IVS-Caffe-Hardware-Oriented Neural Network Model Development.

This article proposes a hardware-oriented neural network development tool, called Intelligent Vision System Lab (IVS)-Caffe. IVS-Caffe can simulate the hardware behavior of convolution neural network inference calculation. It can quantize weights, input, and output features of convolutional neural network (CNN) and simulate the behavior of multipliers and accumulators calculation to achieve the bit-accurate result. Furthermore, it can test the accuracy of the chosen CNN hardware accelerator. Besides, this article proposes an algorithm to solve the deviation of gradient backpropagation in the bit-accurate quantized multipliers and accumulators. This allows the training of a bit-accurate model and further increases the accuracy of the CNN model at user-designed bit width. The proposed tool takes Faster region based CNN (R-CNN) + Matthew D. Zeiler and Rob Fergus (ZF)-Net, Single Shot MultiBox Detector (SSD) + VGG, SSD + MobileNet, and Tiny you only look once (YOLO) v2 as the experimental models. These models include both one-stage object detection and two-stage object detection models, and base networks include the convolution layer, the fully connected layer, and the modern advanced layers, such as the inception module and depthwise separable convolution. In these experiments, direct quantization of layer-I/O fixed-point models to bit-accurate models will have a 2% mean average precision (mAP) drop of accuracy in the constraint that all layers' accumulators and multipliers are quantized to less or equal to 14 and 12 bit, respectively. After retraining of these quantized models with the proposed IVS-Caffe, we can achieve less than 1% mAP drop in accuracy in the constraint that all layers' accumulators and multipliers are quantized to less or equal to 14 and 11 bit, respectively. With the proposed IVS-Caffe, we can analyze the accuracy of the target model when it is running at hardware accelerators with different bit widths, which is beneficial to fine-tune the target model or customize the hardware accelerators with lower power consumption. Code is available at https://github.com/apple35932003/IVS-Caffe.

Overexpression of Histone H3 Lysine 27 Trimethylation Is Associated with Aggressiveness and Dedifferentiation of Thyroid Cancer.

A variety of epigenetic dysregulations are observed in thyroid malignancies. EZH2, the catalytic subunit of polycomb repressive complex 2, is upregulated in advanced thyroid cancers. EZH2 can catalyze trimethylation of histone H3 at lysine 27 (H3K27me3) and contribute to transcriptional silencing of target genes. Here, we investigated the immunohistochemical expression of H3K27me3 in neoplastic and normal thyroid tissues. Normal thyroid epithelial cells typically exhibited nuclear staining of moderate intensity. A similar expression pattern was observed in nodular goiters and follicular adenomas. By contrast, strong H3K27me3 expression was evident in 80% (8/10) lymphocytic thyroiditis, 63% (80/127) papillary thyroid cancer, 41% (7/17) follicular thyroid cancer, and 73% (8/11) poorly differentiated and anaplastic thyroid cancer. In differentiated thyroid cancer, strong H3K27me3 expression was associated with extrathyroidal extension (p < 0.001), lymphovascular invasion (p = 0.029), lymph node metastasis (p = 0.006), and higher risk of recurrence (p = 0.003). Our results indicate that H3K27me3 overexpression may be implicated in aggressiveness and dedifferentiation of thyroid cancer. In addition to prognostication, the predictive value of H3K27me3 expression deserves further investigation given the recent development of epigenetic targeting agents.

Dark channel prior based video dehazing algorithm with sky preservation and its embedded system realization for ADAS applications.

Dark Channel Prior (DCP) is one of the significant dehazing methods based upon the observation of the key features of the haze-free images. But it has disadvantages; high computational complexity, over-enhancement in the sky region, flickering artefacts in video processing, and poor dehazing. Therefore, we propose improved solutions to solve the aforementioned drawbacks. First, we adopt the fast one-dimensional filter, look-up table, and program optimization to reduce the computational complexity. Next, we follow by using a part of the guided filter for sky detection and to preserve the sky region from noise by avoiding over recovery. Then, we propose an airlight update strategy and adjust the radius of a guided filter to reduce the flickering artifacts and also propose an airlight estimation method to produce the better dehazing result as the final step of our algorithm. The improved results from our proposed algorithm are stable and are obtained from the real-time processing suitable for ADAS, surveillance, and monitoring systems. The implementation of the proposed algorithm has yielded a processing speed of 75 fps and 23 fps respectively on an NVIDIA Jetson TX1 embedded platform and Renesas R-Car M2, both on D1 (720x480) resolution videos.

FANCA D1359Y mutation in a patient with gastric polyposis and cancer susceptibility: A case report and review of literature.

Gastric polyposis is a rare disease. Not all polyps progress to cancer. Monoallelic mutation in Fanconi anemia (FA) genes, unlike biallelic gene mutations that causes typical FA phenotype, can increase risks of cancers in a sporadic manner. Aberrations in the FA pathway were reported in all molecular subtypes of gastric cancer. We studied a patient with synchronous gastric cancer from gastric polyposis by conducting a 13-year long-term follow up. Via pathway-driven massive parallel genomic sequencing, a germline mutation at FANCA D1359Y was identified. We identified several recurrent mutations in DNA methylation (TET1, V873I), the ß-catenin pathway (CTNNB1, S45F) and RHO signaling pathway (PLEKHG5, R203C) by comparing the genetic events between benign and malignant gastric polyps. Furthermore, we revealed gastric polyposis susceptible genes and genetic events promoting malignant transformation using pathway-driven targeted gene sequencing.

Comparison of topical mupirocin and gentamicin in the prevention of peritoneal dialysis-related infections: A systematic review and meta-analysis.

BACKGROUND: Topical antibiotics have been shown to reduce exit-site infection and peritonitis. The aim of this study was to compare infection rates between mupirocin and gentamicin. METHODS: Multiple comprehensive databases were searched systematically to include relevant randomized controlled trials and observational studies. Pooled risk ratios (RRs) and 95% confidence intervals were calculated for the incidences of exit-site infection and peritonitis. RESULTS: Seven studies (mupirocin group n = 458, gentamicin group n = 448) were analyzed for exit-site infection. The risk of gram-positive exit-site infection was similar between the groups. Gram-negative exit-site infection rate was higher in the mupirocin group (RR = 2.125, P = 0.037). Six studies were assessed the peritonitis risk. There was no difference in the gram-positive and -negative peritonitis rate. CONCLUSIONS: Topical use of gentamicin is associated with fewer exit-site infections caused by gram-negative organisms. Gentamicin has comparable efficacy to mupirocin for peritonitis and gram-positive exit-site infection.

Effects of age and diabetes mellitus on clinical outcomes in patients with peritoneal dialysis-related peritonitis.

BACKGROUND: Peritoneal dialysis (PD)-related peritonitis is the most common complication and the leading cause of technique failure for patients on PD therapy. The aim of this study was to review the episodes and outcomes of PD-related peritonitis, with special reference to the effects of age and diabetes status. METHODS: The study comprised 204 consecutive peritonitis episodes between January 2001 and June 2011 from our registry database. Factors associated with treatment failure and mortality were analyzed with multi-variable logistic regression modeling. RESULTS: The overall peritonitis rate was one episode per 65 patient-months or 0.187 episodes/patient-year. Older patients had lower probability of remaining free of peritonitis (log rank, p<0.001). Elderly patients and diabetic subjects had a higher risk of peritonitis-related dropout. Age (odds ratio [OR], 1.37; p=0.041), diabetes mellitus (OR, 3.64; p=0.005), and gram-negative peritonitis (OR, 3.55; p=0.011) were independent determinants of catheter removal. Age (OR, 1.93; p=0.026) and diabetes (OR, 5.98; p=0.015) were the only predictors of death from peritonitis. CONCLUSIONS: Old age and diabetes mellitus are two major risk factors for adverse outcomes of PD-related peritonitis. Although they are not contraindications to PD, our results suggest that these patients need meticulous care when peritonitis occurs.

Severe acute pancreatitis with abscess after extracorporeal shock wave lithotripsy: a rare complication.

Extracorporeal shock wave lithotripsy (ESWL) is a widely accepted procedure for urolithiasis. However, the shock waves do not pass through the body without damage. Here, we reported a 57-year-old man who underwent ESWL four times before, and immediately developed acute pancreatitis and peritoneal abscess after ESWL for a right renal stone. Although the possibility of post-ESWL acute pancreatitis is extremely low, urologists must be aware of this vital complication.