Zum Inhalt gehen
MicrosoftTeams-image-7
Emerging technologies

Improving Classification of Pneumonia in Chest X-ray (CXR) images

With the rise of COVID-19, medical facilities were operating beyond peak capacity. Capgemini has developed a Deep Learning based classification system that can distinguish between pneumonia and non-pneumonia cases as well as between viral (including COVID-19) and bacterial cases in seconds.

The aim of the project is to support the diagnosis of pneumonia and thus reduce the overall workload of medical staff. This MVP was presented on the AIforGood 2020 Summit organized by ITU.

The algorithm thus developed as part of our solution led to an improved classification of X-ray in the distinction between pneumonia and non-pneumonia cases, which in turn led to a reduction in the number of false negatives.

Although initially developed for accurately classifying pneumonia cases, this solution is also relevant for other fields of classification.

Download the story to know more or reach out to our experts to explore further possibilities.

Kontaktieren Sie unseren Experten

Dr. Eldar Sultanow

Enterprise Architecture
Dr. Eldar Sultanow ist Software-Architekt. Er hat langjährige Praxiserfahrung in der Softwareindustrie, insbesondere in den Bereichen JEE, Electronic/Mobile Commerce, Track-&-Trace und Auto-ID im Pharmabereich. In einem zwischenstaatlichen Projekt hat er eine Plattform mit konzipiert, an der internationale Finanzinstitute angeschlossen sind. Aktuell ist Eldar Sultanow als technischer Chefdesigner in einem der größten öffentlichen IT-Verfahren aktiv, das hunderttausende Transaktionen pro Tag mit einem Jahresvolumen von über 25 Milliarden EUR vollzieht.