Have a look at some of the projects we are currently working on
FHIR-PYrate is an in-house Python package that provides a comprehensive functionality to query FHIR servers for desirable resources such as diagnostic reports, observations, radiologic imaging data, and many more. The content of diagnostic reports can be filtered with RegEx or the advanced natural language processing library spaCy. Additionally, DICOM studies and series can be downloaded and anonymized.
Body Composition Analyses (BCA) involve automatical extraction of CT-based biomarkers, such as muscle, fat, and bone volumes to use as predictive features in the determination of clinical outcomes. Presently, biomarkers remain largely unused, but are valuable by-products of CT scans, allowing a baseline comparison for disease patients and a comprehensive analysis of disease progression.
Prof. Felix Nensa and Prof. Arzu Oezcelik report on how AI can improve the care of transplant patients. Significantly faster than humans, AI can calculate the size of the liver in order to improve the safety of donors and recipients.
Pydicom-seg is a library to facilitate the conversion process of DICOM-SEG files into ITK compatible data formats by providing a Python native implementation of reading and writing functionality with support for numpy and SimpleITK. Additionally, common use cases like loading multi-class segmentations are supported out-of-the-box.
The constantly growing amount of health data makes it possible to develop intelligent and personalized applications for early health detection, diagnostics, treatment and aftercare. Above all, AI-based systems hold enormous potential, which SmartHospital.NRW wants to leverage and make usable for hospitals in North Rhine-Westphalia.