Annalise Enterprise CXR

annalise.ai

Annalise Enterprise CXR is an AI clinical decision-support solution for chest X-ray, assisting clinicians to interpret CXR studies by detecting up to 124 findings. It acts as a second pair of eyes providing notification of suspected findings.

Features include:
• analysing up to three images per study including frontal and lateral images
• a confidence bar displaying the likelihood of the finding and uncertainty of the AI model
• customisable user interface that integrates seamlessly into PACS and RIS.
• worklist triage

Available for clinical use in EU (MDR), UK, ANZ, India, UAE, Singapore, Thailand, Vietnam, Malaysia, Indonesia, Hong Kong.

**In the US, FDA 510(k) clearance for triage and notification for a subset of the findings, including pneumothorax, tension pneumothorax, pleural effusion, pneumoperiotneum, and vertical compression fracture.**
Product specifications Information source: Vendor
Last updated: Sept. 28, 2023
General
Product name Annalise Enterprise CXR
Company annalise.ai
Subspeciality Chest
Modality X-ray
Disease targeted 124 findings present in the emergent, urgent, and non-urgent care settings including: air space opacity, interstitial thickening, volume loss, effusions and lung masses, pneumothorax, malpositioned lines and tubes, pneumoperitoneum, acute bony trauma
Key-features Detection of up to 124 chest findings, worklist triage
Suggested use Before: adapting worklist order, flagging acute findings
During: perception aid (prompting all abnormalities/results/heatmaps), interactive decision support (shows abnormalities/results only on demand)
Data characteristics
Population All chest x-rays for patients over 16 years of age
Input Frontal (PA or AP), plus optional lateral chest X-ray images. Can process up to 3 images in a single study.
Input format DICOM
Output Indication of presence of finding, segmentation overlay, confidence and threshold score/bar
Output format AI Viewer. Worklist – HL7 or API based output for worklist triage (prioritisation)
Technology
Integration Integration in standard reading environment (PACS), Integration RIS (Radiological Information System), Integration via AI marketplace or distribution platform, Stand-alone third party application
Deployment Locally on dedicated hardware, Locally virtualized (virtual machine, docker), Cloud-based, Hybrid solution
Trigger for analysis Automatically, right after the image acquisition
Processing time 10-30 seconds
Certification
CE
Certified, Class IIb , MDR
FDA
 510(k) cleared, Class II
Market presence
On market since 10-2020
Distribution channels Wellbeing Software, Nuance PIN, Human Bytes
Countries present (clinical, non-research use) 10+
Paying clinical customers (institutes) 300+
Research/test users (institutes) 10+
Pricing
Pricing model Subscription
Based on Number of analyses
Evidence
Peer reviewed papers on performance

  • Commercially Available Chest Radiograph AI Tools for Detecting Airspace Disease, Pneumothorax, and Pleural Effusion (read)

  • Assessment of the effect of a comprehensive chest radiograph deep learning model on radiologist reports and patient outcomes: a real-world observational study (read)

  • Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study (read)

  • Diagnostic accuracy of a commercially available deep learning algorithm in supine chest radiographs following trauma (read)

  • Evaluation of an artificial intelligence model for detection of pneumothorax and tension pneumothorax in chest radiographs (read)

  • Do comprehensive deep learning algorithms suffer from hidden stratification? A retrospective study on pneumothorax detection in chest radiography (read)
Non-peer reviewed papers on performance

  • Poster BIR Annual Congress: Radiologist reporting productivity benefits from AI-assisted triage of CXR studies in clinical practice (read)

  • Poster RSNA: How normal is a normal chest X-ray: Does a comprehensive artificial intelligence model identify significant findings in chest radiographs interpreted as normal in clinical practice? (read)

  • Poster UKIO: Insights from implementation of an artificial intelligence assist device across a national radiology network (read)

  • Poster ECR: Radiologist’s feedback post implementation of a comprehensive AI assist device for CXR across a large radiology network (read)

  • Poster ECR: Remarkable vs Unremarkable Triage of Chest x-rays based on a comprehensive AI model – validation on a ground truthed, real world dataset. (read)
Other relevant papers

  • Abstract: Designing Effective Artificial Intelligence Software (read)