The Dynamic Job Shop Scheduling Problem (DJSSP) is a typical scheduling task that requires rescheduling in the presence of unexpected events, such as random job arrivals and urgent orders. However, due to the varying scales of scheduling problems, existing rescheduling methods struggle to effectively reuse trained scheduling strategies or benefit from previous transfer learning from previous models. To address this challenge, we propose a Double Priority Experience Replay (DPER) mechanism integrated within the Proximal Policy Optimization (PPO) framework (DPER-PPO). First, we introduce a generalized disjunctive graph to model random job arrivals and combine it with an extensible state representation consisting of 10 distinct features to optimize completion time, thereby meeting the dynamic and adaptive requirements of DJSSP. Next, we develop a comprehensive multidimensional action space with adaptive weighting rules, enhancing the action coverage and improving the global optimization capability of the algorithm. Finally, the proposed DPER mechanism, integrated within the PPO framework, enhances elite sample utilization and accelerates agent learning in dynamic environments. Static experimental results on classic benchmark instances demonstrate that our scheduling model outperforms existing Deep reinforcement learning (DRL) methods in terms of average performance. Furthermore, dynamic scheduling experiments show that, when encountering unexpected events such as random job arrivals and urgent orders, our model achieves better results than the Priority scheduling rules (PDR) scheduling method and other DRL approaches within a reasonable time frame. In addition, the results of the analysis of variance (ANOVA) test further confirm the statistical significance and effectiveness of our proposed method.

This paper presents REACH, an advanced machine learning framework to deliver comprehensive decision support for older patient prioritisation. The framework employs a Mixture of Experts (MoE) architecture, integrating multiple specialised predictive models to simultaneously address four critical dimensions: complex care pathway classification, aged residential care prediction, early supported discharge assessment, and mortality risk evaluation. The MoE architecture features a context-aware attention-based gating mechanism that dynamically adjusts expert contributions based on patient characteristics and operational factors. The framework’s implements an automated model selection, and hyperparameter optimisation through a Combined Algorithm Selection and Hyperparameter-tuning methodology. This study is a conceptual theory extending on the fundamentals of REACH to create a multi-dimensional model. This work addresses a critical gap in healthcare delivery by providing a comprehensive, data-driven approach to optimising care pathways for older patients while considering resource constraints and operational efficiency.

Objectives Artificial Intelligence (AI) in healthcare has advanced rapidly in prediction, classification, and pattern recognition, yet high-stakes clinical and operational decisions are not prediction problems alone. Clinical, operational and policy choices also require traceable reasoning, guideline fidelity, attention to institutional constraints, and accountability. This critical narrative review aims to synthesise evidence on neuro-symbolic and knowledge-infused AI for healthcare decision-making, with particular emphasis on decision-maker-governed systems: hybrid architectures in which clinicians, managers, or policymakers author, approve, or govern the symbolic knowledge structures that materially influence system behaviour. Methods We conducted a critical narrative review informed by a structured search of ten databases (last updated May 2026), screening 3,892 de-duplicated records to 101 sources retained for in-depth synthesis against three pre-specified analytic questions. The review proposes an operational definition for this category, distinguishes it from hybrid systems that merely import fixed external knowledge, and appraises evidence across clinical decision support, information extraction, hospital operations, and population-health governance. Results The literature suggests three cautious conclusions. Hybrid designs appear most credible in rule-bound, audit-sensitive, and workflow-dependent settings, including documentation validation, guideline-based recommendation, multimodal oncology workflows, and service optimisation under explicit constraints. Yet the empirical base remains limited; prospective evidence is sparse, comparisons between stakeholder-authored and non-stakeholder-authored hybrids are rare, and many claims regarding trust, fairness, and actionability remain largely theoretical. Symbolic layers also introduce notable burdens, including knowledge acquisition, ontology misalignment, rule conflict, and maintenance overhead. Conclusions Decision-maker-governed AI should therefore be understood not as a proven solution, but as a promising governance-oriented design strategy that requires prospective evaluation, fairness auditing, knowledge-lifecycle tooling, transparent institutional oversight, and cross-site implementation studies before its translational potential can be fully assessed. Public Interest Summary Artificial intelligence (AI) is increasingly used in healthcare, but making a good prediction is not the same as making a safe decision. Real clinical, operational, and policy choices depend on rules, guidelines, and lines of accountability that a purely data-driven model does not capture. This review examines a class of AI systems in which doctors, hospital managers, and policymakers actively write and govern the rules that shape how the AI behaves, rather than leaving decisions to a black box. We find these systems are most convincing for well-structured tasks such as checking clinical documentation, following guidelines, and scheduling hospital resources, but that strong real-world evidence is still limited. We argue such systems should be seen as a promising, accountability-focused design strategy that now needs careful evaluation in real health services before wide adoption.

The detection of biomatter threats in baggage is a critical task for ensuring biosecurity, especially at international borders. This study introduces a novel methodology for converting 3D computed tomography (CT) volumetric data into 2D representations to enable efficient object detection using state-of-the-art 2D models. Systematic time-series feature engineering techniques were applied to vertical variation sequences from 3D data, transforming complex volumetric structures into compact and informative 2D projections. The proposed method leverages Light Gradient Boosting Machine and Random Forest (RF) estimators, combined with dimensionality reduction techniques such as Principal Component Analysis (PCA) and Partial Least Squares Regression, to select the most discriminative features. Evaluations using the YOLOv10l object detection framework demonstrated high detection accuracy, achieving a mean average precision of 0.851 with RF and PCA. Despite challenges such as class imbalance and computational trade-offs, the methodology offers a scalable, efficient, and highly accurate approach to biomatter detection. This study not only addresses limitations in current 3D image analysis techniques but also highlights potential applications in medical imaging and industrial inspection.

This study investigated the effectiveness of Artificial Intelligence (AI) and Machine Learning (ML) approaches for anomaly detection in rheumatology, focusing on their potential to improve clinical insights and identify deviations in disease patterns and diagnostics. This systematic review adhered to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive search was conducted across PubMed, Cochrane Library, Web of Science, Scopus, and EBSCO databases to identify cohort studies that developed and/or validated AI and ML models for anomaly detection in rheumatology. Data were extracted and studies were critically evaluated using the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) guidelines and the Prediction Model Risk of Bias Assessment Tool (PROBAST), with the search last updated on January 28, 2025. The systematic search yielded 2,089 unique citations, from which 40 studies met the inclusion criteria. AI and ML models included demonstrated high performance measures. Most studies focused on rheumatoid arthritis, utilising imaging data, biomarkers, and electronic health records. The quality assessment revealed that 72.5

Effective baggage screening is crucial for biosecurity, requiring accurate detection of biomatter threats in 3D computed tomography (CT) scans. This study introduces a novel approach that transforms 3D volumetric data into 2D representations using signal processing techniques and feature extraction. Feature selection and dimensionality reduction techniques, including Principal Component Analysis (PCA) and Partial Least Squares Regression (PLSR), are applied to enhance classification performance. The results show that the choice of projection view significantly influences detection accuracy, with the $(z, x)$-view achieving the highest mAP@0.5 score of 0.810. Precision-recall analysis further reveals variability in detection performance across biomatter classes, highlighting the challenges posed by similar density profiles. The findings demonstrate the effectiveness of 2D representation learning for biomatter detection while acknowledging limitations due to information loss. Future work will explore hybrid 2D-3D models to further enhance detection capabilities.

Severe mental illness is linked to poor physical health and shorter life expectancy, yet research on how individuals experiencing mental illness view and on improve their physical health is limited. This study investigates the perceptions of individuals experiencing mental illness regarding their physical health, utilising a mixed-methods approach. Phase I involved quantitative and qualitative data from an online Qualtrics survey, which included the 12-item Short Form (SF-12) survey to measure participants' quality of life and assess self-reported physical and mental health. Key findings from Phase I revealed significant relationships between lower Physical Component Summary (PCS) scores and factors such as the frequency of GP visits. Additionally, exercise preferences were found to significantly impact Mental Component Summary (MCS) scores, with individuals who preferred a mix of exercise settings reporting higher MCS scores compared to those who exercised alone or with a training partner. Phase II explored these findings further through semi-structured interviews, where participants discussed themes including physical health perceptions, the role of medication and the importance of the general practitioner relationship. Thematic analysis revealed five main barriers to improving physical health: accessibility and availability of services, motivation, staff attitudes, medication side effects and the experience of diagnostic overshadowing. Participants reported viewing physical and mental health as interconnected and expressed a desire for more collaborative care. The results suggest that strengthening the relationship with GPs and increasing awareness of medication side effects may improve physical health outcomes for individuals experiencing mental illness. Mental health nurses can play a pivotal role in enhancing physical health outcomes by monitoring, supporting health-improving strategies and facilitating access to primary care services.

Mutations drive genetic variation, fueling both oncogenesis and species evolution. The mutation rate varies across the genome, potentially influenced by chromatin organization through histone modifications and other factors. However, the precise relationship between chromatin structure and mutation rate remains poorly understood and needs further investigation. One such modification, the methylation of histone H3 at lysine 9 (H3K9me), is known to form heterochromatin and repress transcription in euchromatin, thereby maintaining genome stability essential for organism survival. This study aimed to elucidate the effect of H3K9 methylation, in isolation from other histone markers, on the mutation rate in fission yeast. Employing fluctuation assays and statistical analysis, our innovative methodology estimates the mutation rates of a single gene under two different conditions within a single experiment using an isogenic clone in Fission yeast. Our findings highlight a potential association between H3K9 methylation and the phenotypic mutation rate of the same gene, ura4+. For prospective researchers, this study introduces a new experimental approach that offers unprecedented accuracy in gene analysis, with implications for both genetic research and epigenetic therapy.

A systematic feature-engineering approach to generate informative 2D representations of 3D data is introduced. In this method, the sequences of voxels along one axis of the 3D image are treated as spatial variation sequences. These sequences are projected into a 783-dimensional feature space using algorithms from statistics, signal processing, complexity theory as well as time-series forecasting and financial time-series analysis. The resulting two-dimensional image has 783 layers from which the most relevant three layers are chosen using a combination of univariate and multivariate feature selection. This process effectively converts the volumetric data into a two-dimensional three-layer image which can then be used as input to established object detection models. The validation of the method is conducted on an object detection application, involving the identification of biomatter threats in 3D X-ray scans of international travellers’ baggage. The 3D scans were recorded at the Airport in Auckland, New Zealand, and comprised 1525 biomatter threats distributed over 690 different bags. Various object detection models from the YOLO series are tested on this dataset. The YOLOv5l model achieved the highest mAP@0.5 of 0.878 on the validation dataset. Our results demonstrate that the methodologies of time-series classification and pattern recognition can be combined to implement efficient pattern recognition on 3D data sets with small sample sizes.

One of the critical issues in healthcare management is the operating room (OR) scheduling problem. Solutions to this problem consider surgery durations and allocate elective surgeries to OR sessions in order to create surgical lists of high quality. Determining the quality of a surgical list is a key undertaking within OR scheduling and is the focus of this research. Currently, probability- and/or expectation-based measures of surgical lists are used instead of statistical distributions of surgery lists to measure quality. The use of multiple measures, e.g., a combination of expectation and probability to assess a surgical list, complicates OR scheduling, so we introduce a new single measure – the OR scheduling metric – for evaluating surgical lists before their realisations, i.e., for use within OR scheduling. We apply the OR scheduling metric to an actual elective dataset and use simulation to demonstrate its use, including customised scheduling rules. We recommend the adoption of a benchmarked OR scheduling metric by the elective surgical services in hospitals with expected practical benefits in the long run, i.e., simpler OR scheduling and more desirable room utilisation, to be similar to that observed in our simulations.

International transport security policy requires the baggage screening. In airports and railway stations, these operations are processed manually with the help of X-ray or computed tomography machines. There is a need for an automatic system which could reduce the time of the screening process and possibly, increase the accuracy of the detections. More than that, there is a demand for developing and evaluating methodologies for learning on 3D image-like data, which has been addressed only recently, mostly in the field of medical imaging. The main objective of this research is to develop a framework for object detection in 3D computed tomography scans for high-throughput security applications. In this paper, a literature review on the topic of 3D image recognition is presented, and a transfer learning approach is evaluated on the security risk detection task in X-ray images.

Structured Abstract Objectives Current policies for older patients do not adequately address the barriers to effective implementation of optimal care models in New Zealand, partly due to differences in patient definitions and the in-patient pathway they should follow through hospital. This research aims to: (a) synthesise a definition of a complex older patient; (b) identify and explore primary and secondary health measures; and (c) identify the primary components of a care model suitable for a tertiary hospital in the midland region of the North Island of New Zealand. Method This mixed-methods study utilised a convergence model, in which qualitative and quantitative data were investigated separately and then combined for interpretation. Semi-structured interviews ( n =11) were analysed using a general inductive method of enquiry to develop key codes, categories and themes. Univariate data analysis was employed using six years of routinely collected data of patients admitted to the emergency department and inpatient units ( n =261,773) of the tertiary hospital. Results A definition of a complex older patient was determined that incorporates chronic conditions, comorbidities and iatrogenic complications, functional decline, activities of daily living, case fatality, mortality, hospital length of stay, hospital costs, discharge destination, hospital readmission and emergency department revisit and age – not necessarily over 65 years old. Well-performing geriatric care models were found to include patient-centred care, frequent medical review, early rehabilitation, early discharge planning, a prepared environment and multidisciplinary teams. Conclusions The findings of this New Zealand study increase understanding of acute geriatric care for complex older patients by filling a gap in policies and strategies, identifying potential components of an optimal care model and defining a complex geriatric patient. Implications for Public Health The findings of this study present actionable opportunities for clinicians, managers, academics and policymakers to better understand a complex older patient in New Zealand, with significant relevance also for international geriatric care and to establish an effective acute geriatric care model that leads to beneficial health outcomes and provides safeguard mechanisms.

BackgroundThe identification and assessment of sentinel lymph nodes (SLNs) in breast cancer is important for optimised patient management. The aim of this study was to develop an interactive 3D breast SLN atlas and to perform statistical analyses of lymphatic drainage patterns and tumour prevalence.MethodsA total of 861 early-stage breast cancer patients who underwent preoperative lymphoscintigraphy and SPECT/CT were included. Lymphatic drainage and tumour prevalence statistics were computed using Bayesian inference, non-parametric bootstrapping, and regression techniques. Image registration of SPECT/CT to a reference patient CT was carried out on 350 patients, and SLN positions transformed relative to the reference CT. The reference CT was segmented to visualise bones and muscles, and SLN distributions compared with the European Society for Therapeutic Radiology and Oncology (ESTRO) clinical target volumes (CTVs). The SLN atlas and statistical analyses were integrated into a graphical user interface (GUI).ResultsDirect lymphatic drainage to the axilla level I (anterior) node field was most common (77.2%), followed by the internal mammary node field (30.4%). Tumour prevalence was highest in the upper outer breast quadrant (22.9%) followed by the retroareolar region (12.8%). The 3D atlas had 765 SLNs from 335 patients, with 33.3-66.7% of axillary SLNs and 25.4% of internal mammary SLNs covered by ESTRO CTVs.ConclusionThe interactive 3D atlas effectively displays breast SLN distribution and statistics for a large patient cohort. The atlas is freely available to download and is a valuable educational resource that could be used in future to guide treatment.

IntroductionThis study aimed to assess the relationship between preparation times and operative procedures for elective orthopaedic surgery. A clearer understanding of these relationships may facilitate list organisation and thereby contribute to improved operating theatre efficiency.MethodsTwo years of elective orthopaedic theatre data was retrospectively analysed. The hospital medical information unit provided de- identified data for 2015 and 2016 elective orthopaedic cases, from which were selected seven categories of procedures with sufficient numbers to allow further analysis - primary hip and knee replacement, spinal surgery, shoulder surgery (excluding shoulder replacement), knee surgery, foot and ankle surgery (excluding ankle replacement), Dupuytrens surgery and general orthopaedic surgery. The data analysed included patient age, ASA grade, operation, operation time, and preparation time (calculated as the time from the start of the anaesthetic proceedings to the patient's admission to Recovery, with the operating time [skin incision to skin closure] subtracted). Statistical analysis of the data was undertaken.ResultsA total of 1596 procedures performed over the two year period were analysed. Preparation times for the different procedures were assessed, along with the relationship to the procedure complexity. Neither age nor ASA correlated strongly with preparation times. Spine procedures had greater preparation times than hip and knee arthroplasty. Greater uniformity in preparation times for hip and knee arthroplasty was seen across the anaesthetic group than operative times across the surgeon group.DiscussionPreparation times are just one aspect that may be evaluated with regard to theatre utilisation. This study did not address the theatre turn-over time between cases, which includes transfer of the patient from the admitting/pre-operative area into the theatre.ConclusionPreparation times for elective procedures follow a pattern which may be used to inform list planning, with the potential for greater theatre efficiencies with regard to list utilisation and staff allocation.

Abstract Mutations are the driving force behind genetic variation, fueling both the oncogenesis and evolution of species. The mutation rate varies across the genome, potentially in response to chromatin organization by histone modifications and other factors. However, the exact relationship between the two is yet to be fully understood and requires further investigation. One modification involves the methylation of histone H3 at lysine 9, which creates heterochromatin and represses transcription in euchromatin to maintain genome stability for organism survival. This study aimed to determine the effect of H3K9 methylation alone, without other histone markers, on the mutation rate in fission yeast using fluctuation assays and statistical analysis. Our groundbreaking method has been proven to accurately estimate mutation rates of a single gene under two different conditions in a single experiment using one isogenic clone. Our research results demonstrate that the H3K9me markers increase the phenotypic mutation rate of the same gene. For prospective researchers, this study presents an innovative experimental approach that ensures unparalleled accuracy in gene analysis for genetics applications and epigenetic therapy.

Sentinel node biopsy (SNB) is a common staging tool for breast cancer. Initially, peritumoral (PT) injections were used, however subareolar (SA) injections were later introduced to simplify the technique. Controversy remains regarding whether PT and SA injections map the same sentinel lymph nodes (SLNs). This study aimed to determine whether the regional location of breast SLNs differs when using PT versus SA injections using a large dataset from a single institution. A total of 1035 patients who underwent breast SNB (PT injections: n = 858 and SA injections: n = 177) with lymphoscintigraphy and SPECT/CT were included. The identified SLN locations using SA injections were compared with those using PT injections. Differences in drainage proportions and odds ratios (ORs) for each clockface breast region and the whole breast were calculated using a two-proportion z-test and Fisher’s Exact Test. A higher proportion of internal mammary SLNs were identified using PT injections for the whole breast (0.30 versus 0.09) and for all breast regions, with all regions showing statistical significance except the upper outer quadrant. Similarly, ORs showed identification of internal mammary SLNs was significantly higher when using PT injections (4.35, 95

Simulation facilitates the understanding and improvement of complex systems. Conceptual modelling is a key step in simulation studies. It has gained recognition because it may both increase engagement with stakeholders and decrease the time to implement a simulation. This scoping review’s objective is to highlight approaches and platforms for electronically representing models from 1999 to 2020. The motivation is that electronic representations facilitate the sharing of conceptual models. The contribution from the review to the research of conceptual modelling and simulation is to show that conceptual models are electronically represented by broadly speaking either General-Purpose or Domain-Specific Modelling Languages. There is a slight trend towards the latter in order to better deal with application specificities and improve unambiguity in model representations, though. Thus, we identify modelling approaches, platforms, and features for electronically representing conceptual models with the potential to fill the gap between conceptual models and their corresponding simulation implementations.

The Luria-Delbrück fluctuation assay is an essential experiment in calculating mutation rates, especially in genetic and mutation research. Its reliability and accuracy have made it the go-to method for numerous researchers. In this article, we provide R-codes that statistically analyze the assay results more easily and offer the most challenging codes for calculating 95% confidence intervals based on the gold standard method “Ma-Sandri-Sarkar Maximum Likelihood.” Recently, the maximization of the likelihood function through optimization functions in R can be a challenging task. The recursive format of the likelihood function is known to cause memory stack issues. Our findings indicate that utilizing a non-recursive version of the function can increase the tractability of the maximization process. With these codes, future scientists can unlock valuable statistical insights related to the biological mechanisms that drive genetic variation and can, therefore, contribute to developing novel therapeutic interventions and innovative solutions to various biological and medical challenges.