Search the digital library catalog

Abstract: Machine learning (ML) has demonstrated remarkable ability to uncover hidden patterns in data. However, the presence of biases and discrimination originating from the data itself and, consequently, emerging in the ML outcomes, remains a pressing concern. With the exponential growth of unstructured data, such as images, fairness has become increasingly critical, as neural network (NN) models may inadvertently learn and perpetuate societal and historical biases. To address this challenge, we propose a fairness-aware loss function that iteratively prioritizes the worst-performing sensitive group during NN training. This approach aims to balance treatment quality across sensitive groups, achieving fairer image classification outcomes while incurring only a slight compromise in overall performance. Our method, evaluated on the FairFace dataset, demonstrates significant improvements in fairness metrics while maintaining comparable overall quality. These trade-offs highlight that the minor decrease in overall quality is justified by the improvement in fairness of the models.
Keywords: fairness, search-basimage classification, machine learning, supervised learnign, neural networks
Published in DKUM: 23.04.2025; Views: 0; Downloads: 0
Full text (2,01 MB)

Abstract: Accurate automated localization of cephalometric landmarks in skull X-ray images is the basis for planning orthodontic treatments, predicting skull growth, or diagnosing face discrepancies. Such diagnoses require as many landmarks as possible to be detected on cephalograms. Today’s best methods are adapted to detect just 19 landmarks accurately in images varying not too much. This paper describes the development of the SCN-EXT convolutional neural network (CNN), which is designed to localize 72 landmarks in strongly varying images. The proposed method is based on the SpatialConfiguration-Net network, which is upgraded by adding replications of the simpler local appearance and spatial configuration components. The CNN capacity can be increased without increasing the number of free parameters simultaneously by such modification of an architecture. The successfulness of our approach was confirmed experimentally on two datasets. The SCN-EXT method was, with respect to its effectiveness, around 4% behind the state-of-the-art on the small ISBI database with 250 testing images and 19 cephalometric landmarks. On the other hand, our method surpassed the state-of-the-art on the demanding AUDAX database with 4695 highly variable testing images and 72 landmarks statistically significantly by around 3%. Increasing the CNN capacity as proposed is especially important for a small learning set and limited computer resources. Our algorithm is already utilized in orthodontic clinical practice.
Keywords: detection of cephalometric landmarks, skull X-ray images, convolutional neural networks, deep learning, SpatialConfiguration-Net architecture, AUDAX database
Published in DKUM: 27.03.2025; Views: 0; Downloads: 5
Full text (2,46 MB)
This document has many files! More...

Abstract: Calcium signals in pancreatic � cell collectives show a sharp transition from uncorrelated to correlated state resembling a phase transition as the slowly increasing glucose concentration crosses the tipping point. However, the exact nature or the order of this phase transition is not well understood. Using confocal microscopy to record the collective calcium activation of � cells in an intact islet under changing glucose concentration in an increasing and then decreasing way, we first show that in, addition to the sharp transition, the coordinated calcium response exhibits a hysteresis indicating a critical, first-order transition. A network model of � cells combining link selection and coordination mechanisms capture the observed hysteresis loop and the critical nature of the transition. Our results point towards an understanding of the role of islets as tipping elements in the pancreas that, interconnected by perfusion, diffusion, and innervation, cause the tipping dynamics and abrupt insulin release.
Keywords: cellular organization, physiology & dynamics, phase transitions in biological systems, complex networks, endocrine system, optical microscopy
Published in DKUM: 19.03.2025; Views: 0; Downloads: 2
Full text (1,90 MB)
This document has many files! More...

Abstract: In this paper, the Conditional Average Estimator artificial neural network (CAE ANN) was used to analyze the influence of chemical composition in conjunction with selected process parameters on the yield strength and elongation of an extruded 6082 aluminum alloy (AA6082) profile. Analysis focused on the optimization of mechanical properties as a function of casting temperature, casting speed, addition rate of alloy wire, ram speed, extrusion ratio, and number of extrusion strands on one side, and different contents of chemical elements, i.e., Si, Mn, Mg, and Fe, on the other side. The obtained results revealed very complex non-linear relationships between all of these parameters. Using the proposed approach, it was possible to identify the combinations of chemical composition and process parameters as well as their values for a simultaneous increase of yield strength and elongation of extruded profiles. These results are a contribution of the presented study in comparison with published research results of similar studies in this field. Application of the proposed approach, either in the research and/or in industrial aluminum production, suggests a further increase in the relevant mechanical properties.
Keywords: AA6082, hot extrusion, mechanical properties, yield strength, elongation, artificial neural networks, analysis
Published in DKUM: 12.03.2025; Views: 0; Downloads: 4
Full text (4,40 MB)
This document has many files! More...

Abstract: Bridges are typically designed by reputable, specialized engineering and design companies with years of experience. In these firms, experienced engineers share and pass on their knowledge to younger colleagues. However, when these experts retire, some of the knowledge is lost forever. As a subset of artificial intelligence methods, artificial neural networks (ANNs) can solve the problem of acquiring, transferring, and preserving specialized expert knowledge. This article describes the possible application of CAE ANN to acquire knowledge and to assist in the design of incrementally launched prestressed concrete bridges. Therefore, multidimensional graphs in the form of iso-curves of equal values were created, allowing practicing engineers to understand complex relationships between design parameters. The graphs also contain information about the reliability of the results, which is defined by an estimated parameter. The general rule is that results based on a larger number of actual data points are more reliable. Finally, an ANN BD assistant is proposed as an application that assists engineers and designers in the early stages of design and/or established engineers and designers in variant studies and design parameter optimization.
Keywords: artificial neural networks, bridge design, incremental launching method, expert knowledge, reliability of predictions, prestressed concrete bridges
Published in DKUM: 10.03.2025; Views: 0; Downloads: 12
Full text (5,54 MB)
This document has many files! More...