- Home
- Solutions informatiques et logicielles
- Artificial Intelligence
- AI campaign
- AI-Rad Companion Chest CT
AI-Rad Companion Chest CT
Your imaging decision support companion for chest CT interpretation
AI-Rad Companion Chest CT is an AI-powered radiology assistant that supports clinicians by performing automatic measurements and preparing the results in the form of valuable clinical images, quantifications and medical reports. It can differentiate between the various structures of the chest, highlight them individually, and mark and measure potential abnormalities automatically. This applies equally to organs such as the heart and lungs, aorta, and vertebral bodies. Through AI-powered algorithms, the AI-Rad companion assists radiologists in their daily clinical routine by undertaking repetitive and time-consuming image recognition tasks.
AI-Rad Companion also provides a structured dataset with all measurements, which can be imported to other database systems in the hospital and used for research related purposes where such measurements are important.
AI-Rad Companion Chest CT interpret standard DICOM CT Images of the thorax.
AIRC Chest CT
AI-Rad Companion Chest CT supports radiologists by preparing valuable clinical information to enhance the image interpretation workflow. It automatically performs measurements, prepares results which could be used for reports, and helps to handle the workload with more ease.
AI-Rad Companion Chest CT automatically highlights abnormalities, characterizes anatomies, and matches results with reference values as well, to support more confident case evaluation. AI-Rad Companion acts as an assistant for radiologists working at their PACS workstation applying the following approach:
- Provide a first reading to radiologists to support final clinical decision.
- Provide a second reading to support radiologist clinical decision.
Enabled by the teamplay digital health platform and using state-of-the-art image processing algorithms supported by artificial intelligence, AI-Rad Companion Chest CT is integrated into the radiologist's workflow and delivers value in four key areas:
- Accelerated interpretation and workflow efficiency
- Improved clinical outcomes and increased accuracy
- Provision of additional clinically relevant information and visual highlighting
- Standardized results while helping to reduce interreader variability
The goal of AI-Rad Companion is to seamlessly integrate into your hospitals existing environment as well as the radiologist workflows. We want to augment the radiologist’s/clinical expert’s interpretation process, not hinder it.
Clinical examples
AI-Rad Companion focuses on three main parts of the thorax: the lungs, the cardiovascular system and the Spine.
It provides the following features with respect to lung-analysis:
- Detection and segmentation of lung nodules and localization with respect to lung lobes
- Analysis of the lung parenchyma based on segmented lung lobes with respect to:
- areas of low attenuation (low attenuation volume, or LAV)
- areas of opacity
- volume of lung lobes
Certain parts of this extension were used in the prototype Siemens Healthineers offered to fight COVID-19, to analyze ground-glass opacities and consolidations. High opacity abnormalities were shown to correlate with lungs of COVID-19 patients.
The AI-Rad Companion Chest CT detects and highlights lung nodules. After segmentation of the lung nodules, the volume, maximum 2D diameter, maximum 3D diameter and tumor burden are automatically calculated
Cinematic Rendering adds clarity to the location of a detected lung nodule. The visualization makes it easy for the referring physician to get a clear picture of the nodule location within the lung.
Certain parts of this extension were used in the prototype Siemens Healthineers offered to fight COVID-19, to analyze ground-glass opacities and consolidations. High opacity abnormalities were shown to correlate with lungs of COVID-19 patients.
The AI-Rad Companion Chest CT detects and highlights lung nodules. After segmentation of the lung nodules, the volume, maximum 2D diameter, maximum 3D diameter and tumor burden are automatically calculated
Cinematic Rendering adds clarity to the location of a detected lung nodule. The visualization makes it easy for the referring physician to get a clear picture of the nodule location within the lung.
Certain parts of this extension were used in the prototype Siemens Healthineers offered to fight COVID-19, to analyze ground-glass opacities and consolidations. High opacity abnormalities were shown to correlate with lungs of COVID-19 patients.
For the radiological assessment of the cardiovascular system a large variety of dedicated CT scan protocols exist depending on the clinical indication. The protocols differ mainly with respect to the cardiac phase in which the acquisition is performed (controlled via ECG-gating) and the type and timing of contrast enhancement. AI-Rad Companion is designed to work with any of these protocols, particularly the most generic non-gated and non-contrast-enhanced Chest CT scans.
The features are:
- Measurement of heart volume and coronary calcium volume (on unenhanced data only)
- Segmentation of aorta and diameter measurements at 9 landmarks according to AHA-guidelines (on both native and contrast-enhanced data)
This 3D overview of the thoracic aorta has been automatically created by the AI-Rad Companion Chest CT. It includes the measurement of relevant diameters, based on medical guidelines and detected anatomical landmarks. This clearly shows how the AI-Rad Companion Chest CT can support the increase of accuracy of your reporting.
The AI-Rad Companion Chest CT automatically recognizes coronary calcium, calculates its volume, and creates an enhanced clinical image including quantification of the heart volume. This additional information about a cardiac condition is valuable clinical knowledge to a routine Chest CT study.
This 3D overview of the thoracic aorta has been automatically created by the AI-Rad Companion Chest CT. It includes the measurement of relevant diameters, based on medical guidelines and detected anatomical landmarks. This clearly shows how the AI-Rad Companion Chest CT can support the increase of accuracy of your reporting.
The AI-Rad Companion Chest CT automatically recognizes coronary calcium, calculates its volume, and creates an enhanced clinical image including quantification of the heart volume. This additional information about a cardiac condition is valuable clinical knowledge to a routine Chest CT study.
This 3D overview of the thoracic aorta has been automatically created by the AI-Rad Companion Chest CT. It includes the measurement of relevant diameters, based on medical guidelines and detected anatomical landmarks. This clearly shows how the AI-Rad Companion Chest CT can support the increase of accuracy of your reporting.
Osteoporosis manifests as loss of bone density e.g. in the spine, and consequently in vertebral compression fractures. Some studies showed that the HU-values of the spine obtained from CT data acquired for other indications can be used to identify osteoporotic patients and called this approach “opportunistic screening for osteoporosis”. AI-Rad Companion (Musculoskeletal) provides:
- Labeling and segmentation of thoracic vertebrasµ
- Measurements of vertebrae heights
- Mean vertebral body HU-density
On this image you can see how the height of the vertebrae bodies is measured and compared. The AI-Rad Companion Chest CT automatically processes these measurements and marks abnormalities with easy-to-understand color-coding. Deviations are highlighted, while bone density is reported in Hounsfield Units.
Customer experiences
“Saving time is one of the many benefits of AI-assisted tools such as the AI-Rad Companion Chest CT. Manual measurements, which until recently took up valuable time, are now performed automatically and extremely accurately by the tool in a very short time. The findings provided by the AI-Rad Companion report supports our team to write our own report faster and with greater accuracy, reaching certain diagnoses faster.”
Dr. Louis Flamée, Radiologist at Jan Yperman Hospital
Additional content
Contact
See the other topics here