ImportanceThe protein-based SARS-CoV-2 vaccines FINLAY-FR-2 (Soberana 02) and FINLAY-FR-1A (Soberana Plus) showed good safety and immunogenicity in phase 1 and 2 trials, but the clinical efficacy of the vaccine remains unknown.ObjectiveTo evaluate the efficacy and safety of a 2-dose regimen of FINLAY-FR-2 (cohort 1) and a 3-dose regimen of FINLAY-FR-2 with FINLAY-FR-1A (cohort 2) in Iranian adults.Design, Setting, and ParticipantsA multicenter, randomized, double-blind, placebo-controlled, phase 3 trial was conducted at 6 cities in cohort 1 and 2 cities in cohort 2. Participants included individuals aged 18 to 80 years without uncontrolled comorbidities, coagulation disorders, pregnancy or breastfeeding, recent immunoglobulin or immunosuppressive therapy, and clinical presentation or laboratory-confirmed COVID-19 on enrollment. The study was conducted from April 26 to September 25, 2021.InterventionsIn cohort 1, 2 doses of FINLAY-FR-2 (n = 13 857) or placebo (n = 3462) were administered 28 days apart. In cohort 2, 2 doses of FINLAY-FR-2 plus 1 dose of FINLAY-FR-1A (n = 4340) or 3 placebo doses (n = 1081) were administered 28 days apart. Vaccinations were administered via intramuscular injection.Main Outcomes and MeasuresThe primary outcome was polymerase chain reaction–confirmed symptomatic COVID-19 infection at least 14 days after vaccination completion. Other outcomes were adverse events and severe COVID-19. Intention-to-treat analysis was performed.ResultsIn cohort 1 a total 17 319 individuals received 2 doses and in cohort 2 5521 received 3 doses of the vaccine or placebo. Cohort 1 comprised 60.1% men in the vaccine group and 59.1% men in the placebo group; cohort 2 included 59.8% men in the vaccine group and 59.9% in the placebo group. The mean (SD) age was 39.3 (11.9) years in cohort 1 and 39.7 (12.0) years in cohort 2, with no significant difference between the vaccine and placebo groups. The median follow-up time in cohort 1 was 100 (IQR, 96-106) days and, in cohort 2, 142 (137-148) days. In cohort 1, 461 (3.2%) cases of COVID-19 occurred in the vaccine group and 221 (6.1%) in the placebo group (vaccine efficacy: 49.7%; 95% CI, 40.8%-57.3%) vs 75 (1.6%) and 51 (4.3%) in cohort 2 (vaccine efficacy: 64.9%; 95% CI, 49.7%-59.5%). The incidence of serious adverse events was lower than 0.1%, with no vaccine-related deaths.Conclusions and RelevanceIn this multicenter, randomized, double-blind, placebo-controlled, phase 3 trial of the efficacy and safety of FINLAY-FR-2 and FINLAY-FR-1A, 2 doses of FINLAY-FR-2 plus the third dose of FINLAY-FR-1A showed acceptable vaccine efficacy against symptomatic COVID-19 as well as COVID-19–related severe infections. Vaccination was generally safe and well tolerated. Therefore, Soberana may have utility as an option for mass vaccination of the population, especially in resource-limited settings, because of its storage condition and affordable price.Trial Registrationisrctn.org Identifier: IRCT20210303050558N1

An application specific programmable processor is designed based on the analysis of a set of greedy recovery Compressive Sensing (CS) algorithms. The solution is flexible and customizable for a wide range of problem dimensions, as well as algorithms. The versatility of the approach is demonstrated by implementing Orthogonal Matching Pursuits, Approximate Messaging Passing and Normalized Iterative Hard Thresholding algorithms, all using a high-level language. Transported Triggered Architecture (TTA) framework is employed for the efficient implementation of macro operations shared by the algorithms. The performance of the CS algorithms on ARM Cortex-A15 and NIOS II processors has also been investigated, and empirical comparisons are presented. The flexible hardware design implemented on an FPGA achieves up to 7.80Ksample/s recovery at a power dissipation of 42μJ/sample and beats both ARM and NIOS in total power consumption.

An energy efficient architecture for TPUs that is based on reduced voltage operation. The errors are captured and corrected by utilizing ABFT and hence aggressive voltage scaling is made possible.

Low-level sensory data processing in many Internet-of-Things (IoT) devices pursue energy efficiency by utilizing sleep modes or slowing the clocking to the minimum. To curb the share of stand-by power dissipation in those designs, near-threshold/sub-threshold operational points or ultra-low-leakage processes in fabrication are employed. Those limit the clocking rates significantly, reducing the computing throughputs of individual processing cores. In this contribution we explore compensating for the performance loss of operating in near-threshold region (Vdd =0.6V) through massive parallelization. Benefits of near-threshold operation and massive parallelism are optimum energy consumption per instruction operation and minimized memory roundtrips, respectively. The Processing Elements (PE) of the design are based on Transport Triggered Architecture. The fine grained programmable parallel solution allows for fast and efficient computation of learnable low-level features (e.g. local binary descriptors and convolutions). Other operations, including Max-pooling have also been implemented. The programmable design achieves excellent energy efficiency for Local Binary Patterns computations.

Operating at reduced voltages offers substantial energy efficiency improvement but at the expense of increasing the probability of computational errors due to hardware faults. In this context, we targeted Deep Neural Networks (DNN) as energy hungry building blocks in embedded systems. Without an error feedback mechanism, blind voltage down-scaling will result in degraded accuracy or total system failure. In this paper, solutions based on the inherent properties of Self-Supervised Learning (SSL) and Algorithm Based Fault Tolerance (ABFT) techniques were investigated. A DNN model trained on MNIST data-set was deployed on a Field Programmable Gate Array (FPGA) that operated at reduced voltages and employed the proposed schemes. The SSL approach provides extremely lowoverhead fault detection at the cost of lower error coverage and extra training, while ABFT incurs less than 8% overheads at run-time with close to 100% error detection rate. By using the solutions, substantial energy savings, i.e., up to 48%, without compromising the accuracy of the model was achieved.