Swansea University

• Jointly models spatial, channel, and topological redundancies, moving beyond conventional spatial-only designs. • Embedding Mamba layers locally within specialized subcomponents instead of as a global backbone, enabling structured context modeling. • Achieves efficient long-range modeling with linear complexity, yielding a smaller model and faster decoding while outperforming baselines. Existing learned point cloud compression frameworks face two major limitations: (1) they focus almost exclusively on spatial redundancy and (2) rely on architectures built around local-global transformers or global Mamba blocks. Transformers incur quadratic complexity, while global Mamba lacks the granularity to capture structured correlations across multiple dimensions. We propose OctMamba, the first unified framework to jointly exploit spatial, channel, and topological redundancies, dimensions previously overlooked in point cloud geometry compression. Our approach introduces a new architectural principle: embedding Mamba modules within specialized subcomponents rather than applying them globally, challenging existing design paradigms. OctMamba combines two modules: Spatial-Channel Coupled Grouping Mamba (SCCGM) for spatial-channel fusion and Local Graph CNN-Mamba (LGCM) for topological encoding. This design enables efficient long-range modeling with linear complexity, delivering a smaller model and faster decoding while outperforming transformer-based and global Mamba baselines. On SemanticKITTI, OctMamba reduces bitrate by 60.2% over GPCC (D1 PSNR) and achieves state-of-the-art performance across LiDAR and dynamic human point cloud benchmarks with practical speed and scalability. By introducing multi-dimensional redundancy modeling, OctMamba has the potential to influence future research on efficient point cloud compression. Source code will be released.

The effectiveness of virtual reality (VR) simulation training for improving robotic surgical skills is not firmly established. This systematic review and meta-analysis aimed to evaluate the impact of VR simulator training on surgeons' robotic surgical performance in the operating room (OR). To synthesize evidence regarding the association between VR-based robotic surgery training and subsequent technical performance inside the OR. A comprehensive literature search was conducted in PubMed, Scopus, Embase, and CINAHL databases, adhering to PRISMA 2020 guidelines. Studies evaluating VR simulator training for robotic surgeons with subsequent performance assessment in the OR were included. Risk of bias assessment was performed using the Medical Education Research Study Quality Instrument and the modified Newcastle-Ottawa Scale for Education (NOS-E), along with funding source appraisal. A total of 294 records were screened, resulting in 13 studies (281 participants) included in the systematic review, and 4 studies (72 participants) suitable for meta-analysis. Metaanalysis was conducted using a random-effects model to pool correlation coefficients between preand post-training performance. Of the included studies, nine reported positive evidence, three found no evidence, and one found negative evidence regarding VR training's role. The meta-analysis of before-after studies revealed a significant positive correlation ( r = 0.717, P < 0.05) between simulation performance and intraoperative outcomes. The small number of studies included in the meta-analysis and methodological heterogeneity may limit generalisability of results. VR simulator training is associated with improved robotic surgical performance in the OR. Incorporation of VR simulation into robotic surgical curricula is likely beneficial for skill acquisition and operative readiness.

Six experiments with human participants explored the effects of a reinforcer devaluation procedure on the microstructure of free-operant schedule performance, which is theoretically and empirically novel. The goal was to test whether different aspects of the microstructure of free-operant responding would be differentially affected by the manipulation to test the view that bout-initiation responses can be regarded as stimulus-driven habits both susceptible to devaluation, and within-bout responses as goal-directed actions that are susceptible to such effects. Humans responded to a fictitious investment game in which responses made investments in two countries with different currencies. These currencies had an exchange rate into Great British Pound (GBP), which could be manipulated to devalue the currency (reinforcer). In Experiments 1a and 2, random ratio schedules showed overall devaluation effects, which were limited to within-bout rates and more pronounced on longer ratios. In Experiments 1b and 3, random interval schedules showed limited devaluation effects, overall, except for with shorter intervals. This effect was seen on some schedules for within-bout responding. Experiments 4 and 5 manipulated the response-reinforcer feedback function (Experiment 4) and participants' ability to experience contingency variations by pretraining to be variable or static in response rates (Experiment 5). Only when there was a stronger response-reinforcer relationship (Experiment 4), which could be experienced due to variable responding (Experiment 5), did the devaluation effect occur, and these effects were limited to within-bout responses. These data suggest that an account of free-operant responding based on there being, at least, two classes of response-bout initiations that are stimulus-driven habits, and within-bout responses that are goal-directed actions-may explain many schedule effects. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

Abstract Aims Free tissue transfer is a common plastic surgical procedure with a wide array of applications. There is limited consensus on how free flap ischaemia time impacts flap survival. This systematic review aims to evaluate the association between free flap ischaemia time and acute flap failure rate. Methods A systematic review and meta-analysis in compliance with PRISMA statement standards was conducted following registration on the PROSPERO Register (CRD420251052841). A search strategy was developed and the search conducted in September 2024. Studies eligible for inclusion reported length of ischaemia time for free flap surgery and acute flap failure rate. 78 articles were eligible for inclusion and were published between 1997 and 2024. Study populations ranged from 7 to 3926 patients, with a pooled total of 26 374 free flaps in 23 615 patients. The meta-analysis showed overall risk of flap failure was 3.5% (95% c.i. 3%–4.1%). Meta-regression using ischaemia time as a continuous variable showed an increasing risk of acute flap failure from 3% (2.3%–3.9%) at 30 min ischaemia time to 7.5% (4.4%–12.5%) at 240 min. The variables study type, flap type, indication and risk of bias did not account for between study heterogeneity which was substantial. Conclusions This study has systematically assessed existing literature to show a significant association between prolonged ischaemia time and acute free flap loss. This association can guide microsurgical training as carrying out microsurgery in a time efficient manner has a positive impact on free flap survival.