| Abstract | Svrha ovog rada bila je procijeniti pouzdanost DIAGNOdent ureĎaja za detekciju
karijesa, in vivo i in vitro te vizualne ICDAS metode (The International Caries Detection
and Assessment System) u usporedbi s histološkim nalazom. Ispitivanje je provedeno na
krunama 40 ljudskih trajnih zuba, predviĎenih za vaĎenje zbog oteţanog nicanja,
ortodontskih ili parodontoloških razloga. Ispitane su 72 mjerne točke na okluzalnim
plohama koristeći ICDAS kriterije i DIAGNOdent pen ureĎaj. Nakon ekstrakcije,
ponovljena su mjerenja i zubi su histološki klasificirani. Podaci su statistički obraĎeni
Chi-kvadrat testom i MeĎuklasnim korelacijskim testom, pri razini značajnosti p<0,05.
Izračunata je osjetljivost, specifičnost i točnost metoda detekcije karijesa.
Ponovljivost mjerenja je bila najveća za DIAGNOdent in vitro uvjetima
(Interclass Correlation Coefficient - 0,963). PronaĎena je statistički značajna razlika
izmeĎu in vivo i in vitro mjerenja. Najsnaţnija korelacija izmeĎu ispitivane metode i
histološkog nalaza izraţena Cohenovom Kappom postignuta je kod vizualne ICDAS
metode (0,613), potom kod DIAGNOdenta in vivo (0,292 ) i isti ureĎaj in vitro (0,222).
Najtočnijom se pokazala ICDAS metoda (95,9%), slijedi DIAGNOdent in vivo (88,9%)
te in vitro (84,7%). ICDAS metoda pokazuje i veću specifičnost (76,9%) od
DIAGNOdent ureĎaja in vivo (62,5%) te in vitro (44,4%).
Temeljem ovog istraţivanja moţe se zaključiti da je ICDAS metoda pouzdanija u
dijagnostici karijesa u odnosu na istraţivani elektronski ureĎaj. |
| Abstract (english) | The purpose of this reasearch is to evaluate clinical perfomance of DIAGNOdent
device for caries detection, in vivo and in vitro, and the ICDAS method in caries
detection using histological gold standard. The study was made on occlusal surfaces of
40 human permanent teeth which were indicated for extraction because of teeth
erruption problems, orthodontic and periodontic reasons. Using the ICDAS criteria and
the DIAGNOdent pen device, 72 measurement points on occlusal surfaces were
examined. After a tooth extraction, measurements were repeated and the samples were
histologically evaluated. The results were statistically analyzed with the Chi-square test
and Interclasse correlation test, with significance level p<0.05. The sensitivity,
specificity and accuracy for these caries detection methods were calculated. The
repeatibility of measurements was highest for DIAGNOdent in vivo (Interclass
Correlation Coefficient – 0.963). The statistically significant difference was found
between in vivo and in vitro measurements. The strongest correlation between the
examined method and histological validation expressed by Cohen’s Kappa is achieved in
the visual ICDAS method (0.613), followed by DIAGNOdent in vivo (0.292) and in
vitro (0.222). The most accurate was the ICDAS method (95.9%), followed by
DIAGNOdent in vivo (88.9%) and in vitro (84.7%). The ICDAS method shows higher
specificity (76.9%) than the DIAGNOdent device in vivo (62.5%) and in vitro (44.4%).
Based on this research, the ICDAS method is more reliable in caries detection compared
to the tested electronic device |
