X-Ray Tomography Reconstruction Accelerated on FPGA Through High-Level Synthesis Tools.

Model-Based Iterative Reconstruction (MBIR) algorithms iteratively use expensive computational operators of forward and backward projections. The irregular memory access pattern of these operators makes them a memory-bound application. Their computation time must be reduced to meet clinical routine constraints. This article proposes a hardware accelerator architecture based on Field Programmable Gate Arrays (FPGAs) through high-level language, as an alternative to GPU architecture. This acceleration is based on an offline memory access analysis to address the main bottleneck of the algorithm and maximize the data reuse rate. The offline analysis allows for the tuning of the architecture parameters so that they converge to an optimal solution. Then, the Berkeley Roofline model guides our optimization steps by iteratively analyzing the design performance. Our design flow significantly improved the algorithm's computational intensity and overcame the memory bottleneck. Thus, our architecture takes advantage of the FPGA local memory to achieve significant memory bandwidth and efficiently harness the pipeline without stalling the computation. Furthermore, we present the scaling-up strategy from mid-range FPGA to high-end FPGA and any concerns of portability. We used two Intel FPGA devices to implement the algorithm, and then we compared the results with our GPU implementation in terms of speedup and energy efficiency. Our experimental results show that our design has achieved better computational throughput than the works on FPGA architectures reported in the literature.

West African female sex workers in Mali: reduction in HIV prevalence and differences in risk profiles of sex workers of differing nationalities of origin.

BACKGROUND: Female sex workers (FSW) in Mali are highly vulnerable to HIV. Their prevalence in 2009 was 9 times higher (24.2%) than that among pregnant women (2.7%). METHODS: Four Integrated HIV/sexually transmitted infection (STI) Surveillance and Behavioral Surveys among FSW in Mali (2000, 2003, 2006, and 2009) tracked demographic characteristics, behavior, and HIV and STI prevalence. Logistic regression using generalized estimating equations to control for the cluster effect identified factors associated with HIV-positive serostatus adjusting for potential confounding. RESULTS: Of 2430 FSW, 40.8% were Nigerian, 36.8% were Malian, and 22.4% were from other neighboring countries. Between 2003 and 2009, HIV prevalence dropped from 44.14% to 28.49% (P < 0.0001) among Malians, from 21.33% to 12.71% (P = 0.0082) among Nigerians, and from 43.42% to 33.67% (P = 0.0442) among "others." Between 2000 and 2009, condom availability increased (89.18%-99.3%; P < 0.0001) as did HIV testing (40%-75%; P < 0.0001). Consistent condom use with clients improved for Malians (72.3%-81.5%; P = 0.0092), but not among Nigerians (92.7%-90.94%; P = 0.8240) and "others" (88.9%-88.48%; P = 0.8452). Consistent condom use with boyfriends was low and improved only for Nigerians (9.8%-28.4%; P = 0.0003). Factors associated with HIV prevalence in the multivariate model were older age, study year (2003 and 2006), nationality, lack of education, mobility, STI symptoms, gonorrhea prevalence, and younger age at first sex. CONCLUSIONS: This study documents progress in the fight against HIV among FSW in Mali. The different vulnerabilities to HIV found for different nationality FSW should be considered in programming and future research.