No abstract

Background: Vitamin D deficiency is linked to adverse childhood health outcomes, yet data on the distribution and quantifiable determinants of neonatal 25-hydroxyvitamin D₃ (25OHD) concentration, a vitamin D biomarker, are limited. Objective: Our aim was to identify determinants of neonatal 25OHD concentration, measured using neonatal dried blood spots (DBS). Methods: A total of 259 ethnically diverse children aged 0-16 years born in Victoria, Australia, were recruited. Data included maternal sun exposure, skin type, 25OHD concentration on stored neonatal DBS, and genotypes at the target genes. Associations were investigated using multiple linear regression models. Results: The median 25OHD concentration was 29.2 nmol/l (IQR 18.0-47.4). Measured 25OHD was <50 nmol/l in almost half of the neonatal sample. Ambient ultraviolet radiation (UVR) 6 weeks before birth was the strongest predictor of neonatal 25OHD, accounting for 23% of its variation. A further 10% was explained by infant genetic variants at GC (rs2282679), the gene encoding the vitamin D binding protein, and DHCR7 (rs12785878), a gene required for synthesis of 7-dehydrocholesterol, a precursor to 25OHD. DBS age explained 7%, and patterns of maternal sun exposure and clothing choices accounted for 4%. A child's skin colour was strongly associated with GC gene variants and not independent of these variants in predicting 25OHD. The final model explained 43% of the total variance in neonatal 25OHD concentration. Conclusion: Maternal lifestyle factors and infant genetic variants predict neonatal 25OHD levels; the importance of maternal UVR exposure in late pregnancy is highlighted.

Weakly-supervised classification of histopathology slides is a computationally intensive task, with a typical whole slide image (WSI) containing billions of pixels to process. We propose Discriminative Region Active Sampling for Multiple Instance Learning (DRAS-MIL), a computationally efficient slide classification method using attention scores to focus sampling on highly discriminative regions. We apply this to the diagnosis of ovarian cancer histological subtypes, which is an essential part of the patient care pathway as different subtypes have different genetic and molecular profiles, treatment options, and patient outcomes. We use a dataset of 714 WSIs acquired from 147 epithelial ovarian cancer patients at Leeds Teaching Hospitals NHS Trust to distinguish the most common subtype, high-grade serous carcinoma, from the other four subtypes (low-grade serous, endometrioid, clear cell, and mucinous carcinomas) combined. We demonstrate that DRAS-MIL can achieve similar classification performance to exhaustive slide analysis, with a 3-fold cross-validated AUC of 0.8679 compared to 0.8781 with standard attention-based MIL classification. Our approach uses at most 18% as much memory as the standard approach, while taking 33% of the time when evaluating on a GPU and only 14% on a CPU alone. Reducing prediction time and memory requirements may benefit clinical deployment and the democratisation of AI, reducing the extent to which computational hardware limits end-user adoption.

The use of human excreta in agricultural settings has the potential to meet crop nutrient requirements and improve soil health, while also providing a sustainable end use for fecal material. Previous reviews have focused on sewage sludge biosolids from wastewater treatment plants, but with on-site sanitation systems overtaking sewers as the leading sanitation system worldwide, greater attention to fecal sludge is warranted. This systematic Review is the first to compile the results of crop trials which utilized fecal amendments from on-site sanitation systems and includes 47 experiments. Overall, fecal amendments increased crop growth compared to unamended control plots and also produced comparable yields to synthetic fertilizers. Biological and physical soil parameters were underrepresented in the literature, which made a holistic assessment of soil health impossible. However, some improvements in chemical characteristics were observed, most notably for soil organic carbon. Inconsistent experimental design made aggregation of results and detailed statistical analysis difficult, highlighting the need for a more standardized approach for testing the efficacy of amendments and reporting results. Regardless, this Review compiles our collective existing knowledge to provide tentative results for the effect of fecal amendments on crop growth and soil health and offers recommendations for future work.