Retroperitoneal Fibrosis, Retroperitoneal Sarcoma
Conditions
Keywords
radiomics, auto-segmentation
Brief summary
A retrospective study utilizing archived CT scans of patients diagnosed with retroperitoneal fibrosis, sarcoma or other malignancies (i.e. lymphoma, germ cell tumors, metastasis, infections, ganglioneuromas) in order to implement a radiomics algorithm which is able to differentiate between these malignancies.
Detailed description
The aim of this project is to develop a radiomics algorithm that can reliably identify retroperitoneal fibrosis (Ormond's disease) and retroperitoneal sarcomas, automatically segment them and differentiate them from other retroperitoneal diseases. Radiomics is a technique that uses artificial intelligence to extract characteristics from radiological image data that are not visible to humans and to identify image morphological patterns of diseases. As it is difficult to differentiate between diseases using image data alone, clinical data such as symptoms and laboratory values are to be correlated with the image data and utilized by the algorithm. Among other things, this should increase the sensitivity, accuracy and specificity of image-based diagnostics in order to enable faster, non-invasive diagnosis.
Interventions
OTHERRadiomics Algortihm
A radiomics algorithm designed to distinguish retroperitoneal fibrosis from other retroperitoneal tumors and provide recommendations for clinical treatment decisions.
Sponsors
Heidelberg University
Study design
Observational model
OTHER
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
* Patients of any age or gender. * CT scans confirming the presence of a retroperitoneal mass. * Confirmed diagnosis of retroperitoneal fibrosis, sarcoma or other malignancies (i.e. lymphoma, germ cell tumors, metastasis, infections, ganglioneuromas) through pathology reports or clinical follow-up.
Exclusion criteria
* Poor quality CT scans where the region of interest is not clearly visible. * Previous treatments or surgeries that might alter the radiomic features of the tumors.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Radiomic accuracy for retroperitoneal fibrosis | 6 months | Accuracy of the algorithm in differentiating between retroperitoneal fibrosis and other retroperitoneal diseases |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Radiomic accuracy for retroperitoneal sarcomas | 10 Months | Accuracy of the algorithm in differentiating between retroperitoneal sarcoma and other retroperitoneal diseases using CT images |
Countries
China, Germany
Outcome results
None listed