Naslov
Artificial intelligence based analysis of met-enkephalin induced immunomodulation in multiple sclerosis.

Autori
Štambuk, Nikola ; Mašić, Nikola ; Brinar, Vesna ; Štambuk, Vjera ; Rabatić, Sabina ; Marotti, Tanja ; Šverko, Višnja ; Marušić-Della Marina, Branka ; Brzović, Zdravko ; Zurak, Niko ; Svoboda-Beusan, Ivna ; Mažuran, Renata ; Karaman, Ksenija ; Gagro, Alenka ; Rudolf, Maja ; Malenica, Branko ; Konjevoda, Paško ; Pokrić, Biserka

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Book of Abstract Math/Chem/Comp 2000 / Graovac, Ante ; Plavšić, Dean ; Pokrić, Biserka ; Smrečki, Vilko - Zagreb : Math/Chem/Comp, 2000

Skup
Math/Chem/Comp 2000 - the 15th Dubrovnik International Course & Conference on the Interfaces between Mathematics, Chemistry and Computer Sciences.

Mjesto i datum
Dubrovnik, Hrvatska, 19.06.2000. - 24.06.2000

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
met-enkephalin; immunomodulation; multiple sclerosis; decision tree classifier; prognostic parameters
(met-enkephalin; immunomodulation; multiple sclerosis; decision tree classifier; prognostic parameters; neural network)

Sažetak
We present the analysis of met-enkephalin (LUPEX) induced effects on 52 different immunochemical parameters that define cellular and humoral immune response. In addition to the standard statistical analysis we used artificial intelligence data mining methods, i.e. nonlinear machine learning approach with C4.5 decision tree classifier and self organising artificial neural network. It is shown that artificial intelligence based methods of data analysis provide useful alternative to the standard statistical approach, enable the extraction of few relevant immunochemical parameters with their mutual relationships, and ensure accurate prediction of the response to drug treatment.

Izvorni jezik
Engleski

Znanstvena područja
Javno zdravstvo i zdravstvena zaštita, Farmacija

POVEZANOST RADA