Publications

A high-speed, low-power 3D-SRAM architecture

This paper presents a novel 3D-SRAM architecture that can be used to extend the scaling of SRAM. This architecture significantly reduces the bit-line capacitance, achieves 3.4 times reduction in active power consumption and 1.8 times reduction in access time. In this architecture, local bit-lines are vertical and connect through select transistors to the global bit-lines routed on the bottom level. A proof-of-concept 32Kb sub-array emulating the critical path of the 3D-SRAM has demonstrated about 5 times improvement in power-delay over conventional 2D-SRAM.

DoS Pushback

HEGESMA: genome search meta-analysis and heterogeneity testing

Abstract Summary: Heterogeneity and genome search meta-analysis (HEGESMA) is a comprehensive software for performing genome scan meta-analysis, a quantitative method to identify genetic regions (bins) with consistently increased linkage score across multiple genome scans, and for testing the heterogeneity of the results of each bin across scans. The program provides as an output the average of ranks and three heterogeneity statistics, as well as corresponding significance levels. Statistical inferences are based on Monte Carlo permutation tests. The program allows both unweighted and weighted analysis, with the weights for each study as specified by the user. Furthermore, the program performs heterogeneity analyses restricted to the bins with similar average ranks. Availability: http://biomath.med.uth.gr Contact: zintza@med.uth.gr

JAMA

Kinetics of Antibody Concentration and Avidity for the Assessment of Immune Response to Pneumococcal Vaccine among Children with Bone Marrow Transplants

The kinetics of the immune response to the 23-valent pneumococcal polysaccharide vaccine (PPV) were studied in 38 children who received bone marrow transplants (BMTs). Anti-pneumococcal antibody concentrations increased 1 and 3 months after vaccination for all 5 serotypes tested, but, in 21 children, the vaccine was not adequately immunogenic. Children vaccinated <18 months after receiving a BMT had a 4.2-fold increased odds of poor response (P=. 06). Antibody concentrations returned close to baseline levels 9 months after vaccination. Avidity declined significantly as early as 1 month after vaccination and remained low thereafter. Antibody concentration responses to PPV were superior among 9 healthy control children (P=.001); 37 of 38 children with a BMT elicited adequate, persistent immune responses to Haemophilus influenzae conjugate vaccine. Immune responses to PPV in children with a BMT are suboptimal, short lived, and associated with declining avidity. The different kinetics of antibody concentration and avidity indicate that both markers should be used for evaluating pneumococcal vaccines in this high-risk population.

Probability of major depression diagnostic classification using semi-structured versus fully structured diagnostic interviews

Background Different diagnostic interviews are used as reference standards for major depression classification in research. Semi-structured interviews involve clinical judgement, whereas fully structured interviews are completely scripted. The Mini International Neuropsychiatric Interview (MINI), a brief fully structured interview, is also sometimes used. It is not known whether interview method is associated with probability of major depression classification. Aims To evaluate the association between interview method and odds of major depression classification, controlling for depressive symptom scores and participant characteristics. Method Data collected for an individual participant data meta-analysis of Patient Health Questionnaire-9 (PHQ-9) diagnostic accuracy were analysed and binomial generalised linear mixed models were fit. Results A total of 17 158 participants (2287 with major depression) from 57 primary studies were analysed. Among fully structured interviews, odds of major depression were higher for the MINI compared with the Composite International Diagnostic Interview (CIDI) (odds ratio (OR) = 2.10; 95% CI = 1.15–3.87). Compared with semi-structured interviews, fully structured interviews (MINI excluded) were non-significantly more likely to classify participants with low-level depressive symptoms (PHQ-9 scores ≤6) as having major depression (OR = 3.13; 95% CI = 0.98–10.00), similarly likely for moderate-level symptoms (PHQ-9 scores 7–15) (OR = 0.96; 95% CI = 0.56–1.66) and significantly less likely for high-level symptoms (PHQ-9 scores ≥16) (OR = 0.50; 95% CI = 0.26–0.97). Conclusions The MINI may identify more people as depressed than the CIDI, and semi-structured and fully structured interviews may not be interchangeable methods, but these results should be replicated. Declaration of interest Drs Jetté and Patten declare that they received a grant, outside the submitted work, from the Hotchkiss Brain Institute, which was jointly funded by the Institute and Pfizer. Pfizer was the original sponsor of the development of the PHQ-9, which is now in the public domain. Dr Chan is a steering committee member or consultant of Astra Zeneca, Bayer, Lilly, MSD and Pfizer. She has received sponsorships and honorarium for giving lectures and providing consultancy and her affiliated institution has received research grants from these companies. Dr Hegerl declares that within the past 3 years, he was an advisory board member for Lundbeck, Servier and Otsuka Pharma; a consultant for Bayer Pharma; and a speaker for Medice Arzneimittel, Novartis, and Roche Pharma, all outside the submitted work. Dr Inagaki declares that he has received grants from Novartis Pharma, lecture fees from Pfizer, Mochida, Shionogi, Sumitomo Dainippon Pharma, Daiichi-Sankyo, Meiji Seika and Takeda, and royalties from Nippon Hyoron Sha, Nanzando, Seiwa Shoten, Igaku-shoin and Technomics, all outside of the submitted work. Dr Yamada reports personal fees from Meiji Seika Pharma Co., Ltd., MSD K.K., Asahi Kasei Pharma Corporation, Seishin Shobo, Seiwa Shoten Co., Ltd., Igaku-shoin Ltd., Chugai Igakusha and Sentan Igakusha, all outside the submitted work. All other authors declare no competing interests. No funder had any role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication.

Architectural support for copy and tamper resistant software

Although there have been attempts to develop code transformations that yield tamper-resistant software, no reliable software-only methods are know. This paper studies the hardware implementation of a form of execute-only memory (XOM) that allows instructions stored in memory to be executed but not otherwise manipulated. To support XOM code we use a machine that supports internal compartments---a process in one compartment cannot read data from another compartment. All data that leaves the machine is encrypted, since we assume external memory is not secure. The design of this machine poses some interesting trade-offs between security, efficiency, and flexibility. We explore some of the potential security issues as one pushes the machine to become more efficient and flexible. Although security carries a performance penalty, our analysis indicates that it is possible to create a normal multi-tasking machine where nearly all applications can be run in XOM mode. While a virtual XOM machine is possible, the underlying hardware needs to support a unique private key, private memory, and traps on cache misses. For efficient operation, hardware assist to provide fast symmetric ciphers is also required.

Heritability and Genome-Wide Association Study to Assess Genetic Differences between Advanced Age-related Macular Degeneration Subtypes

Implementing Pushback: Router-Based Defense Against DDoS Attacks

Pushback is a mechanism for defending against distributed denial-of-service (DDoS) attacks. DDoS attacks are treated as a congestion-control problem, but because most such congestion is caused by malicious hosts not obeying traditional end-to-end congestion control, the problem must be handled by the routers. Functionality is added to each router to detect and preferentially drop packets that probably belong to an attack. Upstream routers are also notified to drop such packets (hence the term Pushback) in order that the router's resources be used to route legitimate traffic. In this paper we present an architecture for Pushback, its implementation under FreeBSD, and suggestions for how such a system can be implemented in core routers.

Screening performance of first-trimester nuchal translucency for major cardiac defects: a meta-analysis

JAMA

Statistical Guidance to Authors at Top-Ranked Journals across Scientific Disciplines

Scientific journals may counter the misuse, misreporting, and misinterpretation of statistics by providing guidance to authors. We described the nature and prevalence of statistical guidance at 15 journals (top-ranked by Impact Factor) in each of 22 scientific disciplines across five high-level domains (N = 330 journals). The frequency of statistical guidance varied across domains (Health &amp; Life Sciences: 122/165 journals, 74%; Multidisciplinary: 9/15 journals, 60%; Social Sciences: 8/30 journals, 27%; Physical Sciences: 21/90 journals, 23%; Formal Sciences: 0/30 journals, 0%). In one discipline (Clinical Medicine), statistical guidance was provided by all examined journals and in two disciplines (Mathematics and Computer Science) no examined journals provided statistical guidance. Of the 160 journals providing statistical guidance, 93 had a dedicated statistics section in their author instructions. The most frequently mentioned topics were confidence intervals (90 journals) and p-values (88 journals). For six “hotly debated” topics (statistical significance, p-values, Bayesian statistics, effect sizes, confidence intervals, and sample size planning/justification) journals typically offered implicit or explicit endorsement and rarely provided opposition. The heterogeneity of statistical guidance provided by top-ranked journals within and between disciplines highlights a need for further research and debate about the role journals can play in improving statistical practice.

Independent discussion sections for improving inferential reproducibility in published research

In‐house editorials and journalistic pieces comprise a massive corpus in the scientific literature that can be improved

Abstract In‐house editorials and journalistic pieces are massively published in peer‐reviewed scientific journals. This corpus has remained outside the efforts of evidence‐based medicine and research reform, and it can be imbued with unchecked biases. High‐impact journals publish such pieces massively and may generate strong support for specific narratives and perspectives. Pieces with a political slant are also a major issue. Besides high‐impact journals, across the entire scientific corpus, such pieces may be (mis)used to boost impact factors, create implausibly prolific CVs (occasionally even fraudulent) and can be powerful instruments of opinion making favouring some sponsors. Here we propose how this influential literature corpus may be strengthened to maximize its benefits and diminish its potential harms. Helpful measures to consider may include bolstering transparency (on authorship, financial compensation, disclosures of publication‐specific and generic conflicts of interest, handling of political issues, peer‐review, commissioning and timing); self‐regulation with limits per author, improvement of subject matter expertise (with experts, meta‐researchers and methodologists); balance of perspectives (with debates and for choice of topics); and post‐publication review, audit, correction and potential retraction, as needed. A systematic research agenda is needed to study better this phenomenon and also the effectiveness of proposed interventions.

Decision letter for "A Generalizable Speech Emotion Recognition Model Reveals Depression and Remission"

The language and conceptual framework of "research reproducibility" are nonstandard and unsettled across the sciences.In this Perspective, we review an array of explicit and implicit definitions of reproducibility and related terminology, and discuss how to avoid potential misunderstandings when these terms are used as a surrogate for "truth.

Automated Design Space Exploration of CGRA Processing Element Architectures using Frequent Subgraph Analysis

The architecture of a coarse-grained reconfigurable array (CGRA) processing element (PE) has a significant effect on the performance and energy efficiency of an application running on the CGRA. This paper presents an automated approach for generating specialized PE architectures for an application or an application domain. Frequent subgraphs mined from a set of applications are merged to form a PE architecture specialized to that application domain. For the image processing and machine learning domains, we generate specialized PEs that are up to 10.5x more energy efficient and consume 9.1x less area than a baseline PE.

Extended-Interval Aminoglycosides in Children: More Guidance Is Needed: In Reply

In Reply.—We thank Best et al for their comments. We fully agree that safety aspects in randomized trials of aminoglycoside dosing, as in the vast majority of biomedical fields,1 is suboptimal and that the situation needs to be improved.2 Large-scale studies on the ototoxicity profile of single daily doses would be welcome. However, realistically, these studies are likely to be mostly, if not exclusively, nonrandomized at this point.We disagree that the current evidence does not suggest that we can confidently recommend universal change in aminoglycoside dosing for children. Of course, evidence can never be final, but there are few topics in pediatrics for which so many trials have been done with uniformly reassuring results. We disagree about the fear of an “incomplete evidence base.” The incomplete evidence base pertains mostly to the inappropriate continued use of multiple daily doses in children. In fact, multiple daily dosing of aminoglycosides was universally adopted for both adults and children a long time ago based on absolutely no evidence. Multiple daily dosing has remained a common, if not prevalent, strategy for both children and adults despite the fact that evidence from many dozens of randomized trials and meta-analyses has failed to show its superiority and have even suggested its inferiority.3As far as subgroup populations are concerned, there is no evidence to suggest that multiple daily dosing is indicated in any patient subgroup. Perpetuating emphasis on spurious subgroup differences4 is only likely to inappropriately delay the wider adoption of single daily dosing.Finally, therapeutic dose monitoring is a different question, and robust evidence is needed to decide when (or if ever) it is indeed indicated to optimize single daily dosing.5 Single daily dosing translates to lower trough values and should simplify the need for monitoring anyhow.6

SCENIC international consensus statement on surveillance and management of dysplasia in inflammatory bowel disease