Iskanje po katalogu digitalne knjižnice

Opis: 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.
Ključne besede: detection of cephalometric landmarks, skull X-ray images, convolutional neural networks, deep learning, SpatialConfiguration-Net architecture, AUDAX database
Objavljeno v DKUM: 27.03.2025; Ogledov: 0; Prenosov: 5
Celotno besedilo (2,46 MB)
Gradivo ima več datotek! Več...

Opis: Because of the prevalence of depression, its often-chronic course, relapse and associated disability, early detection and non-intrusive monitoring is a crucial tool for timely diagnosis and treatment, remission of depression and prevention of relapse. In this way, its impact on quality of life and well-being can be limited. Current attempts to use artificial intelligence for the early classification of depression are mostly data-driven and thus non-transparent and lack effective means to deal with uncertainties. Therefore, in this paper, we propose an end-to-end framework for extracting observable depression cues from diary recordings. Furthermore, we also explore its feasibility for automatic detection of depression symptoms using observable behavioural cues. The proposed end-to-end framework for extracting depression was used to evaluate 28 video recordings from the Symptom Media dataset and 27 recordings from the DAIC-WOZ dataset. We compared the presence of the extracted features between recordings of individuals with and without a depressive disorder. We identified several cues consistent with previous studies in terms of their differentiation between individuals with and without depressive disorder across both datasets among language (i.e., use of negatively valanced words, use of first-person singular pronouns, some features of language complexity, explicit mentions of treatment for depression), speech (i.e., monotonous speech, voiced speech and pauses, speaking rate, low articulation rate), and facial cues (i.e., rotational energy of head movements). The nature/context of the discourse, the impact of other disorders and physical/psychological stress, and the quality and resolution of the recordings all play an important role in matching the digital features to the relevant background. In this way, the work presented in this paper provides a novel approach to extracting a wide range of cues relevant to the classification of depression and opens up new opportunities for further research.
Ključne besede: digital biomarkers of depression, facial cues, speech cues, language cues, deep learning, end-to-end pipeline, artificial intelligence
Objavljeno v DKUM: 17.01.2025; Ogledov: 0; Prenosov: 5
Celotno besedilo (2,34 MB)

Ključne besede: cities green mobility, bike sharing demand prediction, cable car demand prediction, machine learning, deep learning
Objavljeno v DKUM: 22.08.2024; Ogledov: 76; Prenosov: 10
Celotno besedilo (4,33 MB)

Opis: This paper presents the design of a high-voltage pulse-based radar and a supervised data processing method for soil moisture estimation. The goal of this research was to design a pulse-based radar to detect changes in soil moisture using a cross-hole approach. The pulse-based radar with three transmitting antennas was placed into a 12 m deep hole, and a receiver with three receive antennas was placed into a different hole separated by 100 m from the transmitter. The pulse generator was based on a Marx generator with an LC filter, and for the receiver, the high-frequency data acquisition card was used, which can acquire signals using 3 Gigabytes per second. Used borehole antennas were designed to operate in the wide frequency band to ensure signal propagation through the soil. A deep regression convolutional network is proposed in this paper to estimate volumetric soil moisture using time-sampled signals. A regression convolutional network is extended to three dimensions to model changes in wave propagation between the transmitted and received signals. The training dataset was acquired during the period of 73 days of acquisition between two boreholes separated by 100 m. The soil moisture measurements were acquired at three points 25 m apart to provide ground truth data. Additionally, water was poured into several specially prepared boreholes between transmitter and receiver antennas to acquire additional dataset for training, validation, and testing of convolutional neural networks. Experimental results showed that the proposed system is able to detect changes in the volumetric soil moisture using Tx and Rx antennas.
Ključne besede: ground penetrating radar, cross-hole, L-band, deep learning, convolutional neural network, soil moisture estimation
Objavljeno v DKUM: 03.04.2024; Ogledov: 448; Prenosov: 26
Celotno besedilo (3,22 MB)
Gradivo ima več datotek! Več...

Opis: With the rise of deep learning technology, the field of medical image segmentation has undergone rapid development. In recent years, convolutional neural networks (CNNs) have brought many achievements and become the consensus in medical image segmentation tasks. Although many neural networks based on U-shaped structures and methods, such as skip connections have achieved excellent results in medical image segmentation tasks, the properties of convolutional operations limit their ability to effectively learn local and global features. To address this problem, the Transformer from the field of natural language processing (NLP) was introduced to the image segmentation field. Various Transformer-based networks have shown significant performance advantages over mainstream neural networks in different visual tasks, demonstrating the huge potential of Transformers in the field of image segmentation. However, Transformers were originally designed for NLP and ignore the multidimensional nature of images. In the process of operation, they may destroy the 2D structure of the image and cannot effectively capture low-level features. Therefore, we propose a new multi-scale cross-attention method called M-VAN Unet, which is designed based on the Visual Attention Network (VAN) and can effectively learn local and global features. We propose two attention mechanisms, namely MSC-Attention and LKA-Cross-Attention, for capturing low-level features and promoting global information interaction. MSC-Attention is designed for multi-scale channel attention, while LKA-Cross-Attention is a cross-attention mechanism based on the large kernel attention (LKA). Extensive experiments show that our method outperforms current mainstream methods in evaluation metrics such as Dice coefficient and Hausdorff 95 coefficient.
Ključne besede: CNNs, deep learning, medical image processing, NLP, semantic segmentation
Objavljeno v DKUM: 14.03.2024; Ogledov: 496; Prenosov: 311
Celotno besedilo (1,46 MB)
Gradivo ima več datotek! Več...