scholarly journals Keratoconus detection of changes using deep learning of colour-coded maps

2021 ◽  
Vol 6 (1) ◽  
pp. e000824
Author(s):  
Xu Chen ◽  
Jiaxin Zhao ◽  
Katja C Iselin ◽  
Davide Borroni ◽  
Davide Romano ◽  
...  
Keyword(s):  
Neural Networks ◽  
New Zealand ◽  
University Hospital ◽  
Healthy Controls ◽  
Scheimpflug Camera ◽  
Stage 1 ◽  
Multicentre Retrospective Study ◽  
Eye Clinic ◽  
Corneal Tomography ◽  
Testing Set

ObjectiveTo evaluate the accuracy of convolutional neural networks technique (CNN) in detecting keratoconus using colour-coded corneal maps obtained by a Scheimpflug camera.DesignMulticentre retrospective study.Methods and analysisWe included the images of keratoconic and healthy volunteers’ eyes provided by three centres: Royal Liverpool University Hospital (Liverpool, UK), Sedaghat Eye Clinic (Mashhad, Iran) and The New Zealand National Eye Center (New Zealand). Corneal tomography scans were used to train and test CNN models, which included healthy controls. Keratoconic scans were classified according to the Amsler-Krumeich classification. Keratoconic scans from Iran were used as an independent testing set. Four maps were considered for each scan: axial map, anterior and posterior elevation map, and pachymetry map.ResultsA CNN model detected keratoconus versus health eyes with an accuracy of 0.9785 on the testing set, considering all four maps concatenated. Considering each map independently, the accuracy was 0.9283 for axial map, 0.9642 for thickness map, 0.9642 for the front elevation map and 0.9749 for the back elevation map. The accuracy of models in recognising between healthy controls and stage 1 was 0.90, between stages 1 and 2 was 0.9032, and between stages 2 and 3 was 0.8537 using the concatenated map.ConclusionCNN provides excellent detection performance for keratoconus and accurately grades different severities of disease using the colour-coded maps obtained by the Scheimpflug camera. CNN has the potential to be further developed, validated and adopted for screening and management of keratoconus.

scholarly journals Oxidative stress in common variable immunodeficiency

2021 ◽  
Vol 19 ◽  
pp. 205873922110024
Author(s):  
Sevgen Tanir Basaranoglu ◽  
Sukru Cekic ◽  
Emine Kirhan ◽  
Melahat Dirican ◽  
Sara S. Kilic
Keyword(s):  
Oxidative Stress ◽  
Common Variable Immunodeficiency ◽  
Chronic Inflammatory Demyelinating Polyneuropathy ◽  
Infectious Complications ◽  
Clinical Syndrome ◽  
University Hospital ◽  
Unknown Etiology ◽  
Healthy Controls ◽  
Oxidative Stress Biomarkers ◽  
Common Variable

Common variable immunodeficiency (CVID) is a heterogenous group of immunologic disorders of unknown etiology. Alterations of the normal cellular balance due to an increase in reactive oxygen species and/or decrease in antioxidant defense may lead to increased oxidative stress. We aimed to evaluate the levels of oxidative stress biomarkers in patients with CVID who had different presentations. We investigated the serum catalase (CAT), erythrocyte superoxide dismutase (SOD), erythrocyte reduced glutathione as antioxidants and serum malondialdehyde levels as lipid peroxidation marker in patients with CVID in Uludag University Hospital Department of Pediatric Allergy and Immunology’s outpatient clinics. In the analysis, there were 21 patients and 27 matched healthy controls. The median levels of CAT in patients with CVID was significantly lower than in healthy controls ( p = 0.04). Among the patients with CVID, 19% had autoimmune disease, one had Sjögren’s syndrome, one had autoimmune alopecia, one had juvenile rheumatoid arthritis, and one had chronic inflammatory demyelinating polyneuropathy. Patients with autoimmune complications had significantly lower CAT levels compared to the ones without autoimmune diseases ( p = 0.03). The patients without non-infectious complications (NICs) had lower SOD levels than the patients with NICs ( p = 0.05). The analysis of oxidative stress markers in the patients with CVID suggested a series of abnormalities in the anti-oxidant system. The clinical syndrome associations may be a useful tool for future studies to set prediction markers for the prognosis of patients with CVID.

scholarly journals Is suicidal ideation associated with allergic asthma and allergic rhinitis?

2018 ◽  
Vol 44 (1) ◽  
pp. 31-35 ◽  
Author(s):  
Martín Bedolla-Barajas ◽  
Norma Angélica Pulido-Guillén ◽  
Bolívar Vivar-Aburto ◽  
Jaime Morales-Romero ◽  
José Raúl Ortiz-Peregrina ◽  
...  
Keyword(s):  
Allergic Rhinitis ◽  
Suicidal Ideation ◽  
Allergic Asthma ◽  
Allergic Diseases ◽  
Cross Sectional Study ◽  
University Hospital ◽  
Control Group ◽  
Healthy Controls ◽  
Cross Sectional ◽  
And Control

ABSTRACT Objective: To investigate whether there is an association between suicidal ideation (SI) and allergic diseases in adults. Methods: This was a comparative cross-sectional study involving individuals ranging from 20 to 50 years of age recruited from a university hospital in the city of Guadalajara, Mexico. We included patients with a confirmed diagnosis of allergic asthma, those with a confirmed diagnosis of allergic rhinitis, and healthy controls. All subjects completed the Beck Depression Inventory-II (BDI-II), which includes an item that evaluates the presence of suicidal thoughts or desires within the last two weeks, in order to identify SI. Results: The sample comprised 115 patients with allergic asthma, 111 patients with allergic rhinitis, and 96 healthy controls. The number of individuals identified with SI in the three groups were, respectively, 17 (14.8%), 13 (11.7%), and 8 (8.3%). Regarding the presence of SI, no statistically significant association was found in the allergic asthma group (OR = 1.98; 95% CI: 0.78-4.64; p = 0.154) or in the allergic rhinitis group (OR = 1.46; 95% CI: 0.58-3.68; p = 0.424) when they were compared with the control group. However, the presence of depression was associated with SI in the three groups: allergic asthma (OR = 12.36; 95% CI: 2.67-57.15; p = 0.001); allergic rhinitis (OR = 6.20; 95% CI: 1.66-23.14; p = 0.006); and control (OR = 21.0; 95% CI: 3.75-117.36; p < 0,001). Conclusions: In comparison with the control group, no association was found between SI and the groups with allergic diseases. In contrast, there was association between SI and depression in the three groups.

scholarly journals Differentiation Between Anteroposterior and Posteroanterior Chest X-Ray View Position With Convolutional Neural Networks

2020 ◽  
Author(s):  
René Hosch ◽  
Lennard Kroll ◽  
Felix Nensa ◽  
Sven Koitka
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Performance Comparison ◽  
University Hospital ◽  
Hospital Data ◽  
X Ray ◽  
Ensemble Models ◽  
Class Distribution ◽  
Meta Information ◽  
Chest X Ray

Purpose Detection and validation of the chest X-ray view position with use of convolutional neural networks to improve meta-information for data cleaning within a hospital data infrastructure. Material and Methods Within this paper we developed a convolutional neural network which automatically detects the anteroposterior and posteroanterior view position of a chest radiograph. We trained two different network architectures (VGG variant and ResNet-34) with data published by the RSNA (26 684 radiographs, class distribution 46 % AP, 54 % PA) and validated these on a self-compiled dataset with data from the University Hospital Essen (4507, radiographs, class distribution 55 % PA, 45 % AP) labeled by a human reader. For visualization and better understanding of the network predictions, a Grad-CAM was generated for each network decision. The network results were evaluated based on the accuracy, the area under the curve (AUC), and the F1-score against the human reader labels. Also a final performance comparison between model predictions and DICOM labels was performed. Results The ensemble models reached accuracy and F1-scores greater than 95 %. The AUC reaches more than 0.99 for the ensemble models. The Grad-CAMs provide insight as to which anatomical structures contributed to a decision by the networks which are comparable with the ones a radiologist would use. Furthermore, the trained models were able to generalize over mislabeled examples, which was found by comparing the human reader labels to the predicted labels as well as the DICOM labels. Conclusion The results show that certain incorrectly entered meta-information of radiological images can be effectively corrected by deep learning in order to increase data quality in clinical application as well as in research. Key Points:  Citation Format

Association of Elevated Circulating Proteasome Levels at Diagnosis with Poor Outcome in Multiple Myeloma

Blood ◽  
2008 ◽  
Vol 112 (11) ◽  
pp. 5140-5140
Author(s):  
Peter Liebisch ◽  
Stephan Sixt ◽  
Hagen S Bachmann ◽  
Ulrich H Frey ◽  
Juergen Peters ◽  
...  
Keyword(s):  
Multiple Myeloma ◽  
Observation Interval ◽  
Prognostic Significance ◽  
Progression Free Survival ◽  
Ubiquitin Proteasome System ◽  
University Hospital ◽  
Rank Test ◽  
Healthy Controls ◽  
Serum Samples ◽  
Poor Outcome

Abstract Background: The proteasome is a proteolytic complex for intracellular degradation of ubiquitinated proteins which are involved in cell cycle regulation and apoptosis. A constitutively increased proteasome activity has been found in myeloma cells. Recent studies have shown that inhibition of the ubiquitin-proteasome system can be successfully used as a targeted therapy in multiple myeloma (MM). Recent data suggest a significant correlation between circulating proteasome levels (CPL) and outcome in patients with MM. Therefore, we investigated CPL in 110 patients in order to assess the role of CPL in MM. Experimental design: CPL were measured in serum samples from healthy controls (N=10) as well as from patients with monoclonal gammopathies of undetermined significance (N=27), indolent MM (N=15) and symptomatic MM (N=68) using enzyme-linked immunoabsorbent assay (ELISA) techniques detecting circulating 20S proteasome components. All serum samples were collected at the University Hospital in Ulm at time of diagnosis. Results: The median CPL were 123.5 ng/mL (range, 95–185 ng/mL) in healthy controls, 180 ng/mL (range, 100–485 ng/mL) in patients with MGUS (N=27) or indolent MM (N=15), and 227.5 ng/mL (range, 100–985 ng/mL) in patients with symptomatic MM (N=70). The CPL of patients with symptomatic MM were significantly elevated compared with healthy donors (p=0.0017) and to persons with asymptomatic gammopathies (p=0.046). While CPL were also significantly higher in the MGUS/indolent MM cohort as compared to controls (p=0.03), CPL in MGUS and indolent MM were comparable. Using ROC analysis in the symptomatic MM cohort patients with CPL &gt;150ng/mL (N=50) had a significantly shorter progression-free survival (PFS) time than patients (N=18) with CPL&lt;150 ng/mL levels (median PFS: 40 versus 57 months, log rank test p=0.026). Of note, all six patients who died in the observation interval had CPL &gt;150ng/mL. Conclusions: Here we demonstrate that increased CPL at diagnosis correlates with poor outcome in symptomatic MM patients. Evaluation of the prognostic significance of CPL in a larger cohort of uniformly treated patients with symptomatic MM is currently under way. This data will be presented at the meeting.

Symmetry and Antisymmetry in Maori Rafter Designs

1985 ◽  
Vol 3 (1) ◽  
pp. 23-45 ◽  
Author(s):  
J. D. H. Donnay ◽  
Gabrielle Donnay
Keyword(s):  
Group Theory ◽  
New Zealand ◽  
Rotation Axis ◽  
Translational Symmetry ◽  
Original Design ◽  
Main Design ◽  
Geometrical Symmetry ◽  
Color Scheme ◽  
Stage 1 ◽  
Two Stages

The Maori settled in New Zealand in the eighth or ninth centuries A.D.; the Europeans, many centuries later. Maori meeting houses are very ornate: besides carved wood figures, painted black-red-white designs embellish rafters and structural posts. These designs stand out as highly sophisticated examples of the twenty-four known band groups of 2-dimensional 1-translational symmetry G21 and antisymmetry (G21)′ The two independent symmetry elements, periodically repeated by lattice p or antilattice p′, are any two of the following: mirror m and antimirror m′, both either transverse or longitudinal; glide plane a or antiglide plane a′, longitudinal only; and two-fold rotation axis 2 or anti-axis 2′. Pseudosymmetry is common and obviously intentional: desymmetrization is achieved by added motifs, inserted along the edges of the band, at cell boundaries or inside the main design, which possess their own symmetries, always lower than that of the original design. The resulting symmetry of the whole is the symmetry common to the superposed patterns. Desymmetrization is also obtained by topological deformations of the design or deviations from the color scheme imposed by antisymmetry. Such irregularities led to Herbert Williams' hypothesis that the artist worked in two stages: tracing the outline and working in the color [1]. This conjecture finds support in group theory: the geometrical symmetry group involving no change of color (stage 1) is turned into an antisymmetry group (stage 2), as red goes to black and black goes to red under antisymmetry operations. The international symmetry notation of crystallographers is shown to be well suited to describe the Maori designs.

Using artificial neural networks to model the suitability of coastline for breeding by New Zealand fur seals (Arctocephalus forsteri)

2002 ◽  
Vol 148 (2) ◽  
pp. 111-131 ◽  
Author(s):  
Corey J.A Bradshaw ◽  
Lloyd S Davis ◽  
Martin Purvis ◽  
Qingqing Zhou ◽  
George L Benwell
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
New Zealand ◽  
Fur Seals ◽  
Artificial Neural
Sign in / Sign up

Export Citation Format

Share Document