Predicting sequenced dental treatment plans from electronic dental records using deep learning.

BACKGROUND: Designing appropriate clinical dental treatment plans is an urgent need because a growing number of dental patients are suffering from partial edentulism with the population getting older. OBJECTIVES: The aim of this study is to predict sequential treatment plans from electronic dental records. METHODS: We construct a clinical decision support model, MultiTP, explores the unique topology of teeth information and the variation of complicated treatments, integrates deep learning models (convolutional neural network and recurrent neural network) adaptively, and embeds the attention mechanism to produce optimal treatment plans. RESULTS: MultiTP shows its promising performance with an AUC of 0.9079 and an F score of 0.8472 over five treatment plans. The interpretability analysis also indicates its capability in mining clinical knowledge from the textual data. CONCLUSIONS: MultiTP's novel problem formulation, neural network framework, and interpretability analysis techniques allow for broad applications of deep learning in dental healthcare, providing valuable support for predicting dental treatment plans in the clinic and benefiting dental patients. CLINICAL IMPLICATIONS: The MultiTP is an efficient tool that can be implemented in clinical practice and integrated into the existing EDR system. By predicting treatment plans for partial edentulism, the model will help dentists improve their clinical decisions.

Polyunsaturated fatty acids drive neutrophil extracellular trap formation in nonalcoholic steatohepatitis.

Non-alcoholic steatohepatitis (NASH) is the hepatic manifestation of metabolic syndrome. Non-resolving inflammation, triggered by sustained accumulation of lipids, is an important driving force of NASH. Thus, unveiling metabolic immune regulation could help better understand the pathology and intervention of NASH. In this study, we found the recruitment of neutrophils is an early inflammatory event in NASH mice, following the formation of neutrophil extracellular traps (NETs). NET is an initiating factor which exacerbates inflammatory responses in macrophages. Inhibition of NETs using DNase I significantly alleviated inflammation in NASH mice. We further carried out a metabolomic study to identify possible metabolic triggers of NETs, and linoleic acid (LA) metabolic pathway was the most altered pathway. We re-analyzed published clinical data and validated that LA metabolism was highly correlated with NASH. Consistently, both LA and γ-linolenic acid (GLA) were active in triggering NETs formation by oxidative burst. Furthermore, we identified silybin, a hepatoprotective agent, as a potent NETosis inhibitor, which effectively blocked NETs formation both in vitro and in vivo. Together, this study not only provide new insights into metabolism-immune causal link in NASH progression, but also demonstrate silybin as an important inhibitor of NETs and its therapeutical potential in treating NETosis-related diseases.

Curvature-Dependent Binding of Cytochrome c to Cardiolipin.

Cytochrome c binds cardiolipin on the concave surface of the inner mitochondrial membrane, before oxidizing the lipid and initiating the apoptotic pathway. This interaction has been studied in vitro, where mimicking the membrane curvature of the binding environment is difficult. Here we report binding to concave, cardiolipin-containing, membrane surfaces and compare findings to convex binding under the same conditions. For binding to the convex outer surface of cardiolipin-containing vesicles, a two-step structural rearrangement is observed with a small rearrangement detectable by Soret circular dichroism (CD) occurring at an exposed lipid-to-protein ratio (LPR) near 10 and partial unfolding detectable by Trp59 fluorescence occurring at an exposed LPR near 23. On the concave inner surface of cardiolipin-containing vesicles, the structural transitions monitored by Soret CD and Trp59 fluorescence are coincident and occur at an exposed LPR near 58. On the concave inner surface of mitochondrial cristae, we estimate the LPR of cardiolipin to cytochrome c is between 50 and 100. Thus, cytochrome c may have adapted to its native environment so that it can undergo a conformational change that switches on its peroxidase activity when it binds to CL-containing membranes in the cristae early in apoptosis. Our results show that membrane curvature qualitatively affects peripheral protein-lipid interactions and also highlights the disparity between in vitro binding studies and their physiological counterparts where cone-shaped lipids, like cardiolipin, are involved.