BMC Urology - Structured reporting of B-mode, color Doppler, and CEUS in testicular tumor assessment: a reader study with urologist ratings

This study examines whether structured reporting (SR) improves the clinical usability of multiparametric ultrasound reports for testicular tumor assessment. The authors found that, compared with traditional free-text reporting (FTR), structured reports significantly improve readability (97% vs. 10%) and facilitated a more efficient information extraction (99% vs. 92%) for urologists.

However, the authors found that these gains in clarity do not translate into measurable improvements in report completeness or clinical decision-making, with both formats performing similarly in supporting diagnosis and treatment decisions. Additionally, clinician trust in structured reports was slightly lower, likely reflecting missing diagnostic elements and the retrospective generation of SR templates.

Overall, the authors concluded that while structured reporting enhances clarity, consistency, and accessibility, it does not yet outperform free-text reporting in terms of clinical impact. They emphasize that optimizing SR requires interdisciplinary collaboration and clinically tailored templates, as well as prospective validation in real-world workflows. This work highlights that improving radiology reporting is not solely about standardization, but about aligning structure with clinical relevance and completeness. 

BMC Ecology and Evolution - Inter- and intraspecific morphometric divergences in a Seychelles endemic gecko genus

This study tackles a classic challenge in biodiversity science: how to distinguish species when physical differences are subtle or 'cryptic'. Focusing on Ailuronyx geckos from the Seychelles, the authors found that combining advanced 3D skull analysis with traditional body measurements can reveal clear patterns of hidden diversity.

The authors identified strong morphological differences between species, including striking contrasts in skull structure; most notably the heavily sculptured skull of A. trachygaster compared to the smooth skulls of related species. These differences extend to body traits such as head size, toe length, and mass, helping to refine species identification in a group where misclassification has been common.

However, the authors note that variation within species does not follow simple geographic patterns across the archipelago. Instead, most divergence occurs at the island or population level, suggesting that local ecological conditions, not broad regional separation, drive morphological evolution.

Overall, the study highlights how island systems generate fine-scale diversity and underscores the need for integrative approaches combining morphology and genetics. The authors argue that recognising these subtle distinctions is essential for effective conservation of island-endemic species.

BMC Genomics - Menstrual cycle-dependent expression of tRNA-derived fragments in healthy human endometrium and uterine fluid-derived extracellular vesicles

This study explores an emerging class of small regulatory RNAs, tRNA‑derived fragments (tRFs), and their role in the human endometrium across the menstrual cycle. The authors found that tRFs are not only present but show dynamic, phase‑specific expression patterns, varying between the proliferative and secretory phases of the cycle.

The authors highlighted that certain tRFs, particularly 5′ fragments derived from tRNA‑Val, are strongly upregulated during the early and mid‑secretory phases, the critical window for embryo implantation. This coordinated increase is observed both in endometrial tissue and in extracellular vesicles from uterine fluid, suggesting selective production and export of these molecules.

Importantly, the authors discovered evidence that extracellular vesicles mirror the tissue-level RNA profile but display selective enrichment of specific tRFs, supporting a role in intercellular communication within the reproductive tract.

Overall, the study highlights tRFs as previously underappreciated regulators of endometrial biology. The authors suggest these molecules may contribute to embryo–maternal signalling and endometrial receptivity, and could potentially serve as novel non-invasive biomarkers of fertility and implantation success. 

BMC Geriatrics - Lessons learned from Canadian family physicians deprescribing medications in older adults – a five-year retrospective review of medico-legal cases

Polypharmacy remains a growing challenge in ageing populations, and this study takes a grounded look at how deprescribing actually unfolds in primary care. The authors found that, for Canadian family physicians, deprescribing is rarely a simple, guideline-driven decision. Instead, it is shaped by real-world complexities; balancing patient expectations, clinical uncertainty, and limited consultation time.

They highlight that physicians often feel caught between the risks of continuing medications and the potential harms or resistance associated with stopping them. Communication emerged as a central theme: the authors found that successful deprescribing depends heavily on trust and shared decision-making with patients and caregivers. At the same time, systemic barriers, such as fragmented care and lack of clear deprescribing frameworks, can slow progress.

Encouragingly, the authors also found opportunities for improvement. Better clinical guidance, enhanced interdisciplinary collaboration, and stronger system-level support could help clinicians feel more confident when initiating deprescribing conversations. Overall, the study reframes deprescribing not as a single action, but as an ongoing, patient-centred process embedded within everyday clinical practice.

BMC Infectious Diseases - Decision-support system for live detection of Leishmania parasites from microscopic images with deep learning

Diagnosing parasitic diseases like leishmaniasis remains a major challenge, especially in low-resource settings where specialist expertise is limited. In this study, the authors found that a deep learning–based approach could significantly improve how these infections are detected in practice. By training a YOLOv8 model on microscopic images of tissue samples, they demonstrated that automated detection of Leishmania parasites is not only feasible, but highly accurate.

Importantly, the authors found that combining images from both high-end laboratory microscopes and low-cost mobile devices enhanced the model’s performance, suggesting strong potential for real-world deployment. The system achieved around 91% accuracy, alongside high sensitivity and specificity, even when tested on previously unseen data.

Beyond the numbers, the study highlights a practical shift: the authors found that integrating this model into a user-friendly interface enables real-time, point-of-care decision support. This could empower non-experts to identify infections more reliably, particularly in endemic regions where diagnostic resources are scarce. Overall, the work points toward a future where AI-driven tools help bridge gaps in global infectious disease diagnostics.