@misc{Szaleniec_Joanna_Artificial_2014,
 author={Szaleniec, Joanna},
 address={Kraków},
 howpublished={online},
 year={2014},
 school={Uniwersytet Jagielloński. Collegium Medicum. Wydział Lekarski.},
 language={pol},
 abstract={Introduction. Surgical treatment reduces the symptoms of chronic rhinosinusitis (CRS) in most, but not all patients. The choice of optimal treatment modality for each patient can be aided by outcome prediction.Aim. The aim of the study was to build models for individual outcome prediction in patients with CRS.Material and method. The study group comprised of 183 patients who underwent endoscopic sinus surgery for CRS. The preoperative evaluation included anamnesis, laryngological examination, computed tomography and self-assessment of symptoms. The symptoms were reevaluated 3-6 and 12 months after surgery. The results of treatment were predicted using artificial neural networks (self-learning systems inspired by the structure and function of the nervous system).Results. The best obtained models provided correct prognoses of the postoperative symptom reduction in 94% of patients. The individual symptoms were accurately predicted in 70-91% of patients. Effective prognostication did not require data exceeding the routine preoperative investigations.Conclusions. Artificial neural networks are effective tools for the prediction of outcome after surgery for chronic rhinosinusitis and may be adopted in everyday clinical practice.},
 title={Artificial neural networks for prediction of outcome after surgery for chronic rhinosinusitis},
 type={Praca doktorska},
 keywords={chronic rhinosinusitis, artificial neural networks, modelling},
}