Browsing by Author "Hocevar, Alojzija (6506634484)"

Salivary gland ultrasonography (SGUS) has proven to be a promising tool for diagnosing various diseases manifesting with abnormalities in salivary glands (SGs), including primary Sjögren's syndrome (pSS). At present, the major obstacle for establishing SUGS as a standardized tool for pSS diagnosis is its low inter/intra observer reliability. The aim of this study was to address this problem by proposing a robust deep learning-based solution for the automated segmentation of SGUS images. For these purposes, four architectures were considered: a fully convolutional neural network, fully convolutional “DenseNets” (FCN-DenseNet) network, U-Net, and LinkNet. During the course of the study, the growing HarmonicSS cohort included 1184 annotated SGUS images. Accordingly, the algorithms were trained using a transfer learning approach. With regard to the intersection-over-union (IoU), the top-performing FCN-DenseNet (IoU = 0.85) network showed a considerable margin above the inter-observer agreement (IoU = 0.76) and slightly above the intra-observer agreement (IoU = 0.84) between clinical experts. Considering its accuracy and speed (24.5 frames per second), it was concluded that the FCN-DenseNet could have wider applications in clinical practice. Further work on the topic will consider the integration of methods for pSS scoring, with the end goal of establishing SGUS as an effective noninvasive pSS diagnostic tool. To aid this progress, we created inference (frozen models) files for the developed models, and made them publicly available. © 2020 Elsevier Ltd

Salivary gland ultrasonography (SGUS) has proven to be a promising tool for diagnosing various diseases manifesting with abnormalities in salivary glands (SGs), including primary Sjögren's syndrome (pSS). At present, the major obstacle for establishing SUGS as a standardized tool for pSS diagnosis is its low inter/intra observer reliability. The aim of this study was to address this problem by proposing a robust deep learning-based solution for the automated segmentation of SGUS images. For these purposes, four architectures were considered: a fully convolutional neural network, fully convolutional “DenseNets” (FCN-DenseNet) network, U-Net, and LinkNet. During the course of the study, the growing HarmonicSS cohort included 1184 annotated SGUS images. Accordingly, the algorithms were trained using a transfer learning approach. With regard to the intersection-over-union (IoU), the top-performing FCN-DenseNet (IoU = 0.85) network showed a considerable margin above the inter-observer agreement (IoU = 0.76) and slightly above the intra-observer agreement (IoU = 0.84) between clinical experts. Considering its accuracy and speed (24.5 frames per second), it was concluded that the FCN-DenseNet could have wider applications in clinical practice. Further work on the topic will consider the integration of methods for pSS scoring, with the end goal of establishing SGUS as an effective noninvasive pSS diagnostic tool. To aid this progress, we created inference (frozen models) files for the developed models, and made them publicly available. © 2020 Elsevier Ltd

Objective. Ultrasonography (US) is a sensitive tool in the diagnosis of major salivary gland abnormalities in primary Sjögren's syndrome (pSS). The aim of this systematic review was to assess the metric properties of this technique. Methods. PUBMED and EMBASE databases were searched. All publications between January 1988 and January 2013 were considered. Data were extracted from the articles meeting the inclusion criteria according to US definition of salivary gland scoring system and metric properties studied. The type and number of glands tested, study design and metric properties according to OMERACT filter (truth, discrimination, feasibility) were assessed. Results. Of 167 publications identified initially with PUBMED and EMBASE, 31 met the inclusion criteria. The number of pSS patients varied among the studies from 16 to 140. The diagnosis of pSS was in line in most of the cases with the American-European Consensus Group (AECG) classification criteria for Sjögren's syndrome. The US examination was performed in suspected pSS only in studies in which the sensitivity ranged from 45.8 to 91.6% and specificity from 73 to 98.1%. There was heterogeneity in regard to the definition of US in B-mode and few studies used US in colour Doppler. Few studies reported reliability of US and sensitivity to change in pSS. Conclusion. US is a valuable tool for detecting salivary gland abnormalities in pSS. Its reliability has been poorly investigated and there is considerable variation in the definition of US abnormalities. Further studies are required to validate and standardize the US definition of salivary gland in pSS. © The Author 2015.

Salivary gland ultrasonography (SGUS) has shown good potential in the diagnosis of primary Sjögren's syndrome (pSS). However, a series of international studies have reported needs for improvements of the existing pSS scoring procedures in terms of inter/intra observer reliability before being established as standardized diagnostic tools. The present study aims to solve this problem by employing radiomics features and artificial intelligence (AI) algorithms to make the pSS scoring more objective and faster compared to human expert scoring. The assessment of AI algorithms was performed on a two-centric cohort, which included 600 SGUS images (150 patients) annotated using the original SGUS scoring system proposed in 1992 for pSS. For each image, we extracted 907 histogram-based and descriptive statistics features from segmented salivary glands. Optimal feature subsets were found using the genetic algorithm based wrapper approach. Among the considered algorithms (seven classifiers and five regressors), the best preforming was the multilayer perceptron (MLP) classifier (κ = 0.7). The MLP over-performed average score achieved by the clinicians (κ = 0.67) by the considerable margin, whereas its reliability was on the level of human intra-observer variability (κ = 0.71). The presented findings indicate that the continuously increasing HarmonicSS cohort will enable further advancements in AI-based pSS scoring methods by SGUS. In turn, this may establish SGUS as an effective noninvasive pSS diagnostic tool, with the final goal to supplement current diagnostic tests. © 2013 IEEE.

Salivary gland ultrasonography (SGUS) has shown good potential in the diagnosis of primary Sjögren's syndrome (pSS). However, a series of international studies have reported needs for improvements of the existing pSS scoring procedures in terms of inter/intra observer reliability before being established as standardized diagnostic tools. The present study aims to solve this problem by employing radiomics features and artificial intelligence (AI) algorithms to make the pSS scoring more objective and faster compared to human expert scoring. The assessment of AI algorithms was performed on a two-centric cohort, which included 600 SGUS images (150 patients) annotated using the original SGUS scoring system proposed in 1992 for pSS. For each image, we extracted 907 histogram-based and descriptive statistics features from segmented salivary glands. Optimal feature subsets were found using the genetic algorithm based wrapper approach. Among the considered algorithms (seven classifiers and five regressors), the best preforming was the multilayer perceptron (MLP) classifier (κ = 0.7). The MLP over-performed average score achieved by the clinicians (κ = 0.67) by the considerable margin, whereas its reliability was on the level of human intra-observer variability (κ = 0.71). The presented findings indicate that the continuously increasing HarmonicSS cohort will enable further advancements in AI-based pSS scoring methods by SGUS. In turn, this may establish SGUS as an effective noninvasive pSS diagnostic tool, with the final goal to supplement current diagnostic tests. © 2013 IEEE.

Objectives Ultrasonography (US) is sensitive for detecting echostructural abnormalities of the major salivary glands (SGs) in primary Sjögren's syndrome (pSS). Our objectives were to define selected US-SG echostructural abnormalities in pSS, set up a preliminary atlas of these definitions and evaluate the consensual definitions reliability in both static and acquisition US-SG images. Methods International experts in SG US in pSS participated in consensus meetings to select and define echostructural abnormalities in pSS. The US reliability of detecting these abnormalities was assessed using a two-step method. First 12 experts used a web-based standardised form to evaluate 60 static US-SG images. Intra observer and interobserver reliabilities were expressed in κ values. Second, five experts, who participated all throughout the study, evaluated US-SG acquisition interobserver reliability in pSS patients. Results Parotid glands (PGs) and submandibular glands (SMGs) intra observer US reliability on static images was substantial (κ > 0.60) for the two main reliable items (echogenicity and homogeneity) and for the advised pSS diagnosis. PG inter observer reliability was substantial for homogeneity. SMGs interobserver reliability was moderate for homogeneity (κ = 0.46) and fair for echogenicity (κ = 0.38). On acquisition images, PGs interobserver reliability was substantial (κ > 0.62) for echogenicity and moderate (κ = 0.52) for homogeneity. The advised pSS diagnosis reliability was substantial (κ = 0.66). SMGs interobserver reliability was fair (0.20< κ ≤ 0.40) for echogenicity and homogeneity and either slight or poor for all other US core items. Conclusion This work identified two most reliable US-SG items (echogenicity and homogeneity) to be used by US-SG trained experts. US-PG interobserver reliability result for echogenicity is in line with diagnosis of pSS. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Objectives Ultrasonography (US) is sensitive for detecting echostructural abnormalities of the major salivary glands (SGs) in primary Sjögren's syndrome (pSS). Our objectives were to define selected US-SG echostructural abnormalities in pSS, set up a preliminary atlas of these definitions and evaluate the consensual definitions reliability in both static and acquisition US-SG images. Methods International experts in SG US in pSS participated in consensus meetings to select and define echostructural abnormalities in pSS. The US reliability of detecting these abnormalities was assessed using a two-step method. First 12 experts used a web-based standardised form to evaluate 60 static US-SG images. Intra observer and interobserver reliabilities were expressed in κ values. Second, five experts, who participated all throughout the study, evaluated US-SG acquisition interobserver reliability in pSS patients. Results Parotid glands (PGs) and submandibular glands (SMGs) intra observer US reliability on static images was substantial (κ > 0.60) for the two main reliable items (echogenicity and homogeneity) and for the advised pSS diagnosis. PG inter observer reliability was substantial for homogeneity. SMGs interobserver reliability was moderate for homogeneity (κ = 0.46) and fair for echogenicity (κ = 0.38). On acquisition images, PGs interobserver reliability was substantial (κ > 0.62) for echogenicity and moderate (κ = 0.52) for homogeneity. The advised pSS diagnosis reliability was substantial (κ = 0.66). SMGs interobserver reliability was fair (0.20< κ ≤ 0.40) for echogenicity and homogeneity and either slight or poor for all other US core items. Conclusion This work identified two most reliable US-SG items (echogenicity and homogeneity) to be used by US-SG trained experts. US-PG interobserver reliability result for echogenicity is in line with diagnosis of pSS. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.