Microsoft Windows remains the dominant desktop operating system and therefore a frequent focus of digital forensic and incident response investigations. Windows Registry analysis is particularly valuable because it captures persistence mechanisms, execution traces, user activity, device usage, and system configuration changes that are often central to incident reconstruction. Nevertheless, modern investigations are challenged by the scale of Registry data, the fragmentation of evidence across hives and complementary sources, and the need to prioritise investigative actions under time pressure. This paper presents WinRegRL, a hybrid framework that combines a Markov Decision Process (MDP) solved by dynamic programming with bounded Reinforcement Learning (RL) refinement and Rule-based Artificial Intelligence (RB-AI) for automated Windows Registry and timeline-centred forensic analysis. Methodologically, the core planner is a finite-state dynamic-programming solver over an expert-specified model; reinforcement learning enters only as bounded, local tabular refinement for low-support state-action regions, so the framework is positioned as an MDP/dynamic-programming approach with bounded RL rather than as an end-to-end learned agent. The framework models the investigation process as a Markov Decision Process (MDP) with explicitly defined states, actions, transition dynamics, and reward design, and incorporates expert-derived policy graphs to initialise and refine the search strategy. We evaluate the framework on four heterogeneous forensic datasets spanning multiple Windows versions and incident scenarios, and we compare it against analyst-assisted baselines and controlled examiner-led workflows. Under the evaluation protocol adopted in this study, WinRegRL reduced investigation time by up to 68%, increased the number of adjudicated relevant artefacts identified by up to 35%, and achieved high artefact-level precision on the evaluated datasets. Rather than claiming universal superiority, we show that the proposed framework provides a reproducible and explainable decision-support mechanism that improves investigation efficiency while maintaining strong evidential coverage in the tested scenarios. These findings position WinRegRL as a promising decision-support framework for large-scale and time-critical Windows incident response.
Wireless capsule endoscopy (WCE) enables painless, minimally invasive visualization of the gastrointestinal tract. Still, its diagnostic potential is limited by incomplete mucosal coverage and poor transferability of existing navigation methods across patient anatomies. We propose a transferable, anatomical landmark-guided deep reinforcement learning framework for robust autonomous gastric navigation. Leveraging a lightweight edge-contour-depth fusion module, our policy operates on stable, low-dimensional landmark coordinates rather than high-dimensional video streams. This design effectively bridges the sim-to-real visual gap and ensures robustness across diverse anatomies, enabling low-cost deployment by reducing computational overhead. In simulations across eight patient-derived models, the method achieves >97% coverage within 50 s, significantly outperforming vanilla Proximal Policy Optimization, Soft Actor-Critic, and Deep Q-Network agents by enhancing coverage and minimizing variance. To ensure deployment reliability, a two-stage sim-to-real pipeline supported by an adaptive dynamic programming controller actively mitigates physical disturbances, including actuator latency and peristalsis. Ex vivo experiments across five independent scans demonstrate high coverage stability, achieving a mean coverage of 87% and a 53% reduction in procedure time compared with expert manual control. This study establishes a scalable paradigm for autonomous, high‑coverage endoscopic navigation, advancing the clinical deployment of intelligent WCE systems for GI diagnostics.
Community Reinforcement and Family Training (CRAFT) is an evidence-based approach that promotes family skills to encourage people with substance use disorders to seek treatment. To enhance scalable CRAFT implementation, we assessed three digital counselor training models for fidelity, feasibility/acceptability, and implementation potential. Participants (47 counselors) were randomized to one of three digital training programs. Tutorial (T): two-week CRAFT modules. Tutorial + Self Study Materials (TM): two-week tutorial plus 13 modules of training materials (e.g., session videos/checklists) released over 10 weeks after the tutorial. Tutorial + Self Study Materials + Coaching (TMC): two-week tutorial, 10 weeks of additional materials, plus feedback and coaching on ≥one recorded CRAFT sessions. Participants completed assessments at baseline, two weeks, and 12-weeks. The primary outcome was CRAFT fidelity. Secondary outcomes included CRAFT knowledge, feasibility, acceptability, and implementation potential. TMC participants demonstrated the highest CRAFT fidelity and knowledge with 83% of counselors achieving proficiency (T:0%; TM:67%). Feasibility was highest for completing the tutorial (73-100% of participants), then the CRAFT session recording and coaching (79%), and the self-study materials (56-75%). 93% of participants were satisfied with their training program. Implementation potential scores were high across all training groups. The TMC training yielded the best balance of CRAFT fidelity, feasibility/acceptability and implementation potential. TM had the next highest fidelity, and its significantly lower cost/effort may aid implementation in resource-limited settings. T alone was not sufficient to facilitate CRAFT fidelity. Digital training has promise for expanding CRAFT implementation and potentially addressing clinical workforce gaps. gov:NCT05875142.
Protein based interfacial films are a vital component of encapsulated functional food. The stability and mechanical performance of bovine serum albumin (BSA) films can be significantly enhanced through two complementary strategies: (i) electrostatic complexation for example with carboxymethyl cellulose (CMC), and (ii) interfacial covalent cross linking of BSA by genipin (GNP). Quartz crystal microbalance (QCM), interfacial small amplitude oscillatory shear rheology (iSAOS), and atomic force microscopy (AFM) were combined to quantify mass adsorption, viscoelastic film development, and interfacial morphology of BSA, both alone and in mixtures with CMC and/or GNP at pH 3.0 and 7.0. Attractive electrostatic interactions between BSA and CMC led to interfaces with the highest viscoelastic shear moduli (G' = 0.069 ± 0.004 N/m for BSA + CMC at pH 3.0). At neutral pH genepin enhanced the properties of a BSA film from predominantly liquid like to predominantly solid-like. These findings demonstrate that electrostatic complexation and interfacial cross-linking provide two complementary pathways for the reinforcement of protein-based interfacial films and reveal new routes for enhancing mechanical stability of encapsulated ingredients.
Traditional separation of inventory management and Prognostics and Health Management (PHM) often leads to resource misallocation. While Deep Reinforcement Learning (DRL) offers a promising solution for joint decision-making, standard agents typically treat Prognostics and Health Management predictions as deterministic ground truths. However, in real-world scenarios, remaining useful life (RUL) predictions inherently contain stochastic errors. Ignoring this uncertainty leads to risk-blind policies that fail to buffer against sudden failures when prediction confidence is low. To address this, this paper proposes an Uncertainty-Aware collaborative adaptive inventory strategy. First, we introduce a Bayesian uncertainty quantification mechanism using Monte Carlo Dropout to estimate not only the RUL value but also its prediction variance. Second, to overcome the agent's myopic behavior, a novel Asymmetric Cost-Aware Reward Shaping mechanism is designed. By strategically decoupling the training and evaluation reward functions-specifically by introducing safety stock penalties and attenuating holding costs during training-the agent is guided to establish robust inventory buffers against supply chain uncertainties. Simulation results demonstrate that the proposed Risk-Sensitive PPO strategy significantly outperforms deterministic baselines, reducing total costs by 40.3% under high-noise environments.
The proliferation of distributed energy resources (DERs) and the ubiquity of Internet of Things (IoT) devices are driving the integration of mobile edge computing (MEC) into smart grids. This convergence enables real-time data processing for prosumers but introduces a complex cyber-physical coupling: computational offloading decisions directly impact local energy consumption, thereby altering the prosumer's status in the peer-to-peer (P2P) energy market. Conversely, dynamic market prices influence the economic viability of offloading. This paper addresses the joint optimization of computational task offloading and P2P energy trading in an edge-assisted smart grid ecosystem. We formulate the problem as a mixed-integer nonlinear programming (MINLP) model aimed at maximizing long-term system utility, balancing throughput, latency, and economic incentives under strict edge server capacity and community energy neutrality constraints. To tackle the curse of dimensionality and system stochasticity, we propose a hybrid framework combining Deep Q-Networks (DQN) with a constraint-aware heuristic mechanism. The DQN agent learns adaptive offloading policies from high-dimensional states, while a deterministic rule-based layer ensures strict adherence to community energy balance. Simulation results based on real-world solar generation and market data demonstrate that our proposed method outperforms baseline strategies-including local-only execution and greedy heuristics-improving average utility by 12.3% and reducing task delay by 16.5%, while maintaining robust operational feasibility.
The accurate ability to predict the distribution of contact stress under reinforced concrete (RC) footings is important for the safety and serviceability of shallow foundations. Conventional analytical models idealizing the footing as rigid and soil as homogenous fail to capture the stress concentration and redistribution effects, especially under non-uniform loading. Earlier studies are mostly concentrated on sand; however, basalt soil has different mechanical characteristics as it possesses high stiffness, angularity, and interlocking effects. It also ignores stiffness loss due to concrete cracking. This study aims to fill these gaps through an experimental and numerical investigation of RC square footings resting on basaltic soil and the influence of the reinforcement ratio, yield strength of steel and strength of concrete. Within the laboratory conditions, four footings having different reinforcement ratios of 0.19%, 0.36%, 0.54% and 3.43% were tested under monotonic loading. Central and edge displacements were measured. Using a validated finite element model, a parametric study expanded the investigation to include reinforcement ratios of 0.54% to 4.80%, steel yield stresses of 240 MPa to 450 MPa and concrete compressive strengths of 20 MPa to 60 MPa, allowing systematic consideration of these parameters on central contact stress, ultimate load, deformation and energy absorption. The findings revealed that enhancing the reinforcement ratio from 0.54% to 4.80% resulted in an increase of 73.4% in central contact stress, 34.1% in ultimate load, and 55% in energy absorption, respectively. Increase in steel yield stress from 240 MPa to 450 MPa caused a 25.2% increase in central contact stress, 13% in ultimate load and 3.74% in energy absorption in laminated composite panel. The increase of concrete compressive strength from 20 MPa to 60 MPa increased central contact stress by 117.4% ultimate load by 70.4% and energy absorption by 270% showing this factor as dominant. These results show that the performance of footing on stiff basaltic soil mainly depends on the concrete strength and amount of reinforcement whereas careful use of steel yield stress. The insights provided by the study are critical for practical design. Furthermore, non-uniform contact stresses, stiffness degradation, and soil-structure interaction need to be accounted for optimizing strength and ductility.
This study investigates the use of active infrared thermography combined with microwave excitation for identifying and assessing reinforcement bars in concrete structures. The proposed approach integrates numerical modeling and experimental validation to ensure accurate and reliable results. The first phase involves developing a detailed numerical model comprising a microwave source, a broadband antenna, and a concrete structure with specified reinforcement bar arrangements. This model simulates the interaction between microwave excitation and embedded reinforcement, analyzing temperature distributions and thermal responses on the concrete surface. The goal is to optimize the measurement methodology by evaluating parameters such as excitation mode and antenna vs. sample configuration. The second phase focuses on experimental validation of the numerical findings. An experimental setup replicates the modeled conditions to compare real-world thermal patterns with simulated predictions, ensuring consistency and reliability. By combining numerical simulations with experimental testing, this study aims to establish a robust framework for using active infrared thermography with microwave excitation in non-destructive evaluation of reinforced concrete structures. The approach seeks to provide a precise, efficient method for assessing the condition and layout of reinforcement bars in concrete, contributing to advancements in structural inspection techniques.
The impact of a narrow true lumen (NTL) on the outcomes of fenestrated-branched endovascular repair in patients with postdissection thoracoabdominal aortic aneurysms (PD-TAAAs) is underreported. Data from an international, multicenter registry were analyzed, to identify patients treated for PD-TAAAs (2015-2025) at 23 centers. All patients underwent fenestrated-branched endovascular repair using custom or off-the-shelf endografts. NTL was defined by a true lumen diameter <25 mm identified at any aortic level on preoperative computed tomography angiogram. Short-term endpoints compared between NTL and no-NTL patients included technical success, procedural metrics, 30-day mortality, and major adverse events (MAEs). Midterm endpoints included 5-year freedom from aortic adverse events (related mortality, rupture, reintervention, endograft instability) and freedom from target artery instability. Among 544 patients (1705 target vessels), 438 (80%) had an NTL. Device design did not differ between groups (52% branches, 30% fenestrated, and 18% fenestrated-branched combination; P = .053). Patients with an NTL more frequently received bridging stent reinforcement (P < .001), and renal inner branches (P = .038). Septotomy or false lumen occlusion were more often performed in NTLs (27% vs 11%; P = .006). Patients with NTLs had longer operating time (P = .031), fluoroscopy time (P = .007), and a higher dose area product (P = .046). Technical success was 95% in both groups (P = .750). Overall 30-day mortality was 4%, and MAEs occurred in 35%. NTLs did not have a significant impact on MAEs (adjusted odds ratio, 0.84; 95% confidence interval [CI], 0.28-2.76; P = .766). Freedom from any aortic adverse event at 5 years was lower in patents with NTL (73% vs 91%; P = .027), driven primarily to secondary procedures of false lumen embolization (P = .027). Freedom from target vessel instability was 86% ± 4% in the NTL group and 92% ± 4% in the no-NTL group (P = .072). Patients with NTLs had a similar primary patency (97% ± 2% vs 98% ± 2%; P = .380) but lower freedom from target vessel endoleak (89% ± 4% vs 97% ± 3%; P = .006). After adjustment, NTL diameter <10 mm (hazard ratio [HR], 2.45; 95% CI, 1.37-4.36; P = .002) was significantly associated with target artery instability. Use of inner branches (HR, 0.11; 95% CI, 0.02-0.87; P = .035) and bridging stent reinforcement (HR, 0.54; 95% CI, 0.31-0.96; P = .038) were protective. NTL is the most common anatomic presentation in PD-TAAAs and is associated with more complex procedures, but does not affect technical success, mortality, or MAEs. Patients with am NTL experience a higher rate or reinterventions, primarily false lumen embolization. NTL <10 mm is a risk factor for target vessel instability, and reinforcement of bridging stents may be beneficial in these cases.
Conventional polymethyl methacrylate denture base resin exhibits inadequate surface hardness and increased surface roughness, affecting durability and promoting microbial colonization. Therefore, it is of interest to evaluate the effect of different glass fiber forms on surface hardness and roughness of heat-cured polymethyl methacrylate. Hence, eighty specimens were divided into four groups (n=20): control, woven fiber, chopped fiber and longitudinal fiber reinforcement (2.5 wt %) and tested using Vickers microhardness and profilometry. Longitudinal fiber reinforcement showed the highest surface hardness (20.99 ± 0.28 VHN), while woven fibers produced significantly greater surface roughness compared to control (p < 0.001). Thus, we show that glass fiber reinforcement improves hardness but increases roughness, with longitudinal fibers providing the most favorable balance for clinical application.
Glass fiber-reinforced polymer composites (GFRP) are widely used in engineering applications thanks to their high performance. This research investigates how various hybrid reinforcement techniques influence the flexural and shear performance of GFRP laminates. The composite materials were produced by integrating nonwoven glass tissues with woven glass fabrics, along with single and hybrid nanoscale fillers. Silica (SiO₂) and carbon (C) nanoparticles at different weight fractions were selected as nano-reinforcements for their cost-effectiveness. Seven composite variants were made by hand lay-up, including an unmodified GFRP reference and six hybrid configurations. Sonication and magnetic stirring methods were used to disperse nanoparticles in the epoxy matrix uniformly. Flexural performance was assessed by the three-point flexural test, while shear performance was assessed using the short beam shear and Iosipescu tests to measure both in-plane and interlaminar shear strength. When compared to the unmodified GFRP reference, samples with 0.5 wt% carbon black showed better shear and flexural performance. Hybrid reinforcement using 0.25 wt% carbon black combined with 0.25 wt% silica achieved the highest flexural strain increase (21%) and an 8% gain in flexural strength. This indicates improved ductility without compromising shear performance. However, higher concentrations of hybrid fillers negatively affect mechanical behavior.
EUS-guided gastrojejunostomy (EUS-GJ) is an effective minimally invasive treatment for gastric outlet obstruction. However, misdeployment can result in complex fistulas with significant morbidity. We describe successful endoscopic management of a gastrocolojejunal (GCJ) fistula following EUS-GJ. A 69-year-old male developed a GCJ fistula post-EUS-GJ, presenting with septic shock. Initial bridging was performed using 2 telescoping lumen-apposing metal stents and a coaxial fully covered metal stent. Definitive closure was attempted endoscopically using a double-channel gastroscope with OverStitch (Boston Scientific, Marlborough, Mass, USA) and over-the-scope clips (OTSCs). The colojejunal and gastrocolic fistulas were successfully closed with running 2.0 polypropylene sutures and OTSC reinforcement. Fluoroscopy confirmed a complete seal, and follow-up endoscopy showed intact closure without leaks or adverse events. Endoscopic suturing with clip reinforcement is a safe and effective alternative to surgery for complex post-EUS-GJ fistulas in high-risk patients.
Adaptive behavior requires flexibly shifting between exploiting familiar rewards and exploring novel opportunities. These explore-exploit decisions are implemented via a distributed brain network-anchored in frontopolar cortex (FPC) and ventromedial prefrontal cortex (vmPFC)-that computes the total value of a given choice by weighting the immediate value of familiar options against the latent future value of exploration. Capturing the precise temporal dynamics of these neural computations across spatially distributed cortical networks is critical to resolving how the brain manages this tradeoff. Here, we combined magnetoencephalography (MEG) with partially observable Markov decision process (POMDP) modeling during a reinforcement learning task. By decoding POMDP-derived choice policies in MEG source space, we mapped the precise spatiotemporal emergence of explore-exploit decisions across the cortex. We demonstrate that explore-exploit policy implementation unfolds via a hierarchical functional dissociation across the rostral prefrontal cortex. During decision formation, the lateral FPC initiates the strategic shift toward exploration hundreds of milliseconds before choice execution. Conversely, vmPFC and orbitofrontal cortex (OFC) do not exhibit early significant divergence from the baseline exploitative trace on exploration trials, displaying only a delayed, transient peak prior to choice execution. Following feedback, the vmPFC and OFC transition to a sustained representation of the choice's empirical value to update reward expectancies and guide future actions. Mapping reinforcement learning variables onto millisecond-resolved neural data reveals how the human brain resolves the explore-exploit dilemma: through early strategic initiation of exploration in the FPC, a delayed exploratory shift in vmPFC and OFC, and sustained outcome evaluation to optimize future actions.
Efficient patient management in hospitals requires adaptive decision-making under time-varying demand and dynamic service environments. This study proposes a heterogeneous medical patient queueing model that integrates reinforcement learning with stochastic queue dynamics to minimize overall patient waiting time. The model distinguishes between two categories of service providers (SPs): those attending first-time patients and those serving returning patients. Each category may differ in service rate but not in medical specialty. Patient arrivals follow a non-homogeneous Poisson process (NHPP) to capture realistic time-dependent flow variations. A Q-learning framework with a supervised ε-greedy policy is developed to determine optimal operational actions, such as adding or reallocating service providers, based on system state and event type. Separate Q-tables are maintained for arrival and departure events to account for differing cost and reward dynamics. Simulation results demonstrate that the proposed model significantly reduces total waiting time and system cost compared with conventional homogeneous queue models. This approach provides a data-driven mechanism for dynamic hospital queue management and can be extended to broader healthcare resource optimization scenarios.
TolC is the outer membrane component of the tripartite RND efflux pump AcrAB-TolC. The underlying mechanism of TolC-dependent acid survival at physiological pH remains unclear. The present study aimed at understanding mechanism of TolC mediated acid survival in logarithmic phase culture of Enterobacter cloacae subsp. cloacae ATCC 13047 not preconditioned to acidic pH. Of the three distinct loci encoding tolC and tolC like proteins, only ECL_04363 (hereafter referred to as EctolC2) exhibited 86% identity with TolC protein from E. coli K12; deletion of which compromised survival at pH4.0. Tracking the periplasmic and cytoplasmic pH changes with fluorescent protein sensors indicated deletion of EctolC2 resulted in sustained cytoplasmic acidification and loss of pH homeostasis causing cell death. Exogenous supplementation of lysine significantly rescued acid induced lethality in EcΔtolC2 mutants. ΔmarRAB mutants exhibited slower death rate compared to Δrob mutants; underpinning role of the latter in upregulating EctolC2 at early time point of growth at pH4.0. Results of the study construe that EctolC2 provides reinforcement to the outer membrane preventing extreme acidification of the periplasm and thereby restoring cytoplasmic pH homeostasis. This study provides significant insight into mechanism of TolC dependent acid survival in E. cloacae.
The increasing use of cloud computing in hospitals, telemedicine, the Internet of Medical Things (IoMT) and real-time patient monitoring has made for an increasing trend of artificial intelligence-driven cloud security in hospitals. The growing reliance on distributed healthcare clouds layers the cyber-attack surface, however, with critical clinical operations now at risk from ransomware attacks, insider threats, API exploitation, and advanced persistent attacks. This research study introduces a novel AI-integrated cloud security framework tailored for safeguarding mission critical applications in the healthcare sector featuring an intelligent threat detection component, a probabilistic risk evaluation system, and an adaptive response orchestration system. The proposed architecture is built-in by using telemetry normalization, probabilistic behaviour modelling, deep autoencoders for anomaly detection, Bayesian approach for threat probability estimation, multi-objective risk scoring and reinforcement learning for adaptive mitigation. An experimental validation was performed with the CICIDS2017 dataset including around two million samples of network traffic data across various attack categories. The experimental results show that excellent performances have been achieved with an accuracy of 0.96, precision of 0.95, recall of 0.94, F1score of 0.95 and AUC of 0.98 with low latency of around 26 ms and reduced false positive rate of 0.03. A comparative analysis against the current cloud security methods also confirms the effectiveness of the proposed framework in delivering better operational security, response time and the availability of clinical services, providing the continuous clinical service that healthcare organizations require. The research showcases how incorporating AI with responsive cloud security mechanisms can offer a scalable and resilient defense against today's healthcare cloud infrastructures.
Healthcare professionals' support has been identified as an important factor for postpartum weight management, but barriers and facilitators exist among healthcare professionals supporting postpartum women to achieve healthy weight. To explore healthcare professionals' perceptions of barriers and facilitators of postpartum weight management. A systematic review and thematic synthesis of peer-reviewed qualitative studies was conducted. CINAHL, PubMed, EMBASE, PsycINFO, SCOPUS, and Web of Science databases were systematically searched from inception to 29th June 2025 to identify eligible published studies. The search was limited to peer-reviewed articles, with a focus on healthcare professionals' views regarding postpartum weight management. Data were extracted and thematically analysed, and identified themes were subsequently mapped onto the Capability, Opportunity, Motivation, and Behaviour (COMB) model. Four studies met the selection criteria. Within the capability domain, identified factors included limited awareness of postpartum weight management guidelines and deficits in postpartum weight management competencies. Within the opportunity domain, identified factors encompassed time constraints and competing clinical priorities, partner support, and the routine postnatal check opportunities. Within the motivation domain, identified factors included personal engagement in healthy lifestyle behaviours, professional role identity, regular follow-up and positive reinforcement, the sensitive nature of weight management discussions, and misconceptions regarding postpartum weight. The identified barriers and facilitators suggest that effective postpartum weight management interventions should be designed to address healthcare professionals' capability, opportunity, and motivation. The heterogeneity of study locations supports the potential transferability of these findings across diverse healthcare contexts.
Total knee arthroplasty (TKA) is a common and effective procedure for end-stage knee osteoarthritis, yet patients frequently encounter a complex and dynamic symptom experience during the initial period that can significantly impact their rehabilitation and quality of life. This study aimed to explore the symptom experience of patients within 6 weeks after TKA. A longitudinal qualitative study using semi-structured interviews. This study was conducted in an orthopaedics department of a tertiary general hospital in China. Interviews were conducted with 16 patients at 3-5 days postoperatively, 14 patients at 2 weeks postoperatively and 10 patients at 6 weeks postoperatively. Data were collected between July 2024 and November 2024. A purposive sampling method was used to recruit patients. Data were analysed using directed content analysis, with data collection and analysis performed concurrently. Three themes and nine subthemes were identified: (1) symptom perception, including perceived complexity of symptoms, self-identity conflict due to functional limitations, activation of negative emotions and psychological adaptation and interaction and amplification effects among symptoms; (2) symptom evaluation, characterised by the dynamic cognitive reframing of symptom meaning, self-blame tendency and internalisation of responsibility and interference from social and medical information; (3) symptom coping, involving dynamic evolution of active coping strategies and self-efficacy reinforcement, temporal characteristics of passive coping patterns and rehabilitation barriers. The symptom experience of patients who had TKA is complex and dynamic. Healthcare providers should implement tailored interventions based on patients' symptom experiences at different stages to facilitate symptom management, alleviate distress and negative emotions and improve quality of life.
Our understanding of sex differences in reward learning has been limited due to the predominant study of males. Here, we evaluated sex differences in flexible learning in two domains: the learning of stimulus-and action-based associations and their reversals. During action-based learning, rats selected between two identical visual stimuli presented on a touchscreen, where the spatial location predicted a higher probability of reward. For stimulus-based learning, rats chose between two distinct visual stimuli presented in pseudorandom spatial locations, one of which was associated with a higher probability of reward. Reversal phases involved switching reward contingency between the two actions or stimuli. We found that females did not differ in their discrimination or reversal learning accuracy compared to males, yet they collected fewer rewards and were more likely to omit trials than males in both domains. To gain a detailed understanding of differences across conditions, we modeled animals' trial-by-trial choices using reinforcement learning (RL) models and examined their steady-state behavior to capture transitions between distinct behavioral states. Although the estimated parameters of the best-fitting RL model revealed some sex differences, the model that incorporated transitions between different behavioral states provided a better overall fit to the data. This model also revealed that across all reversal phases, females exhibited a higher transition-specific lapse rate than males, indicating greater task disengagement once there was no need for further learning. Together, our fine-grained analysis of behavior adds to a growing literature on sex differences in flexible reward learning.
Bow hunter's syndrome (BHS) causes vertebrobasilar insufficiency when head rotation compresses the vertebral artery (VA).1,2 Although classically craniocervical, subaxial cases from osteophytic V2 compression lack standardized management with treatments ranging from decompression to stenting.3-5 During decompression, intraoperative VA patency confirmation is essential, as static postoperative imaging cannot reliably exclude residual dynamic compression, and delayed confirmation may necessitate reoperation.6,7 We present two cases of subaxial rotational occlusion syndrome (subaxial BHS) treated with anterior cervical decompression and intraoperative angiography in a hybrid operating suite. Case 1: an 80-year-old woman with neck pain, dizziness, tinnitus, near-syncope, and ocular symptoms on leftward head rotation; CT/dynamic angiography showed left VA narrowing at C4-C5 from spondylotic compression. Case 2: a 57-year-old woman with prior Eagle syndrome with vertigo, nausea, and blurred vision on left neck rotation; MRA demonstrated focal left V2 stenosis at C5-C6 from uncovertebral osteophytes and incidental distal VA fenestration. Both underwent anterior V2 exposure, ultrasonic drilling of osteophytes, Gore-Tex/fibrin glue reinforcement, and ACDF (C4-C5; C5-C6) to eliminate residual rotational motion at the decompressed segment. Intraoperative biplane angiography via left radial access (Isovue 300; 6 mL at 6 mL/sec per 2D run; 18 mL at 3 mL/sec for 3D acquisition) with simulated head turning, performed 10-15 minutes after hemostasis under general anesthesia, supplemented by 3D rotational angiography in Case 2 to characterize the VA fenestration, showed no head-turning compression and confirmed complete VA patency.8,9 Both were discharged on postoperative day three and complete symptom resolution was confirmed at 6 weeks.