RESUMO
Hyperfluorescence (HF) and reduced autofluorescence (RA) are important biomarkers in fundus autofluorescence images (FAF) for the assessment of health of the retinal pigment epithelium (RPE), an important indicator of disease progression in geographic atrophy (GA) or central serous chorioretinopathy (CSCR). Autofluorescence images have been annotated by human raters, but distinguishing biomarkers (whether signals are increased or decreased) from the normal background proves challenging, with borders being particularly open to interpretation. Consequently, significant variations emerge among different graders, and even within the same grader during repeated annotations. Tests on in-house FAF data show that even highly skilled medical experts, despite previously discussing and settling on precise annotation guidelines, reach a pair-wise agreement measured in a Dice score of no more than 63-80% for HF segmentations and only 14-52% for RA. The data further show that the agreement of our primary annotation expert with herself is a 72% Dice score for HF and 51% for RA. Given these numbers, the task of automated HF and RA segmentation cannot simply be refined to the improvement in a segmentation score. Instead, we propose the use of a segmentation ensemble. Learning from images with a single annotation, the ensemble reaches expert-like performance with an agreement of a 64-81% Dice score for HF and 21-41% for RA with all our experts. In addition, utilizing the mean predictions of the ensemble networks and their variance, we devise ternary segmentations where FAF image areas are labeled either as confident background, confident HF, or potential HF, ensuring that predictions are reliable where they are confident (97% Precision), while detecting all instances of HF (99% Recall) annotated by all experts.
RESUMO
Phosphorylated alpha-synuclein (p-α-syn) containing Lewy bodies (LBs) and Lewy neurites (LNs) are neuropathological hallmarks of Parkinson's disease (PD) in the central nervous system (CNS). Since they have been also demonstrated in the enteric nervous system (ENS) of PD patients, the aim of the study was to analyze enteric p-α-syn positive aggregates and intestinal gene expression. Submucosal rectal biopsies were obtained from patients with PD and controls and processed for dual-label-immunohistochemistry for p-α-syn and PGP 9.5. p-α-syn positive aggregates in nerve fibers and neuronal somata were subjected to a morphometric analysis. mRNA expression of α-syn and dopaminergic, serotonergic, VIP (vaso intestinal peptide) ergic, cholinergic, muscarinergic neurotransmitter systems were investigated using qPCR. Frequency of p-α-syn positive nerve fibers was comparable between PD and controls. Although neuronal p-α-syn positive aggregates were detectable in both groups, total number and area of p-α-syn positive aggregates were increased in PD patients as was the number of small and large sized aggregates. Increased expression of dopamine receptor D1, VIP and serotonin receptor 3A was observed in PD patients, while serotonin receptor 4 and muscarinic receptor 3 (M3R) were downregulated. M3R expression correlated negative with the number of small sized p-α-syn positive aggregates. The findings strengthen the hypothesis that the CNS pathology of increased p-α-syn in PD also applies to the ENS, if elaborated morphometry is applied and give further insights in altered intestinal gene expression in PD. Although the mere presence of p-α-syn positive aggregates in the ENS should not be regarded as a criterion for PD diagnosis, elaborated morphometric analysis of p-α-syn positive aggregates in gastrointestinal biopsies could serve as a suitable tool for in-vivo diagnosis of PD.