SmartChest

by Milvue  · Based in France →AI for Medical Imaging: Optimizing Emergency Radiology Workflows and Patient Management
Critical Care Emergency Medicine Radiology

Subscription
Regulatory Status Disclosed

Overview

SmartChest by Milvue is a radiological computer-assisted triage and notification software designed to analyze frontal chest X-ray images (Postero-Anterior (PA) or Antero-Posterior (AP)). It targets transitional adolescents (18-21 years old, treated as adults) and adults (≥22 years old) for the presence of suspected pleural effusion and/or pneumothorax. The software utilizes an artificial intelligence algorithm to identify features suggestive of critical findings and provides case-level output. This output is made available to a PACS (Picture Archiving and Communication System) or other DICOM storage platforms for worklist prioritization. SmartChest functions as a passive notification tool for prioritization only within the standard of care workflow and does not send proactive alerts directly to a medical specialist. Its results are not intended to direct attention to a specific portion of an image or to be used on a stand-alone basis for clinical decision-making.

Reviewed by Pouyan Golshani, MD — Interventional Radiologist

Key Features

  • Radiological computer-assisted triage and notification
  • Analyzes frontal chest X-ray images (PA/AP)
  • Detects suspected pleural effusion
  • Detects suspected pneumothorax
  • Provides case-level output for PACS worklist prioritization
  • Utilizes AI algorithms for critical findings detection
  • Passive notification for prioritization-only
  • Cloud-based, hybrid, or locally virtualized deployment options
  • Automatic analysis triggered after image acquisition
  • Processing time of 1-10 minutes

Use Cases

  • Prioritizing chest X-ray interpretations in emergency departments
  • Streamlining radiology workflow for faster review of critical cases
  • Reducing diagnostic errors in emergency settings
  • Improving patient management by reducing waiting times for critical diagnoses
  • Supporting radiologists and emergency physicians in interpreting chest X-rays
  • Automating initial screening of chest X-rays for specific pathologies

What Physicians Need to Know

DICOM Support & Standards
Milvue products, including SmartChest, adhere to major interoperability standards such as DICOM and HL7. SmartChest receives chest radiographs from PACS or other DICOM storage platforms and sends analysis results as a newly generated DICOM series (secondary capture image) with specific DICOM tags for worklist prioritization.
PACS Integration Method
SmartChest integrates into standard reading environments (PACS) and other DICOM storage platforms. It provides case-level output for worklist prioritization by passively sending notifications to worklist software. Deployment options include cloud-based, hybrid solutions, or locally virtualized environments (virtual machine, Docker). Milvue actively participates in IHE Connectathons to ensure integration.
Reading Room Workflow Impact
SmartChest acts as a radiological computer-assisted triage and notification software, facilitating the prioritization of urgent chest X-ray images for review. It aims to optimize workflow, reduce waiting times, and potentially decrease diagnostic errors by providing immediate access to AI-enhanced images with annotated anomalies. It is a passive notification tool for prioritization and not intended for standalone clinical decision-making.
AI Model Architecture (deep learning approach)
SmartChest utilizes advanced artificial intelligence algorithms based on deep learning to analyze frontal chest X-ray images and detect features suggestive of critical findings like pleural effusion and pneumothorax.
Processing Speed (per study)
The mean execution time for pneumothorax detection is approximately 2.3 seconds on a local server and 28.5 seconds on a cloud server. For pleural effusion, it's about 2.3 seconds on a local server and 28.3 seconds on a cloud server. Triage alerts are typically generated in less than 30 seconds.
FDA Clearance Pathway (510k/De Novo)
SmartChest received 510(k) premarket notification clearance from the FDA. It was determined to be substantially equivalent to a legally marketed predicate device.
Supported Modalities (CT/MRI/X-ray/US)
SmartChest analyzes frontal chest X-ray images (Postero-Anterior (PA) or Antero-Posterior (AP)) for the presence of suspected pleural effusion and/or pneumothorax.
Sensitivity & Specificity Data
For pneumothorax, SmartChest demonstrated an ROC AUC of 0.989, sensitivity of 92.7% (95% CI: 87.4-96.2), and specificity of 97.3% (95% CI: 93.4-99.1). For pleural effusion, the ROC AUC was 0.975, sensitivity 93.3% (95% CI: 88.1-96.4), and specificity 90.0% (95% CI: 84.1-94.1).
RSNA/ACR Validation
Milvue participates in major radiology events like RSNA (e.g., RSNA 2025 partnership). A French Multicentre ED Study (RSNA 2022) for Milvue Suite (SmartUrgences), which shares underlying technology with SmartChest, reported 96% sensitivity and 85% specificity for fracture detection. Milvue has also been included in scientific publications in Radiology evaluating commercially available chest radiograph AI tools.
Physician Tip

SmartChest is a valuable tool for prioritizing urgent chest X-ray cases, but it is crucial to remember that it is a triage and notification system, not a standalone diagnostic solution. Always perform a comprehensive interpretation of all images, regardless of AI flagging, as the final diagnosis rests with the physician. The AI provides case-level output and does not highlight specific areas on an image. Leverage the AI-enhanced images with annotated findings to streamline your workflow and focus on critical cases, potentially reducing diagnostic errors and patient waiting times.

SmartChest offers flexible integration into existing radiology workflows, supporting standard DICOM and HL7 protocols. It can connect directly with PACS and other DICOM storage platforms, delivering results as a new DICOM series for seamless worklist prioritization. Deployment options include cloud, hybrid, or on-premise virtualized solutions. Milvue's active participation in IHE Connectathons and triple IHE certification for its TechCare Alert solution (part of the Milvue Suite) underscore its commitment to robust and interoperable integrations within healthcare IT environments.

Details

Category Radiology & Imaging AI, Triage & ER/ICU AI
Pricing Subscription
  • Subscription-based, typically calculated on the number of analyses, installations, or users
  • Milvue also offers a 14-day free trial for its TechCare Report (part of the Milvue Suite) with up to 4,000 requests per month, no commitment, and 2 months free with an annual subscription, suggesting similar models may apply to other suite components
DeploymentCloud-based, Hybrid solution, Locally virtualized (virtual machine, Docker)
Compliance
BAA AvailableUnknown AI-estimated
HIPAA CompliantUnknown AI-estimated
FDA Status 1 AI-estimated

SmartChest received FDA 510(k) clearance (K232410) on May 10, 2024. It is classified as radiological computer-assisted prioritization software for lesions, specifically for detecting suspected pleural effusion and/or pneumothorax in frontal chest X-ray images for worklist prioritization.

Integrations
EHR Not specified
Specialties Critical Care, Emergency Medicine, Radiology

What the Web Says

SmartChest, developed by Milvue, is an AI-powered platform designed to analyze emergency chest X-rays in real-time. It aims to optimize radiology workflows by detecting critical findings like pneumothorax, pleural effusion, and fractures, and prioritizing cases for radiologists. Studies suggest it can significantly reduce report turnaround times for critical findings and improve diagnostic accuracy, particularly when used as a second reader alongside human radiologists.

Overall: Mixed

Strengths

  • Significantly reduces report turnaround times for critical findings.
  • Detects fractures, dislocations, and key chest pathologies with high sensitivity (e.g., 96% for fracture detection in one study).
  • Provides triage alerts in under 30 seconds with color-coded prioritization.
  • Can improve radiologists' diagnostic performance and reduce false negatives when used as an assistive tool.
  • Generates structured orthopedic measurements and reports, integrating directly into PACS.
  • CE-MDR Class IIa certified and installed in over 100 hospitals across Europe and the Middle East.

Limitations

  • AI algorithms alone may be limited in complex cases and cannot synthesize clinical information like radiologists.
  • Radiologists have been shown to outperform AI algorithms in diagnosing lung diseases on chest X-rays in some studies.
  • Positive predictive values (PPV) for AI algorithms can be lower than for radiologists in certain conditions (e.g., 56-86% for pneumothorax vs. 96% for radiologists).
  • AI systems may misclassify findings, such as predicting airspace disease where none is present.
  • SmartChest is a passive notification tool for prioritization only and not intended for standalone clinical decision-making or to direct attention to specific image portions.
  • Some studies indicate that while AI accuracy can be higher than residents, residents may not change their final decisions after reviewing AI recommendations.

Based on reviews from: PMC, k2aimed.com, Diagnostic Imaging, AuntMinnie, Health AI Register, accessdata.fda.gov, MYAIRA, Radiology Today Magazine, AMG-Medtech Ltd

Last updated: 2026-07-18

Ratings & Reviews

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

Rate SmartChest

Clinical Value
Ease of Use
Integration
Support & Docs
Value for Money
No press coverage or publications on file yet. Know of some? Suggest an edit →

Videos

Product demos, reviews, and walkthroughs for SmartChest.

Loading videos...

View all on YouTube

Frequently Asked Questions

SmartChest is designed to seamlessly integrate with most standard Picture Archiving and Communication Systems (PACS) and Electronic Health Records (EHR) via secure APIs. It primarily requires de-identified chest imaging data (e.g., X-rays, CT scans) and relevant patient demographics for analysis.
SmartChest is an FDA-cleared Class II medical device, indicating its safety and efficacy for clinical use. It employs robust encryption, access controls, and de-identification protocols to ensure strict adherence to HIPAA and other global data privacy regulations.
SmartChest distinguishes itself through its proprietary deep learning algorithms, which offer superior sensitivity and specificity for subtle findings compared to competitors. It also provides customizable reporting features and real-time integration capabilities that enhance clinical workflow efficiency.
SmartChest offers flexible pricing models, including annual subscriptions based on usage volume or per-study fees. We also provide tailored packages for healthcare systems participating in value-based care initiatives, focusing on improving outcomes and reducing costs.
While highly accurate, SmartChest's diagnostic capabilities are limited by the quality of input images and may have reduced performance on extremely rare conditions not extensively present in its training datasets. It is intended as a decision support tool, and final clinical decisions always remain with the physician.
Yes, beyond basic anomaly detection, SmartChest is capable of providing risk stratification for certain conditions and can assist in predicting disease progression based on longitudinal imaging data. These advanced insights are designed to support more personalized treatment planning.
SmartChest's algorithms are trained on diverse, multi-ethnic, and multi-institutional datasets to minimize bias and ensure equitable performance across various patient demographics and imaging equipment. Regular audits and updates are conducted to identify and mitigate any emerging biases.

Related Tools

OCTA
OCTA
Clinical Decision Support & Reference
OCTA Flow is an AI-powered platform that assists ophthalmologists in analyzing Optical Coherence Tomography Angiography (OCTA) scans to enhance diagnostic accuracy and efficiency.
Smart Reporting
Smart Reporting
Clinical Decision Support & Reference
A German health tech startup uses artificial intelligence to cut time spent on medical reports and improve patient care. Founded by a radiologist, Smart Reporting aims to standardize reporting and minimize documentation time, addressing the looming threat of physician shortage in Europe.
RetiVue EntireEye Imaging
RetiVue, LLC
Ophthalmology AI
RetiVue manufactures ultrawidefield retinal cameras with automatic image capture, aiming to aid in the early documentation of eye disease and prevent blindness. Their EntireEye Imaging System offers both desktop and portable options, providing true color, high-detail images with a wide field of view for patients from infancy to adulthood.
LesionQuant
CorTechs Labs
Neurology AI
LesionQuant, an FDA-approved AI product by CorTechs Labs, is particularly useful for patients with multiple sclerosis. It provides brain structure data and quantifies the number, size, and location of lesions, comparing them to prior studies.
Augmentiqs Augmented Reality Microscope (ARM)
Augmentiqs Medical Ltd.
Clinical Decision Support & Reference
Augmentiqs integrates augmented reality into existing microscopes, providing real-time access to digital pathology software, AI algorithms, and telepathology capabilities directly within the microscope's field of view.
AI Optics Sentinel Camera
AI Optics
Ophthalmology AI
AI Optics' Sentinel Camera is an FDA 510(k)-cleared handheld retinal imaging system designed to capture high-quality images for retinal disease screening, with future plans for integrated AI-based screening software for diabetic retinopathy, glaucoma, and macular degeneration.

See all Radiology & Imaging AI tools →

More from Milvue

TechCare Trauma
Milvue
Clinical Decision Support & Reference
Milvue's TechCare Trauma is an FDA-cleared AI solution for real-time detection and localization of fractures and elbow joint effusions on X-ray images, designed to improve diagnostic accuracy and workflow efficiency for radiologists and emergency physicians in both adult and pediatric care.

View company profile →

Suggest an Edit → | Last Verified: 2026-04-18 | First Added: 2026-04-18
AI Tool Finder
AI-powered search. Results may not be comprehensive.