The huge number of images shared on the Web makes effective cataloguing methods for efficient storage and retrieval procedures specifically tailored on the end-user needs a very demanding and crucial issue. In this paper, we investigate the applicability of Automatic Image Annotation (AIA) for image tagging with a focus on the needs of database expansion for a news broadcasting company. First, we determine the feasibility of using AIA in such a context with the aim of minimizing an extensive retraining whenever a new tag needs to be incorporated in the tag set population. Then, an image annotation tool integrating a Convolutional Neural Network model (AlexNet) for feature extraction and a K-Nearest-Neighbours classifier for tag assignment to images is introduced and tested. The obtained performances are very promising addressing the proposed approach as valuable to tackle the problem of image tagging in the framework of a broadcasting company, whilst not yet optimal for integration in the business process.
DOCUMENT
AimTo discuss the actual public image of nurses and other factors that influence the development of nurses' self-concept and professional identity.BackgroundNurses have become healthcare professionals in their own right who possess a great deal of knowledge. However, the public does not always value the skills and competences nurses have acquired through education and innovation.DesignDiscussion paper.Data sourcesWe identified 1216 relevant studies by searching MEDLINE, CINAHL and PsycINFO databases in the period 1997–2010. Finally, 18 studies met our inclusion criteria.DiscussionThe included studies show that the actual public image of nursing is diverse and incongruous. This image is partly self-created by nurses due to their invisibility and their lack of public discourse. Nurses derive their self-concept and professional identity from their public image, work environment, work values, education and traditional social and cultural values.Implications for nursingNurses should work harder to communicate their professionalism to the public. Social media like the Internet and YouTube can be used to show the public what they really do.ConclusionTo improve their public image and to obtain a stronger position in healthcare organizations, nurses need to increase their visibility. This could be realized by ongoing education and a challenging work environment that encourages nurses to stand up for themselves. Furthermore, nurses should make better use of strategic positions, such as case manager, nurse educator or clinical nurse specialist and use their professionalism to show the public what their work really entails.
LINK
In contemporary image databases one finds many images with the same image content but perturbed by zooming, scaling, rotation etc. For the purpose of image recognition in such databases we employ features based on statistics stemming from fractal transforms gray-scale images. We show how the features derived from these statistical aspects can be made invariant to zooming or rescaling. A feature invariance measure is defined and described. The method is especially suitable for images of textures. We produce numerical results which validate the approach.
DOCUMENT
This paper reports on an experiment comparing students’ results on image-rich numeracy problems and on equivalent word problems. Given the well reported problematic nature of word problems, the hypothesis is that students score better on image-rich numeracy problems than on comparable word problems. To test the hypothesis a randomized controlled trial was conducted with 31,842 students from primary, secondary, and vocational education. The trial consisted of 21 numeracy problems in two versions: word problems and image-rich problems. The hypothesis was confirmed for the problems used in this experiment. With the insights gained we intend to improve the assessment of students’ abilities in solving quantitative problems from daily life. Numeracy, word problem, image-rich problem, randomized controlled trial, assessment
DOCUMENT
Objective: Summarize all relevant findings in published literature regarding the potential dose reduction related to image quality using Sinogram-Affirmed Iterative Reconstruction (SAFIRE) compared to Filtered Back Projection (FBP).Background: Computed Tomography (CT) is one of the most used radiographic modalities in clinical practice providing high spatial and contrast resolution. However it also delivers a relatively high radiation dose to the patient. Reconstructing raw-data using Iterative Reconstruction (IR) algorithmshas the potential to iteratively reduce image noise while maintaining or improving image quality of low dose standard FBP reconstructions. Nevertheless, long reconstruction times made IR unpractical for clinical use until recently.Siemens Medical developed a new IR algorithm called SAFIRE, which uses up to 5 different strength levels, and poses an alternative to the conventional IR with a significant reconstruction time reduction.Methods: MEDLINE, ScienceDirect and CINAHL databases were used for gathering literature. Eleven articles were included in this review (from 2012 to July 2014).Discussion: This narrative review summarizes the results of eleven articles (using studies on both patients and phantoms) and describes SAFIRE strengths for noise reduction in low dose acquisitions while providing acceptable image quality.Conclusion: Even though the results differ slightly, the literature gathered for this review suggests that the dose in current CT protocols can be reduced at least 50% while maintaining or improving image quality. There is however a lack of literature concerning paediatric population (with increased radiationsensitivity). Further studies should also assess the impact of SAFIRE on diagnostic accuracy.
DOCUMENT
The web is widely used by museums as a low-barrier platform to inform people on activities in the museum and publish their collections online. It is not uncommon that this publishing consists of an simple web interface connected to a database that holds records with limited information about the artifacts; information that is more relevant for managing the collection than for informing a wider public. It is not uncommon for a description to have no reference at all to that what is visible in the picture. Moreover this situation is hardly a worst-case scenario. In the Netherlands over 20 million artifacts in museums await a description, artifacts that do have a (scanty) description only half of them is available digitally. Four museums in the Netherlands (Naturalis, Museon, University Museum Utrecht, Dutch Institute of Image & Sound) together with three research and knowledge institutes (University of Applied Science Utrecht, Novay, BMC Group) decided in 2008 to explore the potential of user groups tagging collections and the effects of this on the involvement of these people towards the museum. For this purpose a dedicated social tagging tool was developed and implemented: www.ikweetwatditis.nl
DOCUMENT
Fast and successful searching for an object in a multimedia database is a highly desirable functionality. Several approaches to content based retrieval for multimedia databases can be found in the literature [9,10,12,14,17]. The approach we consider is feature extraction. A feature can be seen as a way to present simple information like the texture, color and spatial information of an image, or the pitch, frequency of a sound etc. In this paper we present a method for feature extraction on texture and spatial similarity, using fractal coding techniques. Our method is based upon the observation that the coefficients describing the fractal code of an image, contain very useful information about the structural content of the image. We apply simple statistics on information produced by fractal image coding. The statistics reveal features and require a small amount of storage. Several invariances are a consequence of the used methods: size, global contrast, orientation.
DOCUMENT
In this paper, fractal transforms are employed with the aim of image recognition. It is known that such transforms are highly sensitive to distortions like a small shift of an image. However, by using features based on statistics kept during the actual decomposition we can derive features from fractal transforms, which are invariant to perturbations like rotation, translation, folding or contrast scaling. Further, we introduce a feature invariance measure, which reveals the degree of invariance of a feature with respect to a database. The features and the way their invariance is measured, appear well suited for the application to images of textures.
DOCUMENT
In a real-world environment a face detector can be applied to extract multiple face images from multiple video streams without constraints on pose and illumination. The extracted face images will have varying image quality and resolution. Moreover, also the detected faces will not be precisely aligned. This paper presents a new approach to on-line face identification from multiple still images obtained under such unconstrained conditions. Our method learns a sparse representation of the most discriminative descriptors of the detected face images according to their classification accuracies. On-line face recognition is supported using a single descriptor of a face image as a query. We apply our method to our newly introduced BHG descriptor, the SIFT descriptor, and the LBP descriptor, which obtain limited robustness against illumination, pose and alignment errors. Our experimental results using a video face database of pairs of unconstrained low resolution video clips of ten subjects, show that our method achieves a recognition rate of 94% with a sparse representation containing 10% of all available data, at a false acceptance rate of 4%.
DOCUMENT
Background: Research in maternity care is often conducted in mixed low and high-risk or solely high-risk populations. This limits generalizability to the low-risk population of pregnant women receiving care from Dutch midwives. To address this limitation, 24 midwifery practices in the Netherlands bring together routinely collected data from medical records of pregnant women and their offspring in the VeCaS database. This database offers possibilities for research of physiological pregnancy and childbirth. This study explores if the pregnant women in VeCaS are a representative sample for the national population of women who receive primary midwife-led care in the Netherlands. Methods: In VeCaS we selected a low risk population in midwife-led care who gave birth in 2015. We compared population characteristics and birth outcomes in this study cohort with a similarly defined national cohort, using Chi Square and two side t-test statistics. Additionally, we describe some birth outcomes and lifestyle factors. Results: Midwifery practices contributing to VeCaS are spread over the Netherlands, although the western region is underrepresented. For population characteristics, the VeCaS cohort is similar to the national cohort in maternal age (mean 30.4 years) and parity (nulliparous women: 47.1% versus 45.9%). Less often, women in the VeCaS cohort have a non-Dutch background (15.7% vs 24.4%), a higher SES (9.9% vs 23.7%) and live in an urbanised surrounding (4.9% vs 24.8%). Birth outcomes were similar to the national cohort, most women gave birth at term (94.9% vs 94.5% between 37 + 0–41+ 6 weeks), started labour spontaneously (74.5% vs 75.5%) and had a spontaneous vaginal birth (77.4% vs 77.6%), 16.9% had a home birth. Furthermore, 61.1% had a normal pre-pregnancy BMI, and 81.0% did not smoke in pregnancy. Conclusions: The VeCaS database contains data of a population that is mostly comparable to the national population in primary midwife-led care in the Netherlands. Therefore, the VeCaS database is suitable for research in a healthy pregnant population and is valuable to improve knowledge of the physiological course of pregnancy and birth. Representativeness of maternal characteristics may be improved by including midwifery practices from the urbanised western region in the Netherlands.
DOCUMENT
