SRAM-Based CIM Architecture Design for Event Detection.

Convolutional neural networks (CNNs) play a key role in deep learning applications. However, the high computational complexity and high-energy consumption of CNNs trammel their application in hardware accelerators. Computing-in-memory (CIM) is the technique of running calculations entirely in memory (in our design, we use SRAM). CIM architecture has demonstrated great potential to effectively compute large-scale matrix-vector multiplication. CIM-based architecture for event detection is designed to trigger the next stage of precision inference. To implement an SRAM-based CIM accelerator, a software and hardware co-design approach must consider the CIM macro's hardware limitations to map the weight onto the AI edge devices. In this paper, we designed a hierarchical AI architecture to optimize the end-to-end system power in the AIoT application. In the experiment, the CIM-aware algorithm with 4-bit activation and 8-bit weight is examined on hand gesture and CIFAR-10 datasets, and determined to have 99.70% and 70.58% accuracy, respectively. A profiling tool to analyze the proposed design is also developed to measure how efficient our architecture design is. The proposed design system utilizes the operating frequency of 100 MHz, hand gesture and CIFAR-10 as the datasets, and nine CNNs and one FC layer as its network, resulting in a frame rate of 662 FPS, 37.6% processing unit utilization, and a power consumption of 0.853 mW.

Application of Real-Time Augmented Reality Laparoscopic Navigation in Splenectomy for Massive Splenomegaly.

OBJECTIVES: To evaluate the clinical impact and technical feasibility of augmented reality laparoscopic navigation (ARLN) system in laparoscopic splenectomy for massive splenomegaly. METHODS: The clinical data of 17 consecutive patients who underwent laparoscopic splenectomy using ARLN (ARLN group) and 26 patients without ARLN guidance (Non-ARLN group) between January 2018 and April 2020 were enrolled. Propensity score matching (PSM) analysis was performed between the patients with and without ARLN guidance at a ratio of 1:1. RESULTS: Mean intraoperative blood loss was significantly lower in the ARLN-group than in the Non-ARLN group (306.6 ml vs. 462.6 ml, p = 0.047). All the patients in the ARLN-group achieved successful splenic artery dissection, while surgical success was achieved in 12 patients in the Non-ARLN group (p = 0.044). Postoperative hospital stay was significantly longer in the Non-ARLN group (3.8 days vs. 4.5 days, p = 0.040). CONCLUSIONS: ARLN can provide feasible and accurate intraoperative image guidance, and it could be helpful in the performance of laparoscopic splenectomy for massive splenomegaly.

A novel method of fluorescent imaging can guide hepatectomy for intrahepatic cholangiocarcinoma with intrahepatic biliary obstruction.

OBJECTIVE: The aim of this study was to present a novel bile-duct obstructed area imaging (BOAI) and to investigate the feasibility and accuracy of this method in guiding hepatectomy for intrahepatic cholangiocarcinoma (ICC) with intrahepatic biliary obstruction. METHODS: From May 2017 to October 2019, eligible patients who underwent hepatectomy guided by BOAI were enrolled. Perioperative outcomes and operative data were analyzed. To assess the feasibility of BOAI and Glissonean pedicle approach, demarcations based on them were compared. To verify the accuracy of BOAI staining of the target territory, simple linear regression analysis, and intraclass correlation coefficient were used to examine the relationship between predicted resected liver volume (PRLV) and actual resected liver volume (ARLV). RESULTS: BOAI staining achieved valid demarcation in 15 (93.8%) of 16 patients, whereas the ischemic line achieved valid demarcation in only nine patients (57.3%; p = .017). ARLV and PRLV had a strong positive correlation (PRLV = 60.06 + 0.925 × ARLV; R = .945; p = .000). Meanwhile, ARLV (intraclass correlation coefficient = .971) achieved an excellent agreement with PRLV (p < .001). CONCLUSIONS: The novel BOAI staining method can provide valid, feasible, and accurate demarcation line and may be an effective method in the surgical treatment of intrahepatic biliary obstruction.

Reduction pathway and mechanism of chloronitrobenzenes synergistically catalyzed by bioPd and Shewanella oneidensis MR-1 assisted by calculation.

Although microbial synthesized palladium nanoparticles (bioPd) have been demonstrated to exhibit a great activity toward dechlorination of several chlorinated pollutants, there is no systematic investigation into the substituent effect on dechlorination. Chloronitrobenzenes are widely used for manufacturing and known as persistent pollutants with recalcitrance of biodegradation for nitro groups. In this work, bioPd was synthesized by Shewanella oneidensis MR-1. The dechlorination of 2-chloronitrobenzene, 4-chloronitrobenzene and 2,4-dichloronitrobenzene catalyzed by bioPd were investigated. Simultaneous dechlorination and nitro reduction were observed by synergistic catalysis of bioPd and S. oneidensis MR-1. Pd concentration was optimized for the reduction. Producing profiles of intermediates changed with the ratio of Pd to cell, supporting a size- or shape-controlled catalytic activity of bioPd. The removal of chloro atoms at para-position was easier than that at ortho-position in 2,4-DCNB, suggesting a position effect on the reduction, which was further supported by the frontier molecular orbital and frontier electron density of 2,4-DCNB according to density functional theory.

The Arg16Gly polymorphism of human beta2-adrenoreceptor is associated with type 2 diabetes in Taiwanese people.

OBJECTIVE: The significance of the association of amino terminal polymorphisms in beta2-adrenoreceptor (ADRB2) with obesity and type 2 diabetes is controversial and differs among ethnic groups. In this study, the association of ADRB2 with risk and age of onset of type 2 diabetes has been examined in a Taiwanese population. DESIGN: The study design is a case-control study to investigate the impact of the two amino acid polymorphisms in ADRB2. PATIENTS AND MEASUREMENTS: This study includes 130 patients with type 2 diabetes [female : male = 1 : 1, age: 52.4 +/- 10.0 years; body mass index (BMI): 24.2 +/- 2.9 kg/m2; mean +/- SD] and 130 controlled subjects matched for gender, age and BMI with normal glucose tolerance (female : male = 1 : 1, age: 51.7 +/- 10.6 years; BMI: 23.9 +/- 2.7 kg/m2). The Arg16Gly and Gln27Glu polymorphisms of ADRB2 were determined by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) assays. The genotypic and allelic frequencies between two groups were compared and the relationship between the genotypes and clinical phenotypes was examined. RESULTS: A difference in genotypic frequency in the Arg16Gly polymorphism was noted between groups in this gender-, age- and BMI-matched case-control study (P = 0.039). Multivariate regression analysis revealed that the Arg16Gly polymorphism was the only independent factor for development of type 2 diabetes (P = 0.021). In addition, we utilized the log-rank test to compare the differences in age of onset between wild-type and nonwild-type polymorphisms. The Arg16Gly polymorphism was independently associated with age of onset in type 2 diabetes (P = 0.017). There was no difference in the Gln27Glu polymorphism between diabetic and control groups in this study. CONCLUSIONS: In a Taiwanese population, homozygosity of Arg16 in the ADRB2 gene was associated with a higher frequency (odds ratio 1.87, 95% confidence interval 1.34-2.40) for development of type 2 diabetes. Moreover, this polymorphism was also associated with an earlier onset of type 2 diabetes. However, the Glu27Gln polymorphism had no impact on either BMI or type 2 diabetes in a Taiwanese population.