Improved deep learning for automatic localisation and segmentation of rectal cancer on T2-weighted MRI.

INTRODUCTION: The automatic segmentation approaches of rectal cancer from magnetic resonance imaging (MRI) are very valuable to relieve physicians from heavy workloads and enhance working efficiency. This study aimed to compare the segmentation accuracy of a proposed model with the other three models and the inter-observer consistency. METHODS: A total of 65 patients with rectal cancer who underwent MRI examination were enrolled in our cohort and were randomly divided into a training cohort (n = 45) and a validation cohort (n = 20). Two experienced radiologists independently segmented rectal cancer lesions. A novel segmentation model (AttSEResUNet) was trained on T2WI based on ResUNet and attention mechanisms. The segmentation performance of the AttSEResUNet, U-Net, ResUNet and U-Net with Attention Gate (AttUNet) was compared, using Dice similarity coefficient (DSC), Hausdorff distance (HD), mean distance to agreement (MDA) and Jaccard index. The segmentation variability of automatic segmentation models and inter-observer was also evaluated. RESULTS: The AttSEResUNet with post-processing showed perfect lesion recognition rate (100%) and false recognition rate (0), and its evaluation metrics outperformed other three models for two independent readers (observer 1: DSC = 0.839 ± 0.112, HD = 9.55 ± 6.68, MDA = 0.556 ± 0.722, Jaccard index = 0.736 ± 0.150; observer 2: DSC = 0.856 ± 0.099, HD = 11.0 ± 10.1, MDA = 0.789 ± 1.07, Jaccard index = 0.673 ± 0.130). The segmentation performance of AttSEResUNet was comparable and similar to manual variability (DSC = 0.857 ± 0.115, HD = 10.0 ± 10.0, MDA = 0.704 ± 1.17, Jaccard index = 0.666 ± 0.139). CONCLUSION: Comparing with other three models, the proposed AttSEResUNet model was demonstrated as a more accurate model for contouring the rectal tumours in axial T2WI images, whose variability was similar to that of inter-observer.

Porcine circovirus type 2 capsid protein induces unfolded protein response with subsequent activation of apoptosis.

Porcine circovirus type 2 (PCV2) has recently been reported to elicit the unfolded protein response (UPR) via activation of the PERK/eIF2α (RNA-activated protein kinase-like endoplasmic reticulum (ER) kinase/eukaryotic initiation factor 2α) pathway. This study attempted to examine which viral protein might be involved in inducing UPR and whether this cellular event would lead to apoptosis of the cells expressing the viral protein. By transient expression, we found that both replicase (Rep) and capsid (Cap) proteins of PCV2 could induce ER stress as shown by increased phosphorylation of PERK with subsequent activation of the eIF2α-ATF4 (activating transcription factor 4)-CHOP (CCAAT/enhancer-binding protein homologous protein) axis. Cap expression, but not Rep, significantly reduced anti-apoptotic B-cell lymphoma-2 (Bcl-2) and increased caspase-3 cleavage, possibly due to increased expression of CHOP. Since knockdown of PERK by RNA interference clearly reduced Cap-induced CHOP expression, caspase-3 cleavage, and apoptotic cell death possibly by partially rescuing Bcl-2 expression, we propose that there is connection between Cap-induced UPR and apoptosis via the PERK/eIF2α/ATF4/CHOP/Bcl-2 pathway. This study, together with our earlier studies, provides insight into the mechanisms underlying PCV2 pathogenesis.

Porcine Circovirus Type 2 Activates CaMMKß to Initiate Autophagy in PK-15 Cells by Increasing Cytosolic Calcium.

Porcine circovirus type 2 (PCV2) induces autophagy via the 5' adenosine monophosphate-activated protein kinase (AMPK)/extracellular signal-regulated kinase (ERK)/tuberous sclerosis complex 2 (TSC2)/mammalian target of rapamycin (mTOR) pathway in pig kidney PK-15 cells. However, the underlying mechanisms of AMPK activation in autophagy induction remain unknown. With specific inhibitors and RNA interference (RNAi), we show that PCV2 infection upregulated calcium/calmodulin-dependent protein kinase kinase-beta (CaMKKß) by increasing cytosolic Ca(2+) via inositol 1,4,5-trisphosphate receptor (IP3R). Elevation of cytosolic calcium ion (Ca(2+)) did not seem to involve inositol 1,4,5-trisphosphate (IP3) release from phosphatidylinositol 4,5-bisphosphate (PIP2) by phosphoinositide phospholipase C-gamma (PLC-γ). CaMKKß then activated both AMPK and calcium/calmodulin-dependent protein kinase I (CaMKI). PCV2 employed CaMKI and Trp-Asp (WD) repeat domain phosphoinositide-interacting protein 1 (WIPI1) as another pathway additional to AMPK signaling in autophagy initiation. Our findings could help better understanding of the signaling pathways of autophagy induction as part of PCV2 pathogenesis. Further research is warranted to study if PCV2 interacts directly with IP3R or indirectly with the molecules that antagonize IP3R activity responsible for increased cytosolic Ca(2+) both in PK-15 cells and PCV2-targeted primary cells from pigs.