Filetype
Radiotherapy (RT) datasets can suffer from variations in annotation of organ at risk (OAR) and target structures. Annotation standards exist, but their description for prostate targets is limited. This restricts the use of such data for supervised machine learning purposes as it requires properly annotated data. The aim of this work was to develop a modality independent deep learning (DL) model fo
This paper proposes an iterative algorithm to reconstruct missing samples from non-stationary signals. The proposed algorithm is based on the well-known amplitude-modulation frequency-modulation model for non-stationary signals. The method initially estimates the instantaneous frequencies of the observed multi-component signal. The estimated IFs are then used to de-chirp the corresponding componen
We compared the risk of severe COVID-19 during two periods 2021 and 2022 when Delta and Omicron, respectively, were the dominating virus variants in Scania county, Sweden. We adjusted for differences in sex, age, comorbidities, prior infection and vaccination. Risk of severe disease from Omicron was markedly lower among vaccinated cases. It was also lower among the unvaccinated but remained high (
We study the task of embodied visual active learning, where an agent is set to explore a 3d environment with the goal to acquire visual scene understanding by actively selecting views for which to request annotation. While accurate on some benchmarks, today's deep visual recognition pipelines tend to not generalize well in certain real-world scenarios, or for unusual viewpoints. Robotic perception
Newly diagnosed breast cancer (BC) patients with clinical T1–T2 N0 disease undergo sentinel-lymph-node (SLN) biopsy, although most of them have a benign SLN. The pilot noninvasive lymph node staging (NILS) artificial neural network (ANN) model to predict nodal status was published in 2019, showing the potential to identify patients with a low risk of SLN metastasis. The aim of this study is to ass
Joint time-frequency representations are important tools when estimating the instantaneous frequency. The widely used spectrogram is known to have poor energy localisation, which the reassignment method improves. However, the reassignment method is sensitive to noise. In this paper we present a multitaper reassigned spectrogram (MTRS) that is robust to noise and tailored to short duration transien
In this paper we consider the epipolar geometry between orthographic and perspective cameras. We generalize many of the classical results for the perspective essential matrix to this setting and derive novel minimal solvers, not only for the calibrated case, but also for partially calibrated and non-central camera setups. While orthographic cameras might seem exotic, they occur naturally in many a
The underwater localization of a broadband target in the presence of strong interference and noise has been widely investigated. A novel clutter suppression approach based on oblique projections is proposed, exploiting the prior information of the expected target response. The method uses a generalised likelihood ratio formulation to select the oblique projection best matching the measured data. T
Nanowire growth enables creation of embedded heterostructures, where one material is completely surrounded by another. Through materials-selective post-growth oxidation it is also possible to combine amorphous oxides and crystalline, e.g. III-V materials. Such oxide-embedded structures pose a challenge for compositional characterization through transmission electron microscopy since the materials
A corpus study and a combined behavioural and neurophysiological study tested how phonetic and phonological features of the Danish creaky voice feature ‘stød’ influence predictive processing. Being associated with certain word endings, stød and its modal voice counterpart non-stød can cue upcoming speech. Stød has two phases. The first shows phonetic differences in pitch while the second, characte
Introduction: Rates of delivery by cesarean section have gradually risen in many parts of the world, and it is regarded as a safe surgical procedure with expanded indications. We assessed maternal complications within 6 weeks postpartum after planned cesarean section and after planned vaginal delivery among patients without medical indication for cesarean section. Material and methods: This was a
This paper introduces the first minimal solvers that jointly estimate lens distortion and affine rectification from repetitions of rigidly-transformed coplanar local features. The proposed solvers incorporate lens distortion into the camera model and extend accurate rectification to wide-angle images that contain nearly any type of coplanar repeated content. We demonstrate a principled approach to
Speaker localization using microphone arrays depends on accurate time delay estimation techniques. For decades, methods based on the generalized cross correlation with phase transform (GCC-PHAT) have been widely adopted for this purpose. Recently, the GCC-PHAT has also been used to provide input features to neural networks in order to remove the effects of noise and reverberation, but at the cost
While prenatal exposure to ambient air pollution has been shown to be associated with reduced birth weight, there is substantial heterogeneity across studies, and few epidemiological studies have utilized source-specific exposure data. The aim of the present study was, therefore, to investigate the associations between local, source-specific exposure to fine particulate matter (PM2.5) during pregn
Regulations restrict UAVs to fly only within direct view of the pilot, limiting their ability to support critical societal functions. One potential way to move beyond this limitation is by placing a 360-degree camera on the vehicle and using its feed to provide operators with a view that is the equivalent to being on the vehicle. This necessitates a cockpit user interface (UI) that amongst other t
Today, the advances in airborne LIDAR technology provide high-resolution datasets that allow specialists to detect archaeological features hidden under wooded areas more efficiently. Still, the complexity and large scale of these datasets require automated analysis. In this respect, artificial intelligence (AI)-based analysis has recently created an alternative approach for interpreting remote sen
The filtered backprojection (FBP) algorithm is viewed as a preferred candidate for general bistatic synthetic aperture radar (BiSAR) imaging since it does not pose any restrictions on SAR configurations or flight paths. However, high-efficient autofocus methods such as phase gradient autofocus (PGA) or Mapdrift (MD) cannot be effectively integrated with the FBP algorithm due to the unknown propert
Most 3d human pose estimation methods assume that input – be it images of a scene collected from one or several viewpoints, or from a video – is given. Consequently, they focus on estimates leveraging prior knowledge and measurement by fusing information spatially and/or temporally, whenever available. In this paper we address the problem of an active observer with freedom to move and explore the
Existing state-of-the-art estimation systems can detect 2d poses of multiple people in images quite reliably. In contrast, 3d pose estimation from a single image is ill-posed due to occlusion and depth ambiguities. Assuming access to multiple cameras, or given an active system able to position itself to observe the scene from multiple viewpoints, reconstructing 3d pose from 2d measurements becomes
