Search

The aerodynamics of floating turbines is complicated by large motions which are permitted by the floating foundation, and the interaction between turbine, wind, and wake is not yet fully understood. The object of this paper is a wind tunnel campaign finalized at characterizing the aerodynamic response of a 1 V 100 scale model of the IEA 15MW subjected to imposed platform motion. The turbine aerodynamic response is studied focusing on thrust force, torque, and wake at 2.3D downwind the rotor. Harmonic motion is imposed in the surge, sway, roll, pitch, and yaw directions with several frequencies and amplitudes, which are selected to be representative of the two 15MW floating turbines developed within the COREWIND project. Thrust and torque show large-amplitude oscillations with surge and pitch motion, the main effect of which is an apparent wind speed; oscillations in thrust and torque are negligible with the other motions, the main effect of which is to alter the wind direction. The thrust and torque response measured in the experiment is compared with predictions of a quasi-steady model, often used for control-related tasks. The agreement is good in the case of low-frequency surge motion, but some differences are seen in the pitch case. The quasi-steady model is not predictive for the response to wave-frequency motion, where blade unsteadiness may take place. Wake was measured imposing motion in five directions with frequency equal to the wave frequency. The axial speed is slightly lower with motion compared to the fixed case. The turbulence kinetic energy is slightly lower too. Wave-frequency motion seems to produce a more stable and lower flow mixing. [ABSTRACT FROM AUTHOR]

Background: CT is commonly used to image patients with ischaemic stroke but radiologist interpretation may be delayed. Machine learning techniques can provide rapid automated CT assessment but are usually developed from annotated images which necessarily limits the size and representation of development data sets. We aimed to develop a deep learning (DL) method using CT brain scans that were labelled but not annotated for the presence of ischaemic lesions., Methods: We designed a convolutional neural network-based DL algorithm to detect ischaemic lesions on CT. Our algorithm was trained using routinely acquired CT brain scans collected for a large multicentre international trial. These scans had previously been labelled by experts for acute and chronic appearances. We explored the impact of ischaemic lesion features, background brain appearances and timing of CT (baseline or 24-48 hour follow-up) on DL performance., Results: From 5772 CT scans of 2347 patients (median age 82), 54% had visible ischaemic lesions according to experts. Our DL method achieved 72% accuracy in detecting ischaemic lesions. Detection was better for larger (80% accuracy) or multiple (87% accuracy for two, 100% for three or more) lesions and with follow-up scans (76% accuracy vs 67% at baseline). Chronic brain conditions reduced accuracy, particularly non-stroke lesions and old stroke lesions (32% and 31% error rates, respectively)., Conclusion: DL methods can be designed for ischaemic lesion detection on CT using the vast quantities of routinely collected brain scans without the need for lesion annotation. Ultimately, this should lead to more robust and widely applicable methods., Competing Interests: Competing interests: JW declares academic grants but no industry or other CoI. GM declares consultancy fees from Canon Medical. All the other authors declare no CoI., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ.)

Segmentation masks of pathological areas are useful in many medical applications, such as brain tumour and stroke management. Moreover, healthy counterfactuals of diseased images can be used to enhance radiologists' training files and to improve the interpretability of segmentation models. In this work, we present a weakly supervised method to generate a healthy version of a diseased image and then use it to obtain a pixel-wise anomaly map. To do so, we start by considering a saliency map that approximately covers the pathological areas, obtained with ACAT. Then, we propose a technique that allows to perform targeted modifications to these regions, while preserving the rest of the image. In particular, we employ a diffusion model trained on healthy samples and combine Denoising Diffusion Probabilistic Model (DDPM) and Denoising Diffusion Implicit Model (DDIM) at each step of the sampling process. DDPM is used to modify the areas affected by a lesion within the saliency map, while DDIM guarantees reconstruction of the normal anatomy outside of it. The two parts are also fused at each timestep, to guarantee the generation of a sample with a coherent appearance and a seamless transition between edited and unedited parts. We verify that when our method is applied to healthy samples, the input images are reconstructed without significant modifications. We compare our approach with alternative weakly supervised methods on the task of brain lesion segmentation, achieving the highest mean Dice and IoU scores among the models considered.