Advancing Open Science

Prostate-specific membrane antigen (PSMA) is an established diagnostic and therapeutic target in prostate cancer. PSMA-targeted phototherapy comprises mechanistically distinct full-antibody near-infrared photoimmunotherapy (NIR-PIT), antibody-fragment photoimmunotherapy, ligand-targeted photodynamic therapy (PDT), and nanoparticle or multimodal systems. This systematic review evaluated their preclinical efficacy, safety, and translational readiness. MEDLINE/PubMed, Embase, Scopus, Web of Science Core Collection, the Cochrane Library, and ClinicalTrials.gov were searched for studies published from 1 January 2010 to 20 August 2026. Studies were classified into four carrier–mechanism groups before comparison across a seven-domain translational-readiness framework with explicit Early, Developing and Advanced anchors. Owing to substantial heterogeneity in targeting platforms, photoactive agents, models, illumination parameters and outcomes, meta-analysis was inappropriate; synthesis followed SWiM guidance. Animal studies were assessed using SYRCLE, and in vitro or ex vivo studies using an adapted ToxRTool with a decision-relevant reporting overlay. No numerical result used in the synthesis was obtained by figure digitisation. Thirty-three studies were included: 10 were confined to in vitro experiments, 19 combined in vitro and in vivo work, and four were in vivo only; one of the 19 combined studies also assessed fresh human prostate tissue ex vivo. Platforms were analysed as full-antibody NIR-PIT or photoimmunotherapy, antibody-fragment photoimmunotherapy, ligand-targeted PDT, and nanoparticle or multimodal systems. Within these classes, reports generally demonstrated preferential PSMA-dependent uptake and light-induced cytotoxicity, with multiple murine studies reporting tumour-growth inhibition and longer time to humane endpoints. Durable tumour-free responses occurred in selected small cohorts, particularly after repeated or combination treatment. Translational evidence remained limited: one study evaluated orthotopic prostate treatment in four dogs, while another demonstrated treatment-induced apoptosis in fresh prostate cancer tissue from 17 prostatectomy patients. No included full report described therapeutic treatment in patients. The decision-relevant in vitro/ex vivo appraisal classified 17 reports as higher, eight as moderate and four as lower reliability; animal studies frequently lacked sufficient reporting of randomisation, allocation concealment and blinding. Major translational barriers were heterogeneous PSMA expression, non-standardised light delivery, limited long-term safety data and insufficient validation in anatomically and immunologically representative models. PSMA-targeted phototherapy shows preclinical activity but remains clinically immature. Translation should proceed within mechanistically defined therapeutic classes and prioritise clinically realistic light dosimetry, PSMA heterogeneity, pharmacokinetic optimisation, functional safety, independent validation, and staged human evaluation. PROSPERO registration number CRD420261483194.

Background/Objectives: This retrospective benchmark evaluated scores for selecting regions of panoramic tooth segmentations for review. Methods: Three-member U-Net ensembles trained on UFBA-425 were evaluated on 590 held-out images and subsequently on 329 photographed radiographs. The original mixed-granularity pipeline comparison used the calibrated residual ensemble on 100 pediatric images. Cells in a 2 × 4 grid received regional mutual information (MI) scores but shared one whole-image soft Dice confidence (SDC) value. A lower Dice-risk area under the risk–coverage curve (AURC) indicated better ranking against source masks. Interaction tests, quality subset analyses, and prediction-positive sensitivity analyses were post hoc. Image-key bootstrap intervals may be too narrow for patient-level inference because patient independence was unverified. Results: The primary MI-minus-SDC AURC difference was +0.007 (95% confidence interval −0.033 to +0.041), leaving the comparison inconclusive. Subsequent evaluations were exploratory. Analyses of reference-positive cells (containing reference tooth pixels) received co-principal emphasis after outcomes were known. MI had a higher Dice-risk AURC than the replicated SDC in the pediatric main subset (+0.238) and the local SDC in AKUDENTAL (+0.157 residual, +0.134 ResNet18). When 20% of grid cells were reviewed, the mean maximum pixel probability (MSP) captured more error pixels than MI in seven of ten source/model comparisons, including both AKUDENTAL ensembles. On photographed radiographs, the MSP led by 15.73 percentage points for the residual ensemble and MI by 3.66 for ResNet18. For ResNet18, equal weighting of false-positive and false-negative capture favored MSP by 6.59 points. Conclusions: The primary comparison established neither superiority nor equivalence. Exploratory capture varied by source, model, and error mix.

This article presents the implementation of a passive balancing strategy in an energy storage network based on LiPo cells. The proposed system is intended to operate in a 5.6 kWh battery pack for two-wheeled electric vehicle applications. The battery balancer allows voltage equalization between the LiPo cells before and after the charging process, ensuring controlled charging cycle for each cell. The LiPo cell balancer uses a passive method which transforms the excess of electric energy into heat by means of a resistance described in the Joule Effect. In order to minimize energy losses, three operating modes were proposed: post-charge, pre-charge, and storage node balancing. The post-charge mode reduces the cell voltage to 4.20 V, while the voltage is set to 2.8 V in the pre-charge mode. The design and construction details of a six-cell module prototype are described in this paper, along with experimental research. The balancer system was successfully operated in the series modules of the LiPo battery pack using the proposed modes of operation.

A compact dual-polarized magnetoelectric (ME) dipole antenna combining an X-shaped slotted radiator with four corner loading vias is investigated. Simulations of configurations with zero, 24, and four additional loading vias show the trade-off between low-frequency extension and upper-band realized gain. The final four-via configuration exhibits simulated VSWR ≤ 2 bands of 2.25–4.93 GHz (74.6%) for Port 1 and 2.27–4.89 GHz (73.2%) for Port 2, with a peak realized gain of approximately 6.1 dBi. The fabricated 40 mm × 40 mm × 12.6 mm prototype achieves a measured common impedance bandwidth of 2.25–4.90 GHz (74.1%) and a measured peak realized gain of 6.3 dBi. Across the evaluated frequency range of 2.3–4.9 GHz, the measured radiation efficiencies reach approximately 85% and 83% for Ports 1 and 2, respectively. The wavelength-normalized dimensions are 0.30${\lambda }_{\mathrm{L}}$× 0.30${\lambda }_{\mathrm{L}}$× 0.09${\lambda }_{\mathrm{L}}$ at 2.25 GHz. Relative to an equal-footprint reference normalized at its 3.00-GHz lower matched-band edge, the wavelength-normalized aperture-area metric is reduced by 43.8%, while the physical aperture remains unchanged. The measured interport transmission falls below approximately $-10$ dB near 2.5 GHz and is further suppressed over much of the upper band. The results demonstrate a broad common impedance band and forward-directed radiation from a compact electrical aperture, with interport isolation varying across the operating band.

This paper proposes a core-shared inductor for a two-stage low-voltage DC–DC converter (LDC) system to improve power density while satisfying a wide input and output voltage range. In the proposed structure, the 4-switch buck–boost and LLC resonant inductors are integrated into a shared magnetic core assembly to reduce their combined volume. To verify the feasibility of the proposed approach, a 2-kW LDC prototype was experimentally evaluated using both the conventional separate inductors and the proposed core-shared inductor. Experimental results demonstrate that the proposed system maintains conversion efficiency comparable to that of the conventional system, with no appreciable efficiency improvement, while reducing the combined volume and mass of the 4-switch buck–boost and LLC resonant inductors by approximately 9% and 10%, respectively, compared with the conventional design. Therefore, the proposed magnetic integration method provides an effective approach to reducing the volume of magnetic components in wide-range two-stage LDC systems.

Bisphenol A (BPA), a widespread endocrine-disrupting chemical, has been associated with male reproductive health, yet its structural analog bisphenol S (BPS), selected as a primary alternative, may also pose similar risks. Their comparative reproductive toxic mechanisms remain poorly defined. By integrating network toxicology and molecular docking, we predicted and compared BPA- and BPS-related target and pathway profiles in male reproductive toxicity, and identified candidate hub genes potentially dysregulated in testicular pathology using public datasets. Both chemicals shared ESR1 and TP53 as overlapping targets, and enrichment analysis showed that both were enriched in ESR1/AR signaling and endocrine resistance pathways. BPA was preferentially linked to impairment of immune-barrier function with AKT1, TGFB1, TLR4, and STAT3, whereas BPS mapped to metabolic-oxidative stress processes via MDM2, EGFR, and GSTP1. Immune infiltration analysis linked BPA-related genes inversely to NK cell abundance. Transcription factor analysis identified MYCN and PGR as hub nodes with distinct downstream networks. Molecular docking predicted generally favorable binding of both chemicals to key targets, with target-dependent differences in predicted affinity. These in silico results suggest that BPA and BPS may affect male reproductive health through partly distinct mechanisms, with BPA more closely associated with immune-related pathways and BPS with metabolic-oxidative stress pathways.

Climate change may elicit divergent spatial responses among species sharing ecological traits or close taxonomic relationships. Here, we compared six Actinidia species organized into three focal pairs: the cold-tolerant pair (A. arguta–A. kolomikta), the closely related cultivated pair (A. chinensis–A. deliciosa), and the warm-humid pair (A. eriantha–A. latifolia). Species-specific MaxEnt models were used to project potentially suitable habitats in China under SSP126, SSP245, and SSP585 for 2041–2060 and 2081–2100, based on future climate data from the BCC-CSM2-MR model. PCA-env analyses and randomization tests assessed niche overlap and equivalency, while habitat transitions and centroid shifts characterized future spatial responses. Five-fold spatial block cross-validation yielded mean AUC, TSS, and Kappa values of 0.9867–0.9944, 0.8247–0.8727, and 0.4345–0.7676, respectively. BIO4 and BIO16 were influential predictors across most species. The cold-tolerant pair generally showed increases in suitable habitat area and northward centroid shifts. The cultivated pair exhibited contrasting responses, with substantial habitat losses for A. chinensis under late-century SSP585 and scenario-dependent gains and losses for A. deliciosa. A. eriantha remained relatively stable under most scenarios, whereas A. latifolia showed habitat expansion across all scenarios. All three pairs exhibited substantial niche overlap, but niche equivalency was rejected by both Schoener’s D and Warren’s I for the cultivated and warm-humid pairs, and only by Schoener’s D for the cold-tolerant pair. By showing that ecological or taxonomic similarity and substantial niche overlap do not necessarily correspond to parallel future spatial responses, this paired-comparison framework highlights the importance of evaluating species-specific responses to climate change. The projections provide conditional regional estimates of environmental suitability within China under the selected climate model and SSP scenarios.

Background/Objectives: Germectomy is the surgical removal of a developing third molar, usually before one-third of its root has formed. In children and adolescents, it is performed mainly for orthodontic reasons, without an established advantage over retention or later removal. This review summarizes clinical studies published in 2016–2026. Methods: English-language studies from PubMed, Scopus and Web of Science were synthesized narratively, graded with the Strength of Recommendation Taxonomy (SORT) and assessed for risk of bias. Results: Of the 77 records, 14 studies with 654 germectomies were included, and four of them were randomized trials; none reached SORT level 1, and no comparative study was at low risk of bias. The definitions and the staging criteria varied; five studies did not report the developmental stage. The indications were mostly orthodontic, but in the only randomized comparison with retention, germectomy did not shorten the uprighting of impacted second molars. Delayed-onset infection, the most frequent complication, followed 9.2% of 218 germectomies in the largest series and appeared 2–8 weeks after surgery, when seven studies had ended follow-up. Three other studies reported complications after 3.6–4.5% of germectomies; where specified, none was a nerve injury. Pain and oral health-related quality of life were assessed only in the first postoperative week, when pain peaked in the first days. Two maxillary case reports described germ displacement and late partial tooth formation. Conclusions: Germectomy rarely causes serious complications, but its orthodontic benefit is unproven. Families should expect pain in the first days and possible infection weeks later. Prospective comparisons of germectomy and monitoring of the germ, with standardized staging and follow-up of at least one year, should test the orthodontic indication.