Naslov
Using deep-learning artificial intelligence to resolve the mechanistic relationship between Alzheimer's disease and gut microbiota

Autori
Šimić, Goran

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

Izvornik
Book of Abstracts / - , 2017, 31-31

Skup
6th Croatian Neuroscience Congress

Mjesto i datum
Osijek, Hrvatska, 16.09.2017. - 18.09.2017

Vrsta sudjelovanja
Pozvano predavanje

Vrsta recenzije
Domaća recenzija

Ključne riječi
Alzheimer's disease ; gut microbiota ; deep-learning ; artificial intelligence ; nitric oxide synthase ; Lactobacillus ; serotonin ; ABC cholesterol transporters

Sažetak
Recently emerging “deep learning” systems and artificial neural networks need not be programmed exclusively with a human expert’s knowledge. Instead, they can learn on their own, often from big datasets that are far larger than humans can cope with, until they can see patterns. In two recently published studies (Gubiani et al., Expert Syst. Appl., 2017 ; Cestnik et al., Genomics Comp. Biol., 2017) authors used the literature mining tools of OntoGen to document clustering and CrossBee for cross-domain bridging term exploration to search for hidden relations in scientific papers from two domains of interest, Alzheimer’s disease and gut microbiome. The results supported the hypothesis of the neuroinflammatory nature of Alzheimer’s disease and identified several potential links: nitric oxide synthase, Lactobacillus, bile acid (ABC transporters), and serotonin. However, as advanced artificial intelligence systems cannot explain their logic and the computations that lead to an outcome remain hidden, I will make an attempt to open up the black box of machine thinking behind these discoveries, and provide insights about the mechanisms that led to the outcomes to help interpreting the associations observed.

Izvorni jezik
Engleski

Znanstvena područja
Biologija, Temeljne medicinske znanosti, Kliničke medicinske znanosti, Biotehnologija u biomedicini (prirodno područje, biomedicina i zdravstvo, biotehničko područje)

POVEZANOST RADA