20.500.12556/RUL-131381 Vpliv longitudinalnih meritev na zanesljivost algoritma za klasifikacijo mamografskih slik v gostotne skupine po sistemu BI-RADS Impact of longitudinal measurements on reliability of the algorithm for BI-RADS density classification of mammographic images Cilj: Mamografska gostota, ki odraža sestavo dojke, je dober neodvisni dejavnik za napoved ogroženosti za raka dojke. Cilj magistrske naloge je bila izdelava algoritma za napoved mamografske gostote na podlagi procesiranih mamografskih slik in preveriti hipotezo, da upoštevanje zaporednih meritev pomeni izboljšanje zanesljivosti napovedi gostotnega razreda. Podatki in metode: Uporabili smo slike iz podatkovne baze programa DORA. Za izdelavo modela smo uporabili 12190 procesiranih slik, posnetih na aparatih proizvajalca Siemens in 4787 procesiranih slik, posnetih na aparatih proizvajalca Hologic. Model smo uporabili na mamografskih slikah preiskovank, ki so bile slikane vsaj dvakrat, s povprečnim časom med slikanji 2,1 leti. Uporabili smo 34053 procesiranih slik, posnetih na aparatih proizvajalca Siemens in 12601 procesiranih slik, posnetih na aparatih proizvajalca Hologic. Za segmentacijo in izračun značilk smo uporabili program LIBRA. Za selekcijo značilk smo uporabili metodo MRMR, za klasifikator pa multinomsko logistično regresijo. Zanesljivost napovedi smo ocenili s pomočjo Cohenovega koeficienta $\kappa$. Rezultate smo primerjali z rezultati iz literature. Rezultati: Izboljšanje zanesljivosti napovedi mamografske gostote za kombinacijo referenčnih slik in slik, posnetih pred referenčnimi je bilo signifikantno in je znašalo od 0.06 $\pm$ 0.03 za najmanj oddaljene do 0.38 $\pm$ 0.29 za najbolj oddaljene meritve. Zanesljivost napovedi se z upoštevanjem slik, posnetih po referenčnih ni bistveno izboljšala, izboljšanje je znašalo 0.02 $\pm$ 0.01 za najmanj oddaljene in 0.34 $\pm$ 0.53 za najbolj oddaljene meritve. Zaključek: Naš model napovedovanja mamografske gostote je v najboljšem primeru dosegel močno, 0.64 < $\kappa$ < 0.81, v najslabšem pa srednjo, 0.35 < $\kappa$ < 0.63 napovedno moč. Ugotovili smo, da upoštevanje večkratnih zaporednih meritev signifikantno izboljša zanesljivost napovedi. Purpose: Mammographic density, which reflects breast tissue composition, is an important independent breast cancer risk factor. The purpose of this thesis was to build a model for mammographic density classification, based on the mammographic images and testing the hypothesis, that the model reliability can be improved by including longitudinal data in the study. Data and methods: We used images, that were a part of the DORA database. 12190 processed images, recorded with the Siemens mammographs and 4787 processed images, recoreded with the Hologic mammographs were used for making the model for classification. We used the model on the images of women, that were imaged at least twice, with average time between imaging 2.1 years. 34053 processed images, recorded with the Siemens mammographs and 12601 processed images, recorded with the Hologic mammographs were used. Segmentation and feature extraction were performed with the LIBRA software. Feature selection was performed with the MRMR method, and we used multinomial logistic regression for the classifier. Prediction reliability was assessed by calculating Cohen's $\kappa$ coefficient. We compared our results to results from literature. Results: Mammographic density prediction model's reliability was significantly improved by considering the reference images, and images taken prior to the reference. Improvement was 0.06 $\pm$ 0.03 for the closest and 0.38 $\pm$ 0.29 for the furthest measurements. Density predicion reliability was not significantly improved by considering images taken after the reference images. Improvement in that case was 0.02 $\pm$ 0.01 for the closest and 0.34 $\pm$ 0.53 for the furthest measurements. Conslusion: Our density prediction model reaches at most a substantial agreement, 0.64 < $\kappa$ 0.81, and at worst a moderate agreement, 0.35 < $\kappa$ < 0.63. We concluded that considering multiple longitudinal measurements can significantly improve our model's reliability. Mamografija mamografska gostota BI-RADS longitudinalne meritve Cohenov koeficient kappa Mammography mammographic density BI-RADS longitudinal data Cohen's kappa coefficient true false false Slovenski jezik Angleški jezik Magistrsko delo/naloga 2021-09-26 07:16:00 2021-09-26 07:16:21 2022-09-10 04:11:20 0000-00-00 00:00:00 2021 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 121014 78583555 3025.pdf 3025.pdf 1 71C1A936ABB31C86CFBEC86D6481E96F a66b651c90ac3a6ae33eaa17aef2c43c3f8192cf0da9fbcdf862de58f6c6c6d0 a1fb5289-1e88-11ec-a523-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=148695 Fakulteta za matematiko in fiziko 0 0 0