Improvement of intervention information detection for automated clinical literature screening during systematic review.

Systematic literature review (SLR) is a crucial method for clinicians and policymakers to make their decisions in a flood of new clinical studies. Because manual literature screening in SLR is a highly laborious task, its automation by natural language processing (NLP) has been welcomed. Although intervention is a key information for literature screening, NLP models for its detection in previous works have not shown adequate performance. In this work, we first design an algorithm for automated construction of high-quality intervention labels by utilizing information retrieved from a clinical trial database. We then design another algorithm for improving model's recall and F1 score by imposing adaptive weights on training instances in the loss function. The intervention detection model trained on the weighted datasets is tested with the Evidence-Based Medicine NLP (EBM-NLP) corpus, and shows 9.7% and 4.0% improvements respectively in recall and F1 score compared to the previous state-of-the-art model on the corpus. The proposed algorithms can boost automation of literature screening during SLR in the clinical domain.

Transforming growth factor-ß2 is sequestered in preterm human milk by chondroitin sulfate proteoglycans.

Human milk contains biologically important amounts of transforming growth factor-ß2 isoform (TGF-ß2), which is presumed to protect against inflammatory gut mucosal injury in the neonate. In preclinical models, enterally administered TGF-ß2 can protect against experimental necrotizing enterocolitis, an inflammatory bowel necrosis of premature infants. In this study, we investigated whether TGF-ß bioactivity in human preterm milk could be enhanced for therapeutic purposes by adding recombinant TGF-ß2 (rTGF-ß2) to milk prior to feeding. Milk-borne TGF-ß bioactivity was measured by established luciferase reporter assays. Molecular interactions of TGF-ß2 were investigated by nondenaturing gel electrophoresis and immunoblots, computational molecular modeling, and affinity capillary electrophoresis. Addition of rTGF-ß2 (20-40 nM) to human preterm milk samples failed to increase TGF-ß bioactivity in milk. Milk-borne TGF-ß2 was bound to chondroitin sulfate (CS) containing proteoglycan(s) such as biglycan, which are expressed in high concentrations in milk. Chondroitinase treatment of milk increased the bioactivity of both endogenous and rTGF-ß2, and consequently, enhanced the ability of preterm milk to suppress LPS-induced NF-κB activation in macrophages. These findings provide a mechanism for the normally low bioavailability of milk-borne TGF-ß2 and identify chondroitinase digestion of milk as a potential therapeutic strategy to enhance the anti-inflammatory effects of preterm milk.