20.500.12556/RUL-188838 Odkrivanje morfoloških razlik med prečnimi prerezi mišičnih vlaken iztegnjenih in skrčenih mišic miši z računalniško analizo slik Identifying morphological differences of muscle fiber cross-sections in extended and contracted mouse muscles using computer image analysis Cilj tega diplomskega dela je bil ugotoviti, ali je mogoče s strojnim učenjem prepoznati in interpretirati morfološke razlike med prečnimi prerezi mišičnih vlaken pri iztegnjenih in skrčenih mišicah. Za namen raziskave so bili uporabljeni podatki petih miši. Morfološke lastnosti prečnih prerezov mišičnih vlaken, primerne za tovrstno analizo, so bile pridobljene s pomočjo strojne in človeške segmentacije mikroskopskih slik celotnega prereza mišice gastroknemius v iztegnjenem in skrčenem stanju. Po predobdelavi podatkov je bila z uporabo ugnezdenega navzkrižnega preverjanja preverjena uspešnost različnih vrst modelov, nakar je bila za nadaljnjo analizo zaradi najvišje povprečne uravnotežene točnosti in interpretabilnosti izbrana logistična regresija. Ker so rezultati podprli, da modeli na analiziranih podatkih prepoznajo morfometrične razlike med prečnimi prerezi mišičnih vlaken v prej navedenih stanjih, je bila nato potrjena statistična značilnost s pomočjo permutacijskega testa. The aim of this thesis was to determine whether machine learning can be used to identify and interpret morphological differences between muscle fiber cross-sections in extended and contracted muscles. For the purpose of the study data was obtained from five mice specimens. Morphological features of muscle fiber cross-sections suitable for this type of analysis were obtained through machine and human segmentation of microscopic images of the entire cross-section of the gastrocnemius muscle in both extended and contracted states. After preprocessing the data, the performance of various model types was evaluated using nested cross-validation, after which logistic regression was selected for further analysis due to its high average balanced accuracy and interpretability. Since the results supported the finding that the models could identify morphometric differences between muscle fiber cross-sections in the aforementioned states based on the analysed data, statistical significance was then tested and confirmed by using a permutation test. morfologija prečni prerezi mišičnih vlaken segmentacija logistična regresija ugnezdeno navzkrižno preverjanje permutacijski test morphology muscle fiber cross-sections segmentation logistic regression nested cross-validation permutation test true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2026-09-29 09:20:03 2026-09-29 09:20:07 2026-09-30 04:29:19 0000-00-00 00:00:00 2026 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 38916 Veingerl_Tijan_-_Odkrivanje_morfoloskih_razlik_med_precnimi_prerezi_misicnih_vlaken_iztegnjenih_.pdf Veingerl_Tijan_-_Odkrivanje_morfoloskih_razlik_med_precnimi_prerezi_misicnih_vlaken_iztegnjenih_.pdf 1 5F225ECDB0B0DD94D15BF74CE1320181 3c02e82588cdd2b8eaeb1c6b76f9a304e1f30e52ec8cc1b054559edd0dd71700 1b24639b-bbd6-11f1-8bc5-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=249197 Fakulteta za računalništvo in informatiko 0 0 0