Chest X-Ray AI Analysis Package
Embedded intelligence, supporting the clinical workflow
Powered by Lunit INSIGHT CXR & RUBEE® AI
RUBEE® AI and the significance of the Enterprise Imaging Platform
Healthcare systems across the globe are exploring the potential application and benefits of Artificial Intelligence when it comes to improving quality care, focused on outcomes.
The path towards realizing the benefits opens another opportunity: addressing the interoperability and integration aspects. Already, there are hundreds of start-ups and AI developers working in the healthcare arena, each focusing on highly specific applications. Selecting which ones you need, and then integrating them into your system and workflows is far from simple.
RUBEE® AI, as part of your Enterprise Imaging platform, offers a seamless AI experience for your clinicians. Carefully curated ‘packages’ embed best-of-breed AI apps that work seamlessly to support your real clinical workflow from start to finish.
With RUBEE AI, you get more out of your AI investments, while enriching the value of your Enterprise Imaging. It’s a win-win-win for your hospital, your clinicians and your patients.
What is RUBEE AI?
What is RUBEE AI?
With RUBEE AI, you can embed our AI specialty packages into your clinical workflows.
RUBEE visualizes the metadata generated by algorithms such as deep learning, machine learning, image analysis and natural language processing. It also uses that information to automate and optimize your workflows, all within your Enterprise Imaging ecosystem.
Future-proof and clinically relevant
‘Augmented Intelligence’ does just that: offering a set of tools that let your clinicians maximize the value of their own expertise, increase their productivity and enhance diagnosis. But to get the real benefits to your clinicians, the tools need to be embedded right into the workflows and systems they use every day. Your Enterprise Imaging already offers a forward-thinking, multi-specialty platform that consolidates your hospital’s wealth of data.
With RUBEE AI, it also becomes your AI enabled ecosystem. Standards-based workflows are embedded with niche and specialty-focused AI apps, delivering clinical relevance. No need for an additional, complex and costly dedicated AI platform or marketplace. And, instead of worrying about which apps to select and how to integrate them, you are leaping ahead with an ecosystem of seamlessly embedded AI.
“Your Enterprise Imaging already offers a future-proof, multi-specialty platform that consolidates your hospital’s wealth of data. With RUBEE AI, it also becomes your AI enabled ecosystem.”
Best-of-Class AI apps, Specialty Packages
We have taken the guesswork out of your AI journey. Our AI specialty packages have been carefully curated to enable interoperability and integration of best-of-class algorithms.
You can be confident that all the algorithms come from reliable companies, and trained on evidence-based data. Most of all, they are powered by RUBEE AI, ensuring that they meet your specific clinical needs from start to finish. So, there’s no ‘trial and error’: just proven value.
Enrich your Enterprise Imaging workflows
With AI fully embedded in your Enterprise Imaging Platform, your clinicians see benefits all along the line.
- Task assignments and case distribution are smoothly automated, based on the metadata generated from the AI apps.
- Hanging protocols get ‘smart’, with dedicated reading protocols.
- Report automation by auto-including AI results into the reporting workflow.
Offering advanced visualizations, workflow optimization and automation, RUBEE AI helps your clinicians to focus their efforts on cases that require immediate attention.
Chest X-Ray AI Analysis package
Powered by Lunit INSIGHT CXR
Clinical challenges in X-ray Analysis
- High volume of cases assigned to each radiologist, can create backlogs at healthcare facilities
- Unable to intelligently sort or prioritize workflow
- Incidental findings that might be otherwise missed
- Looking for actionable nodules on PACS workstation can be tedious
- Detection of Tuberculosis on CRs remains a labour and time-intensive task that requires and expert’s interpretation (1) (2)
When it comes to detection of lung nodules, machine automation and deep learning algorithms can provide an added value to help address challenges associated with high volume of cases, identification of actionable nodules and workflow prioritization.
Introducing the Chest X-Ray AI Analysis package
Powered by Lunit INSIGHT CXR
- AGFA’s RUBEE AI enables EI Desktop and XERO Universal Viewer AI visualizations
- INSIGHT CXR is CE cleared
- INSIGHT CXR helps detect 10 common chest x-ray findings and generates the analysis result which indicates the presence and the location of chest abnormalities
Detection
Innovative image enhancement technology designed to increase the clarity of chest X-rays by suppressing bone on digital images.
- Atl (Atelectasis)
- Calc (Calcification)
- Csn (Consolidation)
- Fib (Fibrosis)
- Ndl (Nodule)
- Pef (Pleural effusion)
- Ppm (Pneumoperitoneum)
- Ptx (Pneumothorax)
- Cm (Cardiomegaly)
- MW (Mediastinal widening)
- Detection of areas of suspicious abnormal radiologic findings
- Automated analysis of chest radiographs via deep learning technology
- Visualization and quantitative estimation of the likelihood of the presence of each abnormality
Tuberculosis screening option can be turned on
Excellent and consistent performance in the detection of active pulmonary tuberculosis on CR, outperforming physicians, including thoracic radiologists (3)
Chest X-ray AI Analysis Package
- Detects 10 abnormal radiologic findings with 97-99% accuracy
- Supports tuberculosis screening on chest x-ray images.
Product may not be available in all regions, contact your AGFA HealthCare representative for details.
Detection and Analysis
97-99% AUROC
Very high performance level
34%
Reduced reading time
Detected Location
The AI generates the location information of detected lesions in the form of heatmaps and/or contour maps.
Abnormality Score
The AI generates an abnormality score which reflects the AI’s calculation of the actual presence of the detected lesion.
AI report
The AI provides a “case report” that summarizes the overall analysis result. The analysis results are narrowed down to each finding if multiple findings are detected.
Chest X-ray AI Analysis Package – Major features
- Fast triage of normal cases:
Triage normal cases quickly and focus on reading abnormal cases where lesions might exist. - Efficient reading via exam prioritization:
In reference to the abnormality scores on the worklist, radiologists can prioritize exams in their reading order, resulting in a 13% reduction in reading time, and a 33% in reduction time for normal cases. (1) - Improved reading performance:
Non-radiology physicians, general radiologists, and even thoracic radiologists can improve their diagnostic accuracy for major chest abnormalities. (5) (6) (7) (8) (9) (10) (11) - Reduced overlooked lung cancers:
Automatic detection of small and subtle pulmonary nodules overlapped in the hilar shadow, ribs, heart, and diaphragm enables radiologists to reduce overlooked lung cancer cases, especially during regular check-ups. (1) (2) - Streamlined ED workflow:
Radiology residents can improve their diagnostic performance(13) and reduce their reading time(14), which ultimately accelerates the decision-making process and treatment in the ED. - COVID 19 patient triaging and monitoring:
AI-aided chest radiograph interpretation can help medical professionals detect COVID-19 infected pneumonia quickly(15) and accurately(16), enabling prompt isolation and timely treatment.
AI embedded Xray – Major features
Small module detected
63%
Abnormality score
Multiple findings including atelectasis, pleural effusion
98%
Abnormality score
COVID-19 infected pneumonia detected
77 %
Abnormality score
References
(1) Melendez J, Sánchez CI, Philipsen RH, et al. An automated tuberculosis screening strategy combining X-ray-based computer-aided detection and clinical information. Sci Rep 2016; 6:25265.
(2) Hoog AH, Meme HK, van Deutekom H, et al. High sensitivity of chest radiograph reading by clinical officers in a tuberculosis prevalence survey. Int J Tuberc Lung Dis 2011; 15:1308–14.
(3) Lee, J.H., Park, S., Hwang, E.J. et al. Deep learning–based automated detection algorithm for active pulmonary tuberculosis on chest radiographs: diagnostic performance in systematic screening of asymptomatic individuals. Eur Radiol 31, 1069–1080 (2021).
(4)Ju Gang Nam, Minchul Kim, et al. Development and validation of a deep learning algorithm detecting 10 common abnormalities on chest radiographs. European Respiratory Journal. 2020
(5) Ju Gang Nam, Sunggyun Park, et al. Development and Validation of Deep Learning–based Automatic Detection Algorithm for Malignant Pulmonary Nodules on Chest Radiographs. Radiology. 2018
(6) Eui Jin Hwang, Sunggyun Park, et al. Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs. Clinical Infectious Diseases. 2018
(7) Eui Jin Hwang, Sunggyun Park, Kwang-Nam Jin, et al. Development and Validation of a Deep Learning–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs. JAMA Network Open. 2019
(8) Jong Hyuk Lee, Sunggyun Park, et al. Deep learning–based automated detection algorithm for active pulmonary tuberculosis on chest radiographs: diagnostic performance in systematic screening of asymptomatic individuals. European Radiology. 2020
(9) Eui Jin Hwang, Jung Hee Hong, et al. Deep learning algorithm for surveillance of pneumothorax after lung biopsy: a multicenter diagnostic cohort study. European Radiology. 2020
(10) Jong Hyuk Lee, Hye Young Sun, et al. Performance of a Deep Learning Algorithm Compared with Radiologic Interpretation for Lung Cancer Detection on Chest Radiographs in a Health Screening Population. Radiology. 2020
(11) Hyunsuk Yoo, Ki Hwan Kim, et al. Validation of a Deep Learning Algorithm for the Detection of Malignant Pulmonary Nodules in Chest Radiographs JAMA Network Open. 2020
(12) Sowon Jang, Hwayoung Song, et al. Deep Learning–based Automatic Detection Algorithm for Reducing Overlooked Lung Cancers on Chest Radiographs. Radiology. 2020
(13) Eui Jin Hwang, Ju Gang Nam, et al. Deep Learning for Chest Radiograph Diagnosis in the Emergency Department.Radiology. 2019
(14) Jae Hyun Kim, Jin Young Kim, et al. Clinical Validation of a Deep Learning Algorithm for Detection of Pneumonia on Chest Radiographs in Emergency Department Patients with Acute Febrile Respiratory Illness. Journal of Clinical Medicine. 2020
(15) Eui Jin Hwang, Hyungjin Kim, et al. Implementation of a Deep Learning-Based Computer-Aided Detection System for the Interpretation of Chest Radiographs in Patients Suspected for COVID-19. Korean Journal of Radiology. 2020
(16) Se Bum Jang, Suk Hee Lee, et al. Deep-learning algorithms for the interpretation of chest radiographs to aid in the triage of COVID-19 patients: A multicenter retrospective study. PLOS ONE. 2020