Title Umjetna inteligencija u dijagnostici COVID pozitivnih pacijenata
Title (english) Artificial intelligence in the diagnosis of covid positive patients
Author Dora Sladaković
Mentor Frane Mihanović (mentor)
Committee member Diana Aranza (predsjednik povjerenstva)
Committee member Tatjana Matijaš (član povjerenstva)
Committee member Frane Mihanović (član povjerenstva)
Granter University of Split (University Department of Health Studies) Split
Defense date and country 2022-02-17, Croatia
Scientific / art field, discipline and subdiscipline BIOMEDICINE AND HEALTHCARE Clinical Medical Sciences
Abstract Svjetska zdravstvena organizacija 11. ožujka 2020. proglasila je širenje novog Sars-CoV-2 virusa globalnom pandemijom što je izvršilo značajan pritisak i ukazalo na brojne slabosti zdravstvenih sustava. Najčešći simptomi COVID-19 infekcije uključuju povišenu tjelesnu temperaturu, umor, suhi kašalj, gubitak ili promjenu osjeta mirisa ili okusa, kao i teške respiratorne i druge komplikacije koje zahtijevaju hospitalizaciju ili smještanje u jedinice intenzivnog liječenja. Rano dijagnosticiranje presudno je za otkrivanje aktivnih slučajeva, kontrolu širenja infekcije i optimalno zbrinjavanje pacijenata. Zlatni standard za postavljanje dijagnoze je lančana reakcija polimeraze reverznom transkripcijom (RT-PCR). Glavne radiološke metode dijagnosticiranja su radiografsko snimanje pluća i kompjutorizirana tomografija, svaka sa svojim prednostima i ograničenjima.
U uvjetima svjetske zdravstvene krize, zdravstvenom sustavu potrebne su nove tehnologije za praćenje i kontrolu širenja pandemije. Jedna od takvih tehnologija je i umjetna inteligencija, te se u ovom radu pregledom literature nastojalo prikazati potencijalne prednosti sustava umjetne inteligencije koji bi znatno olakšali borbu protiv pandemije i tako utjecali na smanjenje pritiska na zdravstvene sustave diljem svijeta. Početna istraživanja usredotočila su se na dijagnostiku i prognozu infekcije pomoću radiografskih slika ili CT prikaza pluća. Sustavi umjetne inteligencije u tim su istraživanjima uspoređivani s tumačenjima iskusnih radiologa te su postigli obečavajuće rezultate. Ipak, većina razvijenih sustava ograničena je malim skupovima podataka, nije dostupna za javnu upotrebu, bez otvorenog pristupa i tako nedostupna istraživačkoj zajednici za nadogradnju.
Abstract (english) On March 11, 2020, the World Health Organization declared the spread of the new SARS-CoV-2 virus a global pandemic, which put a significant pressure and pointed to a number of weaknesses in health systems. The most common symptoms of COVID-19 infection include fever, fatigue, dry cough, loss or change in sense of smell and taste, as well as sever respiratory and other complications that require hospitalization or placement in intensive care units. Early diagnosis is crucial for detecting active cases, controlling the spread of infection and optimal care for the patients. Polymerase chain reaction by reverse transcription (RT-PCR) is considered the gold standard for diagnosis. The main radiological methods of diagnosis are radiographic imaging and computed tomography of the lungs, each with its advantages and limitations.
In the context of the global health crisis, the health system needs new technologies to monitor and control the spread of the pandemic. One such technology is artificial intelligence, and this paper reviewed the potential benefits of artificial intelligence systems that would greatly facilitate the fight against the pandemic and thus reduce the pressure on health systems around the world. Initial research focused on the diagnosis and prognosis of infection using radiographic images or CT scans of the lungs. Artificial intelligence systems in these studies were compared with the interpretations of experienced radiologists and achieved promising results. However, most developed systems are limited to small data sets, not available for public use and without open access, thus inaccessible to research community for upgrades.
Keywords
COVID-19
umjetna inteligencija
radiografsko snimanje
komputorizirana tomografija
Keywords (english)
COVID-19
artificial intelligence
radiographic imaging
computerized tomography
Language croatian
URN:NBN urn:nbn:hr:176:308613
Study programme Title: Radiologic Technology (university/graduate) Study programme type: university Study level: graduate Academic / professional title: magistar/magistra radiološke tehnologije (magistar/magistra radiološke tehnologije)
Type of resource Text
File origin Born digital
Access conditions Open access
Terms of use
Created on 2022-02-18 10:38:22