with a submission date of May 28, 2025, indicating its intent to assist trained medical professionals by providing initial object contours on axial T1-weighted contrast-enhanced brain MR images to accelerate workflow for radiation therapy treatment planning for known brain metastases with lesions ≥ 10 mm.

Integrations
EHR Not specified
Specialties Neurology, Oncology, Radiology

What the Web Says

TAIMedImg DeepMets is an AI-powered software designed to assist medical professionals in contouring images and automating the segmentation of Gross Tumor Volume (GTV) for brain metastases. It is intended as an adjunctive tool to complement, rather than replace, manual contouring processes, with physicians retaining ultimate responsibility for diagnosis and treatment decisions. The software has undergone performance testing and demonstrated safety and effectiveness, performing comparably to predicate devices.

Overall: Positive

Strengths

  • Automated segmentation of GTV for brain metastases.
  • Adjunctive tool to assist medical professionals.
  • Demonstrated safety and effectiveness in performance testing.
  • Performs comparably to predicate devices.
  • Aids in streamlining the contouring process.

Limitations

  • Not intended to replace manual contouring; requires thorough review by a medical professional.
  • Physician retains ultimate responsibility for diagnosis and treatment.
  • Only valid for DICOM images of adult patients.
  • Exclusions for patients with prior craniotomy, non-BM brain tumors, or excessive motion artifacts.

Based on reviews from: accessdata.fda.gov

Last updated: 2026-07-20

Ratings & Reviews

No reviews yet. Be the first to review this tool!

Rate TAIMedImg DeepMets

Clinical Value
Ease of Use
Integration
Support & Docs
Value for Money

Videos

Product demos, reviews, and walkthroughs for TAIMedImg DeepMets.

No videos found. Search YouTube directly

View all on YouTube

Frequently Asked Questions

TAIMedImg DeepMets is an AI-powered software designed to assist trained medical professionals by providing initial object contours on axial T1-weighted contrast-enhanced brain MR images. Its main goal is to accelerate the workflow for radiation therapy treatment planning for adult patients with known brain metastases (BM) with lesions 10 mm or larger.
TAIMedImg DeepMets has received certification from Taiwan's Food and Drug Administration (TFDA) as a Software as a Medical Device (SaMD) and is compliant with the ISO 13485:2016 international quality standard for medical devices. It is also designed with federated learning, aligning with international regulations like the GDPR and the EU's AI Act to ensure data privacy.
DeepMets is specifically intended for brain metastases with lesions 10 mm or larger and is not for other brain tumors or body parts. It excludes patients with prior craniotomy, clinical imaging diagnosis of brain tumors other than BM, or images with excessive motion artifacts.
TAIMedImg DeepMets utilizes a Federated Learning (FL) approach, which allows AI models to be trained securely across multiple institutions while keeping raw patient data on-premise within each hospital. This method protects patient privacy by sharing only model updates rather than sensitive patient information.
Yes, alternatives exist, such as VBrain, a predicate device that also contours meningiomas and acoustic neuromas in addition to brain metastases. Other AI tools like Robovision BrainMets AI also focus on brain metastasis detection, with some claiming detection of smaller lesions (e.g., sub-3 mm) than DeepMets' specified 10 mm minimum.
The physician retains ultimate responsibility for making the final diagnosis and treatment decisions. DeepMets is an assistive tool, and all automatically generated contours must be thoroughly reviewed, confirmed, or modified by a trained medical professional using external DICOM-RT viewing/editing platforms before any therapy or treatment is delivered.
While specific pricing for TAIMedImg DeepMets is not publicly available, the technology is anticipated to offer benefits such as faster clinical workflows and increased cost-effectiveness of procedures. By accelerating contouring for radiation therapy planning, it aims to reduce the burden on physicians and enhance overall efficiency.

Related Tools

FAITH project (Fatigue Therapy – AI-supported Diagnosis and Therapy of Tumour-associated Fatigue Syndrome)
Fimo Health GmbH (Consortium Leader)
Mental & Behavioral Health AI
The FAITH project is developing an AI-based solution for the diagnosis and therapy of tumor-associated fatigue syndrome in cancer patients. It uses wearable sensors, a smartphone app, and artificial intelligence to provide individualized therapy recommendations and monitor mental health.
AI-based support system for skin cancer diagnostics
German Cancer Research Center (DKFZ)
Clinical Decision Support & Reference
Scientists at the German Cancer Research Center have developed an AI-based support system for skin cancer diagnostics that explains its decisions, increasing doctors' confidence in both the AI and their own diagnoses.
eyonis® LCS
Median Technologies
Oncology AI
eyonis® LCS is an FDA 510(k) cleared AI/machine learning-powered software for the detection and diagnosis of lung cancer, helping clinicians detect and characterize suspicious parenchymal pulmonary nodules earlier on low-dose CT scans.
Lumea Digital Pathology
Lumea
Clinical Decision Support & Reference
Lumea is a specialty-specific digital pathology platform that covers the entire diagnostic workflow, from biopsy to final report, and integrates AI tools from partners like Paige and Mindpeak.
Azra AI
Azra AI
Clinical Decision Support & Reference
Azra AI is an AI platform that swiftly analyzes, identifies, and classifies cancer patients, triaging those needing therapy and follow-up to enhance oncology care. It is deployed at more than 1,500 sites and integrated with leading EHRs.
Trial Library
Trial Library
Drug Discovery & Research
Trial Library's AI-native platform is embedded in community oncology practices to identify eligible patients for oncology clinical trials. It integrates clinical trials into routine care, activating providers, identifying eligible patients, and navigating them through the trial lifecycle across health systems, biopharma, and payers.

See all Oncology AI tools →

Suggest an Edit → | Last Verified: 2026-04-19 | First Added: 2026-04-19
AI Tool Finder
AI-powered search. Results may not be comprehensive.