Accurately characterising the heterogeneous presentation of Parkinson’s disease (PD) requires effective integration of neuroimaging and clinical information, yet many multimodal models suffer from limited interpretability, class imbalance, or suboptimal fusion of imaging and tabular data. We propose the Class-Weighted Sparse-Attention Fusion Network (SAFN), an interpretable deep learning framework designed for robust multimodal Parkinson’s disease classification. SAFN integrates MRI-derived quantitative neuroimaging biomarkers (cortical thickness and volumetry), clinical assessments, and demographic variables using modality-specific encoders and a symmetric cross-attention mechanism to model non-linear interactions between imaging-derived and clinical signal representations. A sparsity-constrained attention-gating fusion layer adaptively prioritises informative modalities, while a Class-Balanced Focal Loss (β=0.999,γ=1.5) addresses dataset imbalance without synthetic oversampling. Evaluated on 703 participants (570 PD, 133 healthy controls) from the Parkinson’s Progression Markers Initiative using subject-wise five-fold cross-validation, SAFN achieves an accuracy of 0.98 ± 0.02 and a PR–AUC of 1.00 ± 0.00, achieving competitive performance relative to established machine-learning and deep-learning baselines. Interpretability analysis reveals clinically coherent attribution patterns, with modality gating assigning approximately 60% of the fused representation to clinical assessments, consistent with Movement Disorder Society diagnostic principles, while MRI-derived biomarkers provide complementary structural information within the multimodal fusion framework. Attribution estimates were validated against SHAP and permutation-based analysis, which independently identified clinical assessments as the dominant modality, and cross-fold ranking stability was quantified explicitly. These results demonstrate SAFN as an interpretable and robust multimodal fusion framework for Parkinson’s disease classification within a harmonised research cohort setting and indicate that further external, prospective, and usability-focused validation is required before considering clinical decision-support use.
Antimicrobial stewardship (AMS) programmes are central to tackling antimicrobial resistance. Comparative evidence on how barriers differ between high-income countries (HICs) and low-/middle-income countries (LMICs) in tertiary care hospitals is limited. The review aims to identify and compare multilevel barriers to implementing AMS programmes in tertiary care hospitals in HICs and LMICs. We conducted a systematic scoping review following PRISMA-ScR and Joanna Briggs Institute guidance. PubMed and Web of Science were searched for English-language primary studies from January 2015 to November 2025. Studies were considered eligible if they examined barriers to implementing AMS in adult inpatient services in tertiary care hospitals. Using a framework-based thematic approach, barriers were grouped into eight themes and mapped to individual, team, organizational, and system levels. Out of 2311 records, 57 studies met inclusion criteria (23 = HICs, 34 = LMICs). Knowledge, education, and confidence gaps were the most frequent barrier theme in both groups, reported in about three-quarters of studies. Staffing and resource constraints and organisational/governance barriers were also highly prevalent, particularly in LMIC hospitals. Workflow and documentation problems and data-system limitations were more prominent in HIC studies, whereas structural health system constraints were more prominent in LMICs. Qualitative and mixed-methods studies identified a broader range of barriers than surveys or observational designs. AMS implementation in tertiary hospitals is constrained by overlapping multilevel barriers that are broadly similar across income groups but differ in emphasis. Tailored strategies that jointly strengthen workforce capacity, infrastructure and governance are needed, with particular attention to system-level constraints in LMICs and workflow and data challenges in HICs.
Changes to fire regimes, including application of prescribed burns, represent a threatening process for species worldwide. A challenge is to determine how the spatial pattern of prescribed burns can be manipulated to minimise impacts on species of concern. We used an experimental approach to assess short-term responses of a threatened marsupial, the brush-tailed phascogale Phascogale tapoatafa, to landscape-scale prescribed burns. We used remote cameras to survey this species in 14 forested landscapes subject to burns that varied in % burn cover. Analysis was undertaken using occupancy modelling at both landscape and plot level (5 plots per landscape). The brush-tailed phascogale was widespread but scarce, recorded in 13/14 landscapes and 25/70 plots. While occupancy was not influenced by fire or habitat covariates, detectability decreased with time since camera deployment, and with increasing burn cover. Thus, although persisting, its activity and presumed population density declined as more of the landscape was burned. A negative relationship between burn cover and burn patchiness in these landscapes suggests a patchy mosaic, rather than complete burn, will moderate detrimental effects. A patchy mosaic can contribute to reducing fuel loads while limiting loss of critical structural features on which many species depend, but which take decades to replace.
Rights of wetlands shifts the people-wetlands relationship towards one recognizing the intrinsic rights and living beingness of wetlands, embodying reciprocity, kinship, and gratitude. Recognition of wetland rights occurs within the broader Rights of Nature movement, which has often been led by Indigenous Peoples and local communities from many cultural traditions. Intrinsic wetland rights centre on the right to exist and incorporate seven other rights, forming a holistic framework that enables preservation of wetland ecosystem integrity. These rights provide the basis for a wetland’s thriving existence. The eight rights are: • the right to exist. • their natural place in the landscape; • natural and connected hydrologic regimes; • natural climatic conditions; • naturally occurring biodiversity; • natural ecosystem processes; • natural water, soil, and air quality (i.e., structural integrity and form); and. • regeneration and a naturally determined future. Moral and ethical responsibilities towards these rights include fostering harmonious relationships in and with wetlands, respecting interdependence of all wetland components, safeguarding and fostering conditions enabling rights of wetlands, and acting in a precautionary manner when faced with risks of harm to wetlands. Legal and policy responsibilities include acknowledging plural worldviews, recognising legal standing and political orientations to represent and defend wetland rights, providing adequate means for wetland representation and defending and advocating for wetlands, and using best available science to restore degraded wetlands and support regeneration, thereby supporting harmonious societal relationships with wetlands. Opportunities for implementation of rights of wetlands at a variety of scales (individual, community, national, and global) are explored, and training/guidance resources are referenced.
Luciola Laporte, 1833 is a taxonomically challenging firefly taxon. Five species occur in Europe: L. italica, L. lusitanica, L. pedemontana, L. mingrelica and L. novaki. Their taxonomic history has been obscured by synonymies, insufficient diagnoses and a lack of integrative studies. This species assembly is particularly important because it includes the type species of Luciola, L. lusitanica. Legacy diagnostic traits have proven ineffective in distinguishing Luciola species across Europe, and accurate species identification is challenging, which hinders further studies and conservation efforts. Three species - L. lusitanica, L. pedemontana and L. mingrelica - share similar external morphology (e.g. pale thorax, femora darkened apically, no pronotal black spot) and have often been treated collectively as L. lusitanica sensu lato, underscoring the need for revised species boundaries. We analyzed COI sequences from 86 L. lusitanica s.l. and six L. italica specimens. We found L. lusitanica s.l. to be paraphyletic (excluding L. italica). Using integrative species delimitation (combining morphological, molecular and behavioural evidence), we found support for the distinctiveness of L. lusitanica, L. pedemontana, L. mingrelica and L. italica, proposing updated diagnoses and distributions for each species. Our richly illustrated and updated diagnoses are intended to significantly facilitate species identification and help stabilize Luciola taxonomy.
