Purpose: The purpose of this study is to investigate the effect of different magnetic resonance (MR) sequences on the accuracy of deep learning-based synthetic computed tomography (sCT) generation in the complex head and neck region. Methods: Four sequences of MR images (T1, T2, T1C, and T1DixonC-water) were collected from 45 patients with nasopharyngeal carcinoma. Seven conditional generative adversarial network (cGAN) models were trained with different sequences (single channel) and different combinations (multi-channel) as inputs. To further verify the cGAN performance, we also used a U-net network as a comparison. Mean absolute error, structural similarity index, peak signal-to-noise ratio, dice similarity coefficient, and dose distribution were evaluated between the actual CTs and sCTs generated from different models. Results: The results show that the cGAN model with multi-channel (i.e., T1 + T2 + T1C + T1DixonC-water) as input to predict sCT achieves higher accuracy than any single MR sequence model. The T1-weighted MR model achieves better results than T2, T1C, and T1DixonC-water models. The comparison between cGAN and U-net shows that the sCTs predicted by cGAN retains additional image details are less blurred and more similar to the actual CT. Conclusions: Conditional generative adversarial network with multiple MR sequences as model input shows the best accuracy. The T1-weighted MR images provide sufficient image information and are suitable for sCT prediction in clinical scenarios with limited acquisition sequences or limited acquisition time.

No abstract

No abstract

Accurate and automatic segmentation of individual tooth is critical for computer-aided analysis towards clinical decision support and treatment planning. Three-dimensional reconstruction of individual tooth after the segmentation also plays an important role in simulation in digital orthodontics. However, it is difficult to automatically segment individual tooth in cone beam computed tomography (CBCT) images due to the blurring boundaries of neighboring teeth and the similar intensities between teeth and mandible bone. In this work, we propose the use of a multi-task 3D fully convolutional network (FCN) and marker-controlled watershed transform (MWT) to segment individual tooth. The multi-task FCN learns to simultaneously predict the probability of tooth region and the probability of tooth surface. Through the combination of the tooth probability gradient map and the surface probability map as the input image, MWT is used to automatically separate and segment individual tooth. Twenty-five dental CBCT scans are used in the study. The average Dice similarity coefficient, Jaccard index, and relative volume difference are 0.936 (±0.012), 0.881 (±0.019), and 0.072 (±0.027), respectively, and the average symmetric surface distance is 0.363 (±0.145) mm for our method. The experimental results demonstrate that the multi-task 3D FCN combined with MWT can segment individual tooth of various types in dental CBCT images.INDEX TERMS Individual tooth segmentation, dental CBCT, deep learning, marker-controlled watershed transform.

BackgroundUrsolic acid (UA), a natural pentacyclic triterpenoid, exerts anti-tumor effects in various cancer types including hepatocellular carcinoma (HCC). However, the molecular mechanisms underlying this remain largely unknown.MethodsCell viability and cell cycle were examined by MTT and Flow cytometry assays. Western blot analysis was performed to measure the phosphorylation and protein expression of p38 MAPK, insulin-like growth factor (IGF) binding protein 1 (IGFBP1) and forkhead box O3A (FOXO3a). Quantitative real-time PCR (qRT-PCR) was used to examine the mRNA levels of IGFBP1 gene. Small interfering RNAs (siRNAs) method was used to knockdown IGFBP1 gene. Exogenous expressions of IGFBP1 and FOXO3a were carried out by transient transfection assays. IGFBP1 promoter activity was measured by Secrete-Pair™ Dual Luminescence Assay Kit. In vivo nude mice xenograft model and bioluminescent imaging system were used to confirm the findings in vitro.ResultsWe showed that UA stimulated phosphorylation of p38 MAPK. In addition, UA increased the protein, mRNA levels, and promoter activity of IGFBP1, which was abrogated by the specific inhibitor of p38 MAPK (SB203580). Intriguingly, we showed that UA increased the expression of FOXO3a and that overexpressed FOXO3a enhanced phosphorylation of p38 MAPK, all of which were not observed in cells silencing of endogenous IGFBP1 gene. Moreover, exogenous expressed IGFBP1 strengthened UA-induced phosphorylation of p38 MAPK and FOXO3a protein expression, and more importantly, restored the effect of UA-inhibited growth in cells silencing of endogenous IGFBP1 gene. Consistent with these, UA suppressed tumor growth and increased phosphorylation of p38 MAPK, protein expressions of IGFBP1 and FOXO3a in vivo.ConclusionCollectively, our results show that UA inhibits growth of HCC cells through p38 MAPK-mediated induction of IGFBP1 and FOXO3a expression. The interactions between IGFBP1 and FOXO3a, and feedback regulatory loop of p38 MAPK by IGFBP1 and FOXO3a resulting in reciprocal pathways, contribute to the overall effects of UA. This in vitro and in vivo study corroborates a potential novel mechanism by which UA controls HCC growth and implies that the rational targeting IGFBP1 and FOXO3a can be potential for the therapeutic strategy against HCC.