Radiomics App
Overview
The ‘Radiomics App’ referenced in the prompt appears to be an internal project identifier or an early component related to Microsoft’s broader efforts in medical imaging AI, primarily embodied by Project InnerEye. Project InnerEye is a research initiative from Microsoft Health Futures focused on building innovative tools for the automatic, quantitative analysis of three-dimensional medical images. It leverages state-of-the-art machine learning technology, including Convolutional Neural Networks, for tasks such as voxel-wise segmentation of medical images.
The goal of Project InnerEye is to democratize AI for medical image analysis, empowering researchers, hospitals, life science organizations, and healthcare providers to build their own medical imaging AI models using Microsoft Azure. The project has released an open-source deep learning toolkit under an MIT license, making it widely available for the global medical imaging community. This toolkit aims to increase productivity for research and development of best-in-class medical imaging AI, facilitating deployment using Microsoft Azure cloud computing, subject to appropriate regulatory approvals.
While Project InnerEye is a research and open-source endeavor, its foundational work in radiomics and medical imaging AI contributes to Microsoft’s commercial healthcare AI offerings, such as the healthcare AI models available in the Microsoft Azure AI model catalog and Microsoft Dragon Copilot for Radiology, which enhance diagnostic workflows and optimize radiology processes.
Reviewed by Pouyan Golshani, MD — Interventional Radiologist
Key Features
- Automatic quantitative analysis of 3D medical images
- Voxel-wise segmentation using Convolutional Neural Networks
- Open-source deep learning toolkit (MIT license)
- Supports image classification, segmentation, and sequential models
- Configuration-based approach for model building
- Integration with Microsoft Azure for scalable training and deployment
- Feature extraction, preprocessing, and analysis tools (via PyRadiomics integration)
- Supports multiple imaging modalities (CT, MR, OCT, x-ray)
- Peer-reviewed research validation
- Ability to combine imaging and non-imaging data inputs
Use Cases
- Radiotherapy planning workflows
- Quantitative radiology for monitoring tumor progression
- Planning for surgery
- Medical imaging research and development
- Building custom medical imaging AI models
- Accelerating image preparation tasks for clinicians
What Physicians Need to Know
Radiomics offers a powerful way to extract quantitative data from medical images, potentially uncovering disease characteristics not visible to the naked eye. This can aid in enhanced cancer diagnosis, prognosis, prediction of treatment response, and patient stratification. When utilizing radiomics tools, consider their integration capabilities with existing PACS and EMR systems, their compliance with HIPAA and data de-identification standards, and the transparency of their AI models (Explainable AI) to build trust and validate clinical reasoning. Be aware that lack of standardization in image acquisition and feature extraction can impact reproducibility and generalizability of radiomics signatures. Therefore, understanding the underlying methodologies and validation of the radiomics application is crucial for reliable clinical use.
Radiomics applications primarily integrate with Picture Archiving and Communication Systems (PACS) for medical image data (DICOM). Integration with Electronic Medical Records (EMR) systems is evolving, often leveraging FHIR for broader patient data exchange. The ability to convert DICOM metadata into FHIR resources is a key aspect of achieving seamless integration into clinical workflows and research repositories. Tools may also integrate with various programming languages (Python, C++, R, Julia) through SDKs or libraries, allowing for custom development and analysis workflows.
Details
| Category | Developer Tools & APIs, Oncology AI, Radiology & Imaging AI |
| Pricing | Unknown |
| Compliance | |
| BAA Available | Unknown |
| HIPAA Compliant | Unknown |
| FDA Status | No AI-estimated |
| Integrations | |
| EHR | Not specified |
| Specialties | Nuclear Medicine, Oncology, Radiology |
Support & Reliability
FDA clearance history
This product’s FDA authorization. From FDA’s list of AI-enabled medical devices; each number links to FDA’s record.
| Decision date | Number | Name as listed by FDA |
| 2017-12-27 | K173420 510(k) | Radiomics App |
Ratings & Reviews
No reviews yet. Be the first to review this tool!
Rate Radiomics App
Videos
Product demos, reviews, and walkthroughs for Radiomics App.
No videos found. Search YouTube directly




