In this paper, we investigate the use of Vision Transformers for processing and understanding visual data in an autonomous driving setting. Specifically, we explore the use of Vision Transformers for semantic segmentation and monocular depth estimation using only a single image as input. We present state-of-the-art Vision Transformers for these tasks and combine them into a multitask model. Through multiple experiments on four different street image datasets, we demonstrate that the multitask approach significantly reduces inference time while maintaining high accuracy for both tasks. Additionally, we show that changing the size of the Transformer-based backbone can be used as a trade-off between inference speed and accuracy. Furthermore, we investigate the use of synthetic data for pre-training and show that it effectively increases the accuracy of the model when real-world data is limited.

Perioperative monitoring of cardiac function is beneficial for early detection of cardiovascular complications. The standard of care for cardiac monitoring performed by trained cardiologists and anesthesiologists involves a manual and qualitative evaluation of ultrasound imaging, which is a time-demanding and resource-intensive process with intraobserver- and interobserver variability. In practice, such measures can only be performed a limited number of times during the intervention. To overcome these difficulties, this study presents a robust method for automatic and quantitative monitoring of cardiac function based on 3D transesophageal echocardiography (TEE) B-mode ultrasound recordings of the left ventricle (LV). Such an assessment obtains consistent measurements and can produce a near real-time evaluation of ultrasound imagery. Hence, the presented method is time-saving and results in increased accessibility. The mitral annular plane systolic excursion (MAPSE), characterizing global LV function, is estimated by landmark detection and cardiac view classification of two-dimensional images extracted along the long-axis of the ultrasound volume. MAPSE estimation directly from 3D TEE recordings is beneficial since it removes the need for manual acquisition of cardiac views, hence decreasing the need for interference by physicians. Two convolutional neural networks (CNNs) were trained and tested on acquired ultrasound data of 107 patients, and MAPSE estimates were compared to clinically obtained references in a blinded study including 31 patients. The proposed method for automatic MAPSE estimation had low bias and low variability in comparison to clinical reference measures. The method accomplished a mean difference for MAPSE estimates of (-0.16±1.06) mm. Thus, the results did not show significant systematic errors. The obtained bias and variance of the method were comparable to inter-observer variability of clinically obtained MAPSE measures on 2D TTE echocardiography. The novel pipeline proposed in this study has the potential to enhance cardiac monitoring in perioperative- and intensive care settings.

With reliable performance, and linear time complexity, Vision Transformers like the Swin Transformer are gaining popularity in the field of Medical Image Computing (MIC). Examples of effective volumetric segmentation models for brain tumours include VT-UNet, which combines conventional UNets with Swin Transformers using a unique encoder-decoder Cross-Attention (CA) paradigm. Self-Supervised Learning (SSL) has also experienced an increase in adoption in computer vision domains such as MIC, in situations where labelled training data is scarce. The Querying Transformer UNet (QT-UNet) model we introduce in this paper brings these advancements together. It is an all-Swin Transformer UNet with an encoder-decoder CA mechanism strengthened by SSL. For the purpose of evaluating the potential of QT-UNet as a generic volumetric segmentation model, it is subjected to extensive testing on several MIC datasets. Our best model achieves a Dice score of 88.61 on average and a Hausdorff Distance of 4.85mm making it competitive with State of the Art in Brain Tumour Segmentation (BraTS) 2021, using 40% fewer FLOPs than the baseline VT-UNet. We found poor results with Beyond The Cranial Vault (BTCV) and Medical Segmentation Decathlon (MSD), but validate the effectiveness of our new CA mechanism and find that the SSL pipeline is most effective when pre-trained with our CT-SSL dataset. The code be can found at https://github.com/AndreasHaaversen/QT-UNet .

New digital technologies furnish retail managers with new means to enhance consumer experiences in omnichannel retailing. Conceptual academic literature and industry emphasize the promising use of immersive digital displays and their potential benefits for retailers. In this research, we present the design of a personal shopping assistance system that is based on optical see-through mixed-reality technology. Microsoft HoloLens 2 was leveraged as the archetype to realize this novel system, facilitating consumer information search and decision making. The design incorporates various shopping assistance elements (i.e., product information, reviews, recommendations, product availability, videos, a virtual cart, and an option to buy). Users can interact with these elements with gesture-based inputs to navigate through the interface. A qualitative study with 35 participants was conducted to collect users’ feedback and perceptions about the mixed-reality shopping assistant system. Derived from the qualitative feedback, we propose seven design principles that aim to support future designs and developments of mixed-reality shopping applications for head-mounted displays in omnichannel retail: rigor, informativeness, tangibility, summary, comparability, flexibility and holism.

Aims: Continuous monitoring of left ventricular (LV) function may improve cardiopulmonary management. Therefore, we have developed 3D autoMAPSE, a novel method that combines 3D transesophageal echocardiography and deep learning to automatically measure mitral annular plane systolic excursion (MAPSE). We hypothesized that 3D autoMAPSE could provide continuous monitoring of LV function in perioperative patients. Methods and results: This prospective observational study included 50 adult intensive care patients after cardiac surgery. Single-beat full-volume 3D recordings were obtained every 5 min over a 120-min period using a hands-free method that simulated continuous monitoring with transesophageal echocardiography. Precision and agreement with manual MAPSE were determined by a test-retest study design during hemodynamic stability. Our results show that continuous monitoring by 3D autoMAPSE had excellent feasibility (99%). It analysed 10 cycles instantaneously (55 ± 15 s) with high precision (least significant change 1.6 mm). 3D autoMAPSE had adequate agreement with manual MAPSE (bias –1.4 mm, limits of agreement −4.0 to 1.2 mm). Continuous 3D autoMAPSE was associated with both N-terminal pro B-type natriuretic peptide (rho = −0.37, P = 0.008) and high-sensitivity troponin-T (rho = −0.28, P = 0.047). This association was slightly stronger than for LV ejection fraction or any other single echocardiographic measurement. Conclusion: 3D autoMAPSE provided continuous monitoring of LV function in perioperative patients by obtaining highly feasible and precise measurements that agree with manual echocardiography and postoperative biomarkers. Thus, continuous 3D autoMAPSE may be an attractive complement to hemodynamic monitoring that can aid cardiopulmonary management.