DELTAVIEW MODEL
Overview
The DELTAVIEW MODEL, originally developed by Riverain Medical Group (now Riverain Technologies), was an early AI-driven medical imaging software. It was specifically designed for the temporal comparison of current and prior chest X-ray images to aid in the detection of subtle changes, such as emerging or enlarging lung nodules that may indicate early-stage lung cancer. The software utilized bone suppression technology to provide a clearer view of soft tissues by electronically removing ribs and clavicles from the radiograph. By aligning and registering two patient images, DELTAVIEW produced a third subtraction image, accentuating differences and making changes more conspicuous for radiologists. This technology was a precursor to Riverain Technologies’ current ClearRead™ suite, which continues to leverage advanced AI and patented suppression technologies to improve diagnostic accuracy and efficiency in cardiothoracic imaging.
Reviewed by Pouyan Golshani, MD — Interventional Radiologist
Key Features
- Temporal comparison of chest X-rays
- Bone suppression technology
- Generation of subtraction images to highlight changes
- Detection of emerging or enlarging lung nodules
- Identification of early-stage lung cancer
- Seamless integration with PACS
- Improved sensitivity for actionable solitary pulmonary nodules
- Aids in detection of soft tissue interval changes
Use Cases
- Early detection of lung cancer
- Monitoring patients with chronic obstructive pulmonary disease (COPD)
- Monitoring patients with chronic pleural thickening from asbestos exposure
- Improving diagnostic accuracy for cardiothoracic conditions
- Enhancing efficiency in radiology workflows
- Reducing missed nodules in chest imaging
What Physicians Need to Know
For optimal performance, use thin-slice CT scans (1.0-1.25 mm) with softer reconstruction kernels, and non-contrast exams. Ensure consistent scan characteristics and avoid axial rotation, gaps in slice spacing, and table height changes. While the system can process less-than-ideal data, adhering to these recommendations will yield the best results. The software is particularly helpful in detecting subtle changes in complex areas like the hilum and apices.
DELTAVIEW MODEL (and the ClearRead Chest suite) is designed for seamless integration with existing PACS, EHR, and reporting systems. It is vendor-neutral and can handle a broad range of acquisition protocols. Recent collaborations, such as with RapidAI, further expand its integration capabilities into enterprise imaging platforms, allowing access to Riverain's solutions through unified workflows.
Details
| Category | Oncology AI, Radiology & Imaging AI |
| Pricing | Unknown |
| Deployment | On-premise (software solution, integrates with PACS) and Cloud-deployable. |
| Compliance | |
| BAA Available | Unknown |
| HIPAA Compliant | Unknown |
| FDA Status | Cleared AI-estimated |
| Integrations | |
| EHR | Not specified |
| Specialties | Oncology, Pulmonology, Radiology |
What the Web Says
Riverain Technologies' DELTAVIEW MODEL, now known as ClearRead +Compare, is a software that electronically compares current and prior chest X-ray images to identify subtle changes that may indicate early-stage lung cancer or other lung diseases. It utilizes bone suppression technology to enhance the visibility of soft tissues by removing ribs and clavicles from the image. The software aims to improve diagnostic accuracy and efficiency for radiologists in detecting and characterizing cardiothoracic diseases.
Overall: PositiveStrengths
- Improves detection of subtle lung cancers and nodules by accentuating changes over time.
- Utilizes bone suppression technology for a clearer view of soft tissues in chest X-rays.
- Reduces radiologist reading times without compromising accuracy.
- Enhances visualization of small pulmonary nodules (≤2 mm).
- Integrates with existing PACS and radiology workflows.
- FDA-cleared and CE marked, used in hundreds of sites worldwide.
Limitations
- May produce false positives or residuals, especially with incomplete fields of view or non-ideal image series.
- Optimal performance relies on ideal image conditions (e.g., thin slices, soft reconstruction kernels, minimal artifacts).
- Concerns about incomplete field of view where vessel suppression may fail in the entire image.
- Techniques are not specifically intended to assist in finding pulmonary nodules in all cases.
- Some general skepticism about AI in radiology regarding false positives and the nuance of diagnosis.
Based on reviews from: blackfordanalysis.com, riveraintech.com, sectra.com, rapidai.com, healthairegister.com, aihealthindex.io, auntminnie.com, indeed.com, builtin.com, parikhradiology.com, ajronline.org, aidoc.com, radiologybusiness.com, fiercepharma.com, ai-imaging.org, reddit.com, theimagingwire.com, youtube.com, pulse2.com, g2.com
Last updated: 2026-10-31
Ratings & Reviews
No reviews yet. Be the first to review this tool!
Rate DELTAVIEW MODEL
Press & Coverage
Videos
Product demos, reviews, and walkthroughs for DELTAVIEW MODEL.



