Home / Clinical Decision Support & Reference / DeepContour

DeepContour

by Wisdom Technologies, Inc. · Based in United States →

Oncology Radiology

Website

Claim this tool

Is this the maker? Claim the company →

Leave this field empty

Correct company / maker

Company website (optional)

Anything else? (optional)

Regulatory Status Disclosed

Overview

DeepContour is an advanced AI-powered automatic CT image organ contour system developed by Wisdom Technologies, Inc. This innovative medical device is designed to streamline and automate the complex process of organ contouring from CT images, which is a critical step in radiotherapy treatment planning. By providing predictive results that can be stored in various formats, DeepContour aims to enhance efficiency and accuracy in clinical workflows. The system is built with a strong emphasis on prioritizing the quality and security of patient information, ensuring that patients receive the safest, highest quality, and most efficient treatment available. DeepContour is part of a suite of radiotherapy-focused products from Wisdom Technologies, Inc., which also includes ArcherQA (Quality Assurance) and DeepPlan (Treatment Planning Systems).

Reviewed by Pouyan Golshani, MD — Interventional Radiologist

Key Features

Automatic CT image organ contouring
AI-powered predictive results
Supports various data storage formats
Streamlines clinical workflows through automation
Prioritizes patient information quality and security
Designed for radiotherapy treatment planning

Use Cases

Automated organ at risk (OAR) contouring in radiotherapy
Enhancing efficiency in radiation treatment planning
Improving accuracy of anatomical structure delineation from CT images
Supporting clinical decision-making in oncology

What Physicians Need to Know

Evidence Base

DeepContour models are trained on large datasets of contoured images, often utilizing custom 3D U-Net deep learning architectures. Clinical contouring guidelines, developed by multidisciplinary panels through literature review and image-based methods, form the basis for expert contours used in training and validation.

Clinical Validation Studies

Multiple studies evaluate DeepContour's effectiveness in radiotherapy, comparing auto-generated contours to expert manual contours using metrics like Dice-Su00f8rensen Coefficient (DSC) and Mean Distance (MD). Significant time savings have been demonstrated, with mean contouring time reductions of approximately 5.9 minutes for prostate and 16.2 minutes for head and neck structures. DeepContour has also shown potential in reducing inter-observer variability for certain organs at risk (OARs).

Override Rate Data

Instead of 'override rate' for alerts, the relevant metric for DeepContour is the rate of manual editing or rejection of auto-generated contours. Studies indicate that while DeepContour significantly reduces initial contouring time, manual editing is often still required. One study reported that 33% of auto-contours needed 'clinically significant' edits, though 65% required only minor adjustments or none.

Clinical Workflow Integration

DeepContour models are designed for seamless integration into existing clinical workflows for radiotherapy planning. This involves transferring AI-generated contours (e.g., DICOM files) into treatment planning systems (TPS) and oncology information systems (OIS). Platforms like deepcOSu00ae facilitate this by integrating AI results into radiology workflows, compatible with PACS, RIS, EHR, EMR, HIS, and CIS, often leveraging APIs for efficient data exchange.

Physician Tip

Physicians should view DeepContour as a powerful assistant for initial contouring, significantly reducing preparation time. However, a critical review and potential manual adjustment of auto-generated contours remain essential, especially for complex or small-volume structures, to ensure clinical accuracy and patient safety. Understanding the model's performance characteristics for different organs and patient anatomies is key to effective utilization. Leverage the time savings to focus on complex cases and optimize treatment plans.

DeepContour's utility is maximized through seamless integration with existing radiology and oncology IT infrastructure, including PACS, RIS, EHR, and various treatment planning systems. API-based integrations are crucial for efficient data flow, enabling real-time access to AI results within the established clinical workflow without requiring clinicians to switch between multiple interfaces. This 'quiet integration' approach is vital for high adoption rates and avoiding workflow disruption.

At a glance

Visit site

Details

Category	Clinical Decision Support & Reference, Oncology AI
Pricing	Unknown
Compliance
BAA Available	Unknown AI-estimated
HIPAA Compliant	Unknown AI-estimated
FDA Status	1 AI-estimated DeepContour received clearance from the FDA on May 8, 2024. It is an automatic CT image organ contour system that provides predictive results, streamlining processes through automation while prioritizing the quality and security of patient information for safe and efficient treatment.
Integrations
EHR	Not specified
Specialties	Oncology, Radiology

Ratings & Reviews

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

Rate DeepContour

Clinical Value

Ease of Use

Integration

Support & Docs

Value for Money

Your Name *

Email *

Your Role

Specialty

Your Review * (min 50 characters)

Press & Coverage

Wisdom Tech

DeepContour received FDA clearance

Wisdom Technologies, Inc. announced on May 8, 2024, that its product, DeepContour, an automatic CT image organ contour system, has received FDA clearance, signifying a major step towards global market penetration.

2024-05

accessdata.fda.gov

DeepContour is a deep learning based medical imaging software that allows trained healthcare professionals to use DeepContour as a tool to automatically process CT images.

DeepContour is a deep learning-based medical imaging software designed to automatically process CT images for trained healthcare professionals, supporting quantitative analysis, organ HU distribution statistics, and transfer of contour files to TPS.

2024-05

Wisdom Tech

Spotlight on Wisdom Tech 's Flagship Products at the 2024 AAPM Annual Meeting( Los Angeles, CA )

Wisdom Tech will showcase its flagship products, including DeepContour, at the 66th AAPM Annual Meeting & Exhibition in Los Angeles from July 21 to 25, 2024, an event that gathers experts in medical physics.

2024-08

DLinRT

AI solution for automatic contour segmentation in radiation therapy planning.

DeepContour is highlighted as an AI solution for automatic contour segmentation in radiation therapy planning, featuring AI-powered segmentation, fast processing, and clinical workflow integration.

unknown

Wisdom Tech

Wisdom Tech Secures Multi-Million-Dollar Investment to Drive Radiotherapy Innovation

Anhui Wisdom Technologies, Inc. completed its third financing round on January 24, 2025, securing a multi-million-dollar investment from Hefei Industrial Investment Group to fuel its innovation and expansion in medical technology.

2025-01

Wisdom Tech

Paper on Monte Carlo Simulations for MRI Guided Proton Radiation Therapy Published in the Medical Physics Journal.

A paper authored by Dr. Shijun Li and Professor George XU on a GPU-based fast Monte Carlo code for proton transport in magnetic fields for radiation therapy was published in the Medical Physics Journal on November 21, 2023.

2023-11

Videos

Product demos, reviews, and walkthroughs for DeepContour.

No videos found. Search YouTube directly

View all on YouTube

Frequently Asked Questions

How does DeepContour integrate into our existing clinical workflow for radiotherapy planning?

DeepContour is designed as an automatic CT image organ contour system that provides predictive results which can be stored in various formats, streamlining the contouring process in radiotherapy planning. It aims to automate the delineation of organs at risk and target volumes, potentially reducing the manual effort and time traditionally spent by clinicians.

What are the primary clinical benefits of using DeepContour for organ contouring in radiotherapy?

The primary benefits include significant time savings in contouring, with studies showing reductions in time for prostate and head and neck structures. DeepContour also aims to improve contouring accuracy and consistency by leveraging deep learning, potentially reducing inter-observer variability.

How accurate is DeepContour, and what are its known limitations or scenarios where manual review is still critical?

DeepContour, utilizing deep active contour networks, has demonstrated effectiveness in precise boundary delineation on public medical image datasets. However, while deep learning auto-segmentation can improve efficiency, studies indicate that a percentage of auto-contours may still require major, clinically significant edits, highlighting the ongoing need for physician review.

What regulatory clearances does DeepContour have, and how does it ensure patient data privacy and security?

DeepContour has received clearance from the FDA, marking a significant step for its global market penetration. As a medical device, it prioritizes the quality and security of patient information to ensure safe and efficient treatment, aligning with general medical device compliance standards.

What are the main alternatives to DeepContour for automated contouring, and how does it compare?

Alternatives to deep learning-based auto-contouring include manual contouring and atlas-based auto-contouring methods. DeepContour and similar deep learning approaches generally outperform manual and atlas-based methods in terms of speed and accuracy, especially with larger datasets, while atlas-based methods can be limited by deformable image registration accuracy and dataset size.

What is involved in implementing DeepContour, and what kind of training is required for our clinical staff?

Implementing DeepContour involves integrating the system into existing workflows, often through DICOM-based tools that route images for automatic outlining and then to the planning system. While the system aims for automation, clinical staff, particularly physicians, would still require training on reviewing and potentially editing the auto-generated contours, and understanding its performance in various clinical scenarios.

What is the typical pricing model for DeepContour, and what is its expected return on investment in terms of time and resource savings?

Specific pricing for DeepContour is not publicly detailed, but the cost of adopting AI solutions in radiology goes beyond the list price and includes procurement, evaluation, and integration. The expected return on investment stems from significant time savings in contouring, which can improve clinic workflow efficiency and potentially reduce resource intensity in treatment planning.