Annalise Enterprise CXR
comprehensive decision-support AI for chest x-ray

Empower clinicians with Annalise CXR,
a second pair of eyes detecting 124 findings.

Some features are not available in all regions,
please check the regulatory status with an annalise.ai employee.

Where we work

Intended to assist clinicians with the interpretation of radiological imaging studies, the solution is CE marked for use as a medical device in the EU, and available for clinical use in Australia , New Zealand and Malaysia. Annalise Enterprise is not available for purchase by the general public.

See the chest like never before
Lateral and frontal views. Three images per study.

Analysing up to three images per study including frontal and lateral images, clinicians can see more thanks to the correlation of findings across multiple views.

See sample cases in web demo
A comprehensive decision-support tool
Detecting 124 findings. By clinicians for clinicians.

The development of the breadth and depth of Annalise CXR’s ontology tree was a clinically driven process, intended to select a comprehensive list of findings that would be most clinically necessary and helpful to clinicians.

Thoughtful design
Flexible user interface (UI). Seamless workflow.

Built by clinicians who understand the importance of avoiding disruptions to the workflow, Annalise CXR’s customisable UI integrates seamlessly into your PACS and RIS. Clinicians feel confident enough to use Annalise CXR in under 10 minutes.

 

Interpret with confidence
Confidence bar

The confidence bar displays the likelihood of the finding and uncertainty of the AI model assisting clinicians to interpret with confidence.

Learn more about confidence and uncertainty in medical imaging AI
Findings

Our algorithm can detect the following findings

Technical Factors
Devices
Skeletal
Cardiomediastinum
Pleuropulmonary
Peripheral
Technical Factors
Patient rotation
Cervical flexion
Underinflation
Underexposed
Overexposed
Incompletely imaged chest
Image obscured
Lines and Tubes
In position central line
In position endotracheal tube
In position Nasogastric tube
In position Pulmonary Arterial Catheter (PAC)
Intercostal drain
Suboptimal Central Line (CVC)
Suboptimal Endotracheal Tube (ETT)
Suboptimal nasogastric tube (NGT)
Suboptimal Pulmonary Arterial Catheter (PAC)
Cardiac Devices
Electronic cardiac devices
Cardiac valve prosthesis
Sternotomy wires
Orthopaedic Implants
Rib fixation
Shoulder fixation
Shoulder replacement
Rotator cuff anchor
Clavicle fixation
Spinal fixation
Surgical Clips and Stents
Mediastinal clips
Neck clips
Axillary clips
Abdominal clips
Lung sutures
Aortic stent
Coronary stent
Airway stent
Oesophageal stent
Biliary stent
Ribs
Acute rib fracture
Chronic rib fracture
Rib resection
Humerus
Acute humerus fracture
Chronic humerus fracture
Shoulder dislocation
Shoulder arthritis
Scapular
Scapular fracture
Clavicle
Acute clavicle fracture
Chronic clavicle fracture
Spine
Kyphosis
Scoliosis
Spinal wedge fracture
Spinal arthritis
Diffuse spinal osteophytes
Osteopaenia
Bone Lesion
Spinal lesion
Scapular lesion
Humeral lesion
Rib lesion
Clavicle lesion
Cardiomediastinum
Widened cardiac silhouette
Inferior mediastinal mass
Superior mediastinal mass
Hilar lymphadenopathy
Calcified hilar lymphadenopathy
Pneumomediastinum
Unfolded aorta
Widened aortic contour
Aortic arch calcification
Pulmonary congestion
Pulmonary artery enlargement
Pericardial fat pad
Airspace Opacity
Focal airspace opacity
Multifocal airspace opacity
Diffuse lower airspace opacity
Diffuse upper airspace opacity
Perihilar airspace opacity
Diffuse airspace opacity
Interstitial
Upper zone fibrotic volume loss
Lower zone fibrotic volume loss
Diffuse interstitial thickening
Upper interstitial thickening
Basal interstitial thickening
Diffuse fibrotic volume loss
Diffuse nodular / miliary lesions
Upper zone bullae
Lower zone bullae
Diffuse bullae
Collapse
Atelectasis
Segmental collapse
Lung collapse
Post resection volume loss
Pulmonary Lesion
Solitary lung nodule
Solitary lung mass
Multiple masses or nodules
Cavitating mass(es)
Cavitating mass with content
Calcified granuloma (< 5mm)
Calcified mass (> 5mm)
Nipple shadow
Airways
Reduced lung markings
Peribronchial cuffing
Hyperinflation
Bronchiectasis
Tracheal deviation
Pneumothorax
Simple pneumothorax
Tension pneumothorax
Pleural Effusion
Simple effusion
Loculated effusion
Pleural Thickening
Calcified pleural plaques
Pleural mass
Diffuse pleural thickening
Diaphragmatic Contour
Diaphragmatic elevation
Diaphragmatic eventration
Chest Wall Contour
Pectus carinatum
Pectus excavatum
Soft Tissues
Subcutaneous emphysema
Mastectomy
Breast implant
Calcified axillary nodes
Calcified neck nodes
Abdomen
Subdiaphragmatic gas
Distended bowel
Hiatus hernia
Gallstones
Gastric band
Suboptimal gastric band
Non-surgical Foreign Body
Internal foreign body
Testimonials
“I realised Annalise CXR behaves, in a sense, like a quiet radiology pal sitting over your shoulder, or not unlike having a keen registrar involved in your reporting session. That is, it hovers in the background, and ensures you don't take anything for granted on the X-ray under review.”
Private clinical practice radiologist
1/2
"If a radiologist could accessorise, Annalise would be the choice to make. Accurate, fast and easily integrated into existing workflow, Annalise showcases AI as a tool for enhanced efficiency in the reporting room."
Pilot program radiologist user
2/2
Powered by clinically robust AI

Broad datasets, extensive training and robust validation

750000
+
unique CXR images
Annalise CXR's AI algorithm was trained on over 750,000 images. Chest x-rays were sourced from broad datasets from three continents, including CXR machine equipment and manufacturers, departmental and portable imaging, patient demographics, and inpatient and outpatient groups.
148
radiologists
Annalise CXR has been trained and validated on datasets hand-labelled by a group of 148 radiologists. Each case in the validation study was reviewed by three thoracic subspecialist radiologists.
280
million
CXR labels
To ensure a consistent and high-standard of labelling across all data-sets, radiologists were trained in standardised labelling procedures. Each case was then hand-labelled independently by three radiologists (triple-labelled) and assessed for the absence or presence of each finding.
Product benefits

The Annalise CXR advantage

Reducing ‘Never Events’

According to the UK’s NHS National Patient Safety Agency*, between 2005 and 2010, 45% of all cases of patient harm caused by a malpositioned nasogastric tube (NGT) were due to misinterpreted CXRs. In 2009 in the UK, complications of a malpositioned NGT became a ‘never event’ – a serious adverse incident which is wholly preventable if clinical safety protocols are in place. Annalise CXR provides the immediate assessment of NGT position to prevent such an incident.

*Patient Safety Alert NPSA/2011/PSA002: Reducing the harm caused by misplaced nasogastric feeding tubes in adults, children and infants. London: National Patient Safety Agency, 2011.

1/2

Managing hidden stratification

The process of interpreting cases with multiple clinical findings lead to complex associations between findings, known as hidden stratification. This is inherent to any model examining multiple findings, and robust examination of the interplay of findings allows for validation of the model and improved model performance. Annalise CXR is the first solution to analyse the hidden stratification effect of a subset of clinically relevant findings.

2/2
Deployment

Customised deployment solutions

Cloud
Simple and quick implementation using a cloud backend deployment while still controlling data privacy within your network. For customers who want to realise the benefits of cloud computing - maximum flexibility, scalability and reliability, with instant upgrade access.
On-premise
An on-premise deployment option allowing for customer specific data processing and storage requirements. For customers who need a whole on-site solution and complete control of the IT infrastructure.
Discover Annalise CXR

Our experienced sales and deployment teams are here to help with any questions you may have or if you'd like to book a private demo.