CCL

Összesen 1 találat.
#/oldal:
Részletezés:
Rendezés:

1.

001-es BibID:BIBFORM101294
035-os BibID:(cikkazonosító)e842 (wos)000804849400001
Első szerző:Kui Balázs
Cím:EASY-APP : an artificial intelligence model and application for early and easy prediction of severity in acute pancreatitis / Kui Balázs, Pintér József, Molontay Roland, Nagy Marcell, Farkas Nelli, Gede Noémi, Vincze Áron, Bajor Judit, Gódi Szilárd, Czimmer József, Szabó Imre, Illés Anita, Sarlós Patrícia, Hágendorn Roland, Pár Gabriella, Papp Mária, Vitális Zsuzsanna, Kovács György, Fehér Eszter, Földi Ildikó, Izbéki Ferenc, Gajdán László, Fejes Roland, Németh Balázs Csaba, Török Imola, Farkas Hunor, Artautas Mickevicius, Ville Sallinen, Shamil Galeev, Elena Ramirez Maldonado, Párniczky Andrea, Erőss Bálint, Hegyi Péter Jenő, Márta Katalin, Váncsa Szilárd, Sutton Robert, Enrique de-Madaria, Elizabeth Pando, Piero Alberti, Maria José Gómez-Jurado, Alina Tantau, Szentesi Andrea, Hegyi Péter, Hungarian Pancreatic Study Group
Dátum:2022
ISSN:2001-1326
Megjegyzések:Background: Acute pancreatitis (AP) is a potentially severe or even fatal inflammation of the pancreas. Early identification of patients, who are at high risk for developing a severe course of the disease is crucial for preventing organ failure and death. Most of the former predictive scores require many parameters or at least 24 hours to predict the severity, so the early therapeutic window is missing. Methods: The early achievable severity index (EASY) is a registered multicentre, multinational, prospective, observational study (ISRCTN10525246). Clinical parameters were collected from 15 countries and 28 medical centres via eCRF. The predictions were made using machine learning models including Decision Tree, Random Forest, Logistic Regression, SVM, CatBoost, and XGBoost. For the modeling, we used the scikit-learn, xgboost, and catboost Python packages. We have evaluated our models using 4-fold cross-validation and the receiver operating characteristic (ROC) curve, the area under the ROC curve (AUC), and accuracy metrics have been calculated on the union of the test sets of the cross-validation. The most important factors and their contribution to the prediction were identified using a modern tool of explainable artificial intelligence, called SHapley Additive exPlanations (SHAP). Using the XGBoost machine learning algorithm for prediction, the SHAP values for the explanation, and the bootstrapping method for the estimation of confidence we have developed a web application in the Streamlit Python-based framework. Results: The prediction model is based on the international cohort of 1184 patients and a validation cohort of 3543 patients. The best performing model has been an XGBoost classifier with an average AUC score of 0.81 and accuracy of 89.1% and the model is improving with experience. The six most influential features are the respiratory rate, body temperature, abdominal muscular reflex, gender, age, and glucose level. Finally, a free and easy-to-use web application was developed (http://easy-app.org/). Conclusions: The EASY prediction score is a practical tool for identifying patients at high risk for severe acute pancreatitis within hours of hospital admission. The easy-to-use web application is available for clinicians and contributes to the improvement of the model.
Tárgyszavak:Orvostudományok Klinikai orvostudományok idegen nyelvű folyóiratközlemény külföldi lapban
folyóiratcikk
severity prediction
acute pancreatitis
artificial intelligence
Megjelenés:Clinical and Translational Medicine. - 12 : 6 (2022), p. 1-13. -
További szerzők:Pintér József (1930-) (urológus) Molontay Roland Nagy Marcell Farkas Nelli Gede Noémi Vincze Áron Bajor Judit Gódi Szilárd Czimmer József Szabó Imre Illés Anita Sarlós Patrícia Hágendorn Roland Pár Gabriella Papp Mária (1975-) (belgyógyász, gasztroenterológus) Vitális Zsuzsanna (1963-) (belgyógyász, gasztroenterológus) Kovács György (1982-) (belgyógyász, gasztroenterológus) Fehér Eszter Földi Ildikó (1981-) (orvos) Izbéki Ferenc Gajdán László Fejes Roland Németh Balázs Csaba Török Imola Farkas Hunor Mickevicius, Artautas Sallinen, Ville Galeev, Shamil Ramírez-Maldonado, Elena Párniczky Andrea (gyermekgyógyász) Erőss Bálint Hegyi Péter Jenő (belgyógyász) Márta Katalin Váncsa Szilárd Sutton, Robert de-Madaria, Enrique Pando, Elizabeth Alberti, Piero Gómez-Jurado, Maria José Tantau, Alina Szentesi Andrea Hegyi Péter (pszichológus) Hungarian Pancreatic Study Group
Internet cím:Intézményi repozitóriumban (DEA) tárolt változat
DOI
Borító:
Rekordok letöltése1