Improving performance of deep learning models using 3.5D U-Net via majority voting for tooth segmentation on cone beam computed tomography.

Deep learning allows automatic segmentation of teeth on cone beam computed tomography (CBCT). However, the segmentation performance of deep learning varies among different training strategies. Our aim was to propose a 3.5D U-Net to improve the performance of the U-Net in segmenting teeth on CBCT. This study retrospectively enrolled 24 patients who received CBCT. Five U-Nets, including 2Da U-Net, 2Dc U-Net, 2Ds U-Net, 2.5Da U-Net, 3D U-Net, were trained to segment the teeth. Four additional U-Nets, including 2.5Dv U-Net, 3.5Dv5 U-Net, 3.5Dv4 U-Net, and 3.5Dv3 U-Net, were obtained using majority voting. Mathematical morphology operations including erosion and dilation (E&D) were applied to remove diminutive noise speckles. Segmentation performance was evaluated by fourfold cross validation using Dice similarity coefficient (DSC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV). Kruskal-Wallis test with post hoc analysis using Bonferroni correction was used for group comparison. P < 0.05 was considered statistically significant. Performance of U-Nets significantly varies among different training strategies for teeth segmentation on CBCT (P < 0.05). The 3.5Dv5 U-Net and 2.5Dv U-Net showed DSC and PPV significantly higher than any of five originally trained U-Nets (all P < 0.05). E&D significantly improved the DSC, accuracy, specificity, and PPV (all P < 0.005). The 3.5Dv5 U-Net achieved highest DSC and accuracy among all U-Nets. The segmentation performance of the U-Net can be improved by majority voting and E&D. Overall speaking, the 3.5Dv5 U-Net achieved the best segmentation performance among all U-Nets.

Discovery, synthesis, biological evaluation and structure-based optimization of novel piperidine derivatives as acetylcholine-binding protein ligands.

The homomeric α7 nicotinic receptor (α7 nAChR) is widely expressed in the human brain that could be activated to suppress neuroinflammation, oxidative stress and neuropathic pain. Consequently, a number of α7 nAChR agonists have entered clinical trials as anti-Alzheimer's or anti-psychotic therapies. However, high-resolution crystal structure of the full-length α7 receptor is thus far unavailable. Since acetylcholine-binding protein (AChBP) from Lymnaea stagnalis is most closely related to the α-subunit of nAChRs, it has been used as a template for the N-terminal domain of α-subunit of nAChR to study the molecular recognition process of nAChR-ligand interactions, and to identify ligands with potential nAChR-like activities.Here we report the discovery and optimization of novel acetylcholine-binding protein ligands through screening, structure-activity relationships and structure-based design. We manually screened in-house CNS-biased compound library in vitro and identified compound 1, a piperidine derivative, as an initial hit with moderate binding affinity against AChBP (17.2% inhibition at 100 nmol/L). During the 1st round of optimization, with compound 2 (21.5% inhibition at 100 nmol/L) as the starting point, 13 piperidine derivatives with different aryl substitutions were synthesized and assayed in vitro. No apparent correlation was demonstrated between the binding affinities and the steric or electrostatic effects of aryl substitutions for most compounds, but compound 14 showed a higher affinity (Ki=105.6 nmol/L) than nicotine (Ki=777 nmol/L). During the 2nd round of optimization, we performed molecular modeling of the putative complex of compound 14 with AChBP, and compared it with the epibatidine-AChBP complex. The results suggested that a different piperidinyl substitution might confer a better fit for epibatidine as the reference compound. Thus, compound 15 was designed and identified as a highly affinitive acetylcholine-binding protein ligand. In this study, through two rounds of optimization, compound 15 (Ki=2.8 nmol/L) has been identified as a novel, piperidine-based acetylcholine-binding protein ligand with a high affinity.