Automated Magnetic Resonance Image Segmentation of Spinal Structures at the L4-5 Level with Deep Learning: 3D Reconstruction of Lumbar Intervertebral Foramen.

OBJECTIVE: 3D reconstruction of lumbar intervertebral foramen (LIVF) has been beneficial in evaluating surgical trajectory. Still, the current methods of reconstructing the 3D LIVF model are mainly based on manual segmentation, which is laborious and time-consuming. This study aims to explore the feasibility of automatically segmenting lumbar spinal structures and increasing the speed and accuracy of 3D lumbar intervertebral foramen (LIVF) reconstruction on magnetic resonance image (MRI) at the L4-5 level. METHODS: A total of 100 participants (mean age: 42.2 ± 14.0 years; 52 males and 48 females; mean body mass index, 22.7 ± 3.2 kg/m2 ), were enrolled in this prospective study between March and July 2020. All participants were scanned on L4-5 level with a 3T MR unit using 3D T2-weighted sampling perfection with application-optimized contrast with various flip-angle evolutions (SPACE) sequences. The lumbar spine's vertebra bone structures (VBS) and intervertebral discs (IVD) were manually segmented by skilled surgeons according to their anatomical outlines from MRI. Then all manual segmentation were saved and used for training. An automated segmentation method based on a 3D U-shaped architecture network (3D-UNet) was introduced for the automated segmentation of lumbar spinal structures. A number of quantitative metrics, including dice similarity coefficient (DSC), precision, and recall, were used to evaluate the performance of the automated segmentation method on MRI. Wilcoxon signed-rank test was applied to compare morphometric parameters, including foraminal area, height and width of 3D LIVF models between automatic and manual segmentation. The intra-class correlation coefficient was used to assess the test-retest reliability and inter-observer reliability of multiple measurements for these morphometric parameters of 3D LIVF models. RESULTS: The automatic segmentation performance of all spinal structures (VBS and IVD) was found to be 0.918 (healthy levels: 0.922; unhealthy levels: 0.916) for the mean DSC, 0.922 (healthy levels: 0.927; unhealthy levels: 0.920) for the mean precision, and 0.917 (healthy levels: 0.918; unhealthy levels: 0.917) for the mean recall in the test dataset. It took approximately 2.5 s to achieve each automated segmentation, far less than the 240 min for manual segmentation. Furthermore, no significant differences were observed in the foraminal area, height and width of the 3D LIVF models between manual and automatic segmentation images (P > 0.05). CONCLUSION: A method of automated MRI segmentation based on deep learning algorithms was capable of rapidly generating accurate segmentation of spinal structures and can be used to construct 3D LIVF models from MRI at the L4-5 level.

Identification of aberrantly methylated differentially expressed genes targeted by differentially expressed miRNA in osteosarcoma.

BACKGROUND: Osteosarcoma (OS) is the most common primary bone tumors diagnosed in children and adolescents. Recent studies have shown a prognostic role of DNA methylation in various cancers, including OS. The aim of this study was to identify the aberrantly methylated genes that are prognostically relevant in OS. METHODS: The differentially expressed mRNAs, miRNAs and methylated genes (DEGs, DEMs and DMGs respectively) were screened from various GEO databases, and the potential target genes of the DEMs were predicted by the RNA22 program. The protein-protein interaction (PPI) networks were constructed using the STRING database and visualized by Cytoscape software. The functional enrichment and survival analyses of the screened genes was performed using the R software. RESULTS: Forty-seven downregulated hypermethylated genes and three upregulated hypomethylated genes were identified that were enriched in cell activation, migration and proliferation functions, and were involved in cancer-related pathways like JAK-STAT and PI3K-AKT. Eight downregulated hypermethylated tumor suppressor genes (TSGs) were identified among the screened genes based on the TSGene database. These hub genes are likely involved in OS genesis, progression and metastasis, and are potential prognostic biomarkers and therapeutic targets. CONCLUSIONS: TSGs including PYCARD, STAT5A, CXCL12 and CXCL14 were aberrantly methylated in OS, and are potential prognostic biomarkers and therapeutic targets. Our findings provide new insights into the role of methylation in OS progression.

NVP-BEZ235 synergizes cisplatin sensitivity in osteosarcoma.

Osteosarcoma(OS) remains a major health concern in childhood and adolescence, although cisplatin is one of the gold standard chemotherapeutic drugs in the treatment of OS, chemoresistant to cisplatin is common. Phosphoinositide 3-kinase (PI3K)-Akt-mammalian target of rapamycin inhibitor (mTOR) pathway and autophagy regulates chemosensitivity incancer cells. In this study, we hypothesized that NVP-BEZ235, a dual inhibitor of PI3K/mTOR, could synergize cisplatin sensitivity in OS. In vitro, NVP-BEZ235 plus cisplatinexerted a synergistic effect on cell proliferation inhibition and apoptosis induction. Cisplatin could activate PI3K-Akt-mTOR pathway activity in early times, whereas, NVP-BEZ235 could inhibit PI3K-Akt -mTOR pathway activity all the times alone or combined with cisplatin. What's more, NVP-BEZ235 could switch function of autophagy induced by cisplatin to synergize cisplatin sensitivity. In vivo, pronounced decrease in tumor cell proliferation and increase in apoptosisin combination-treated mouse xenograft models compared with cisplatin or NVP-BEZ235 treated models. All these results suggest NVP-BEZ235 could synergize cisplatin sensitivity in OS, combination of NVP-BEZ235 with cisplatin could represent a novel therapeutic strategy for treatment of OS.