Pediatric Appendicitis Prediction Tool
This is a research prototype for the tool that shall predict how likely it is that a pediatric patient has appendicitis, will need a surgical treatment and will develop complicated appendicitis.1 This prototype is not a finished tool and is only intended for demonstration purposes regarding the ongoing research. It must therefore not be used for medical advice or decisions.
All entered data will be stored (short-term) for prediction with pre-trained machine learning models and will be erased immediately after performing prediction.
1 This prototype was developed by the Medical Data Science research group, ETH Zurich, Switzerland, in collaboration with the University Children's Hospital Regensburg (KUNO), Germany, Research and Development Campus Regensburg (WECARE), Germany, and Department of Pediatric Surgery and Pediatric Orthopedics, Hospital St. Hedwig of the Order of St. John, Regensburg, Germany.
The Medical Data Science group works on developing and extending machine learning techniques for precision medicine, life sciences, and clinical data analysis. Our ultimate goal is to develop and employ the computational and statistical tools that are necessary to enable personalized medical treatment of patients. This website demonstrates current results of one of our research projects. The tools provided on this page are prototypes. They may not be used for medical purposes (advice, decisions, etc.).
* Mandatory fields.
** "NA" stands for missing values. All missing values will be imputed. Missingness can result in predictions being less accurate! Note: Predictions produced by this prototype must not be used for clinical decisions and serve only for demonstration purposes.
*** Alvarado and pediatric appendicitis scores have to be calculated according to these criteria.
**** Preprocessing involves a best-effort removal of artifacts such as numbers or icons added by the ultrasound machine. Enabling this option increases total runtime significantly.
***** Image file must be either jpg or png format.