Magnetic Resonance Imaging (MRI) and its automatic segmentation are pivotal in assisting physicians with clinical diagnosis. In recent years, with the scarcity of labeled data, significant advancements have been made in semi-supervised segmentation. However, the prediction of many current methods is affected by the presence of false positive regions, which limits their reliability in clinical applications. To tackle this issue, we propose a pseudo-label optimization method based on polar coordinate modeling and prior constraints (PMPC), which refines false positive regions in pseudo-labels by leveraging prior knowledge within the polar coordinate system. Firstly, to improve the efficiency and rationality during polar coordinate modeling, the Adaptive Pole Selection (APS) algorithm is presented to ensure that the pole is located within the foreground region. Secondly, to mitigate false positive regions in pseudo-labels that violate medical anatomical priors, we propose the Prior Knowledge Constraint in Polar Coordinate System (KCP) module to reassign pixel categories in these regions. Finally, the Shape-aware Weighting (SaW) strategy is presented to evaluate the quality of the optimized pseudo-labels based on their shape and then determine their weight in guiding network parameter updates. Experiments on three MRI datasets demonstrate that the proposed method can be effectively integrated with existing pelvic MRI segmentation approaches, significantly reducing false positive rates and further improving segmentation quality.

Our everyday technologies could have appeared terrifying to our ancestors: instantaneous disembodied communication, access to knowledge, objects with ‘intelligence’ that talk to us (and each other). Black boxes and intangible entities are omnipresent in our homes and lives without our necessarily understanding the hidden flows of data, unknown agendas, imaginary clouds, and mysterious rules that govern them. Have humanity's ways of relating to the unknown throughout history gone away, or have they perhaps transmuted into new forms? In an ongoing project, we have inventoried examples, encounters and reflections on contemporary technology, framed through the perspective of the haunted, spectral and otherworldly. In this paper, we excerpt this collection to illustrate the value and opportunity of an unfamiliar, disquieting perspective in helping to frame the frictions, beliefs and myths that are emerging around interactions with everyday technologies. We posit and demonstrate ‘spooky technology’ as an accessible framework to reflect and respond to our increasingly entangled relationships with technology.

Spectrality describes the spatiotemporal relationships between landscape and memories where the past ‘haunt’ the present. We explore how spatialized soundscapes may evoke the spectral memories of physical landscape. To do this, we curated a site-specific mixed reality audio walk at Carrie Furnace, a national historic landmark in Pittsburgh’s ‘Rivers of Steel’ National Heritage area. This audio walk leverages virtual soundscapes, mobile mixed reality, and open-ended narratives. Each visitor takes a unique path through the spectral memories of the site’s industrial past and post-industrial relics. Findings from an eleven-person user study assess our approach to spatialized spectral soundscapes and reveal the dimensions (content, context, experience) that resonate with audiences.