- Browse by Subject
Browsing by Subject "Digital image analysis"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Automatic classification of white regions in liver biopsies by supervised machine learning(Elsevier, 2014-04) Vanderbeck, Scott; Bockhorst, Joseph; Komorowski, Richard; Kleiner, David E.; Gawrieh, Samer; Medicine, School of MedicineAutomated assessment of histological features of non-alcoholic fatty liver disease (NAFLD) may reduce human variability and provide continuous rather than semiquantitative measurement of these features. As part of a larger effort, we perform automatic classification of steatosis, the cardinal feature of NAFLD, and other regions that manifest as white in images of hematoxylin and eosin-stained liver biopsy sections. These regions include macrosteatosis, central veins, portal veins, portal arteries, sinusoids and bile ducts. Digital images of hematoxylin and eosin-stained slides of 47 liver biopsies from patients with normal liver histology (n = 20) and NAFLD (n = 27) were obtained at 20× magnification. The images were analyzed using supervised machine learning classifiers created from annotations provided by two expert pathologists. The classification algorithm performs with 89% overall accuracy. It identified macrosteatosis, bile ducts, portal veins and sinusoids with high precision and recall (≥ 82%). Identification of central veins and portal arteries was less robust but still good. The accuracy of the classifier in identifying macrosteatosis is the best reported. The accurate automated identification of macrosteatosis achieved with this algorithm has useful clinical and research-related applications. The accurate detection of liver microscopic anatomical landmarks may facilitate important subsequent tasks, such as localization of other histological lesions according to liver microscopic anatomy.Item Orange/Red Fluorescence of Active Caries by Retrospective Quantitative Light-Induced Fluorescence Image Analysis(Karger, 2016-05) Gomez, G.F.; Eckert, G.J.; Ferreira Zandona, A.; Biostatistics, School of Public HealthThis retrospective clinical study determined the association of caries activity and orange/red fluorescence on quantitative light-induced fluorescence (QLF) images of surfaces that progressed to cavitation, as determined by clinical visual examination. A random sample of QLF images from 565 children (5-13 years) previously enrolled in a longitudinal study was selected. Buccal, lingual and occlusal surface images obtained after professional brushing at baseline and every 4 months over a 4-year period were analyzed for red fluorescence. Surfaces that progressed (n = 224) to cavitation according to the International Caries Detection and Assessment System (ICDAS 0/1/2/3/4 to 5/6 or filling), and surfaces that did not progress (n = 486) were included. QA2 image analysis software outputs the percentage increase of the red/green components as x0394;R and area of x0394;R (areax0394;R) at different thresholds. Mixed-model ANOVA was used to compare progressive and nonprogressive surfaces to account for correlations of red fluorescence (x0394;R and areax0394;R) between surfaces within a subject. The first analysis used the first observation for each surface or the first available visit if the surface was unerupted (baseline), while the second analysis used the last observation prior to cavitation for surfaces that progressed and the last observation for surfaces that did not progress (final). There was a significant (p < 0.05) association between red fluorescence and progression to cavitation at thresholds x0394;R0, x0394;R10, x0394;R20, x0394;R60, x0394;R70, x0394;R80, x0394;R90 and x0394;Rmax at baseline and for x0394;R0 and x0394;R10 at the final observation. Quantification of orange/red fluorescence may help to identify lesions that progress to cavitation. Future studies identifying microbiological factors causing orange/ red fluorescence and its caries activity are indicated.