Explore the vast range of titles available.

Over the past two decades, support vector machines (SVMs) have become a popular supervised machine learning model, and plenty of distinct algorithms are designed separately based on different KKT conditions of the SVM model for classification/regression with different losses, including convex and or nonconvex loss. In this paper, we propose an algorithm that can train different SVM models in a unified scheme. First, we introduce a definition of the least squares type of difference of convex loss (LS-DC) and show that the most commonly used losses in the SVM community are LS-DC loss or can be approximated by LS-DC loss. Based on the difference of convex algorithm (DCA), we then propose a unified algorithm called UniSVM which can solve the SVM model with any convex or nonconvex LS-DC loss, wherein only a vector is computed by the specifically chosen loss. UniSVM has a dominant advantage over all existing algorithms for training robust SVM models with nonconvex losses because it has a closed-form solution per iteration, while the existing algorithms always need to solve an L1SVM/L2SVM per iteration. Furthermore, by the low-rank approximation of the kernel matrix, UniSVM can solve large-scale nonlinear problems efficiently. To verify the efficacy and feasibility of the proposed algorithm, we perform many experiments on small artificial problems and large benchmark tasks both with and without outliers for classification and regression for comparison with state-of-the-art algorithms. The experimental results demonstrate that UniSVM can achieve comparable performance in less training time. The foremost advantage of UniSVM is that its core code in Matlab is less than 10 lines; hence, it can be easily grasped by users or researchers.

The year 2015 marks the 30th birthday of DC (Difference of Convex functions) programming and DCA (DC Algorithms) which constitute the backbone of nonconvex programming and global optimization. In this article we offer a short survey on thirty years of developments of these theoretical and algorithmic tools. The survey is comprised of three parts. In the first part we present a brief history of the field, while in the second we summarize the state-of-the-art results and recent advances. We focus on main theoretical results and DCA solvers for important classes of difficult nonconvex optimization problems, and then give an overview of real-world applications whose solution methods are based on DCA. The third part is devoted to new trends and important open issues, as well as suggestions for future developments.

Bile acids are soluble derivatives of cholesterol produced in the liver that subsequently undergo bacterial transformation yielding a diverse array of metabolites. The bulk of bile acid synthesis takes place in the liver yielding primary bile acids; however, other tissues have also the capacity to generate bile acids (e.g. ovaries). Hepatic bile acids are then transported to bile and are subsequently released into the intestines. In the large intestine, a fraction of primary bile acids is converted to secondary bile acids by gut bacteria. The majority of the intestinal bile acids undergo reuptake and return to the liver. A small fraction of secondary and primary bile acids remains in the circulation and exert receptor-mediated and pure chemical effects (e.g. acidic bile in oesophageal cancer) on cancer cells. In this review, we assess how changes to bile acid biosynthesis, bile acid flux and local bile acid concentration modulate the behavior of different cancers. Here, we present in-depth the involvement of bile acids in oesophageal, gastric, hepatocellular, pancreatic, colorectal, breast, prostate, ovarian cancer. Previous studies often used bile acids in supraphysiological concentration, sometimes in concentrations 1000 times higher than the highest reported tissue or serum concentrations likely eliciting unspecific effects, a practice that we advocate against in this review. Furthermore, we show that, although bile acids were classically considered as pro-carcinogenic agents (e.g. oesophageal cancer), the dogma that switch, as lower concentrations of bile acids that correspond to their serum or tissue reference concentration possess anticancer activity in a subset of cancers. Differences in the response of cancers to bile acids lie in the differential expression of bile acid receptors between cancers (e.g. FXR vs. TGR5). UDCA, a bile acid that is sold as a generic medication against cholestasis or biliary surge, and its conjugates were identified with almost purely anticancer features suggesting a possibility for drug repurposing. Taken together, bile acids were considered as tumor inducers or tumor promoter molecules; nevertheless, in certain cancers, like breast cancer, bile acids in their reference concentrations may act as tumor suppressors suggesting a Janus-faced nature of bile acids in carcinogenesis.

DC (Difference of Convex functions) programming and DCA (DC Algorithm) play a key role in nonconvex programming framework. These tools have a rich and successful history of thirty five years of development, and the research in recent years is being increasingly explored to new trends in the development of DCA: design novel DCA variants to improve standard DCA, to deal with the scalability and with broader classes than DC programs. Following these trends, we address in this paper the two wide classes of nonconvex problems, called partial DC programs and generalized partial DC programs, and investigate an alternating approach based on DCA for them. A partial DC program in two variables (x,y)∈Rn×Rm takes the form of a standard DC program in each variable while fixing other variable. A so-named alternating DCA and its inexact/generalized versions are developed. The convergence properties of these algorithms are established: both exact and inexact alternating DCA converge to a weak critical point of the considered problem, in particular, when the Kurdyka–Łojasiewicz inequality property is satisfied, the algorithms furnish a Fréchet/Clarke critical point. The proposed algorithms are implemented on the problem of finding an intersection point of two nonconvex sets. Numerical experiments are performed on an important application that is robust principal component analysis. Numerical results show the efficiency and the superiority of the alternating DCA comparing with the standard DCA as well as a well known alternating projection algorithm.

In situ, continuous and real-time monitoring of respiration (R) and respiratory quotient (RQ) are crucial for identifying the optimal conditions for the long-term storage of fresh produce. This study reports the application of a gas sensor (RMS88) and a modular respirometer for in situ real-time monitoring of gas concentrations and respiration rates of strawberries during storage in a lab-scale controlled atmosphere chamber (190 L) and of Pinova apples in a commercial storage facility (170 t). The RMS88 consisted of wireless O2 (0% to 25%) and CO2 sensors (0% to 0.5% and 0% to 5%). The modular respirometer (3.3 L for strawberries and 7.4 L for apples) consisted of a leak-proof arrangement with a water-containing base plate and a glass jar on top. Gas concentrations were continuously recorded by the RMS88 at regular intervals of 1 min for strawberries and 5 min for apples and, in real-time, transferred to a terminal program to calculate respiration rates ( R O 2 and R CO 2 ) and RQ. Respiration measurement was done in cycles of flushing and measurement period. A respiration measurement cycle with a measurement period of 2 h up to 3 h was shown to be useful for strawberries under air at 10 °C. The start of anaerobic respiration of strawberries due to low O2 concentration (1%) could be recorded in real-time. R O 2 and R CO 2 of Pinova apples were recorded every 5 min during storage and mean values of 1.6 and 2.7 mL kg−1 h−1, respectively, were obtained when controlled atmosphere (CA) conditions (2% O2, 1.3% CO2 and 2 °C) were established. The modular respirometer was found to be useful for in situ real-time monitoring of respiration rate during storage of fresh produce and offers great potential to be incorporated into RQ-based dynamic CA storage system.

Contact tracing apps are an essential component of an effective COVID-19 testing strategy to counteract the spread of the pandemic and thereby avoid overburdening the health care system. As the adoption rates in several regions are undesirable, governments must increase the acceptance of COVID-19 tracing apps in these times of uncertainty. Building on the Uncertainty Reduction Theory (URT), this study aims to investigate how uncertainty reduction measures foster the adoption of COVID-19 tracing apps and how their use affects the perception of different risks. Representative survey data were gathered at two measurement points (before and after the app's release) and analyzed by performing covariance-based structural equation modeling (n=1003). We found that uncertainty reduction measures in the form of the transparency dimensions disclosure and accuracy, as well as social influence and trust in government, foster the adoption process. The use of the COVID-19 tracing app in turn reduced the perceived privacy and performance risks but did not reduce social risks and health-related COVID-19 concerns. This study contributes to the mass adoption of health care technology and URT research by integrating interactive communication measures and transparency as a multidimensional concept to reduce different types of uncertainty over time. Furthermore, our results help to derive communication strategies to promote the mass adoption of COVID-19 tracing apps, thus detecting infection chains and allowing intelligent COVID-19 testing.

We propose a DC (Difference of two Convex functions) formulation approach for sparse optimization problems having a cardinality or rank constraint. With the largest-k norm, an exact DC representation of the cardinality constraint is provided. We then transform the cardinality-constrained problem into a penalty function form and derive exact penalty parameter values for some optimization problems, especially for quadratic minimization problems which often appear in practice. A DC Algorithm (DCA) is presented, where the dual step at each iteration can be efficiently carried out due to the accessible subgradient of the largest-k norm. Furthermore, we can solve each DCA subproblem in linear time via a soft thresholding operation if there are no additional constraints. The framework is extended to the rank-constrained problem as well as the cardinality- and the rank-minimization problems. Numerical experiments demonstrate the efficiency of the proposed DCA in comparison with existing methods which have other penalty terms.

Micro RNA (miRNA) research has great implications in uncovering the aetiology of neuropsychiatric conditions due to the role of miRNA in brain development and function. Schizophrenia, a complex yet devastating neuropsychiatric disorder, is one such condition that had been extensively studied in the realm of miRNA. Although a relatively new field of research, this area of study has progressed sufficiently to warrant dozens of reviews summarising findings from past to present. However, as a majority of reviews cannot encapsulate the full body of research, there is still a need to synthesise the diversity of publications made in this area in a systematic but easy-to-understand manner. Therefore, this study adopted bibliometrics and scientometrics, specifically document co-citation analysis (DCA), to review the literature on miRNAs in the context of schizophrenia over the course of history. From a literature search on Scopus, 992 papers were found and analysed with CiteSpace. DCA analysis generated a network of 13 major clusters with different thematic focuses within the subject area. Finally, these clusters are qualitatively discussed. miRNA research has branched into schizophrenia, among other medical and psychiatric conditions, due to previous findings in other forms of non-coding RNA. With the rise of big data, bioinformatics analyses are increasingly common in this field of research. The future of research is projected to rely more heavily on interdisciplinary collaboration. Additionally, it can be expected that there will be more translational studies focusing on the application of these findings to the development of effective treatments.

Introduction: Stereotactic radiosurgery is a highly precise radiotherapy technique widely used for the management of brain metastases. While VMAT enables highly conformal dose distributions, it is often associated with increased plan complexity and longer delivery times. Optimized dynamic conformal arc therapy (OptDCA) represents a less complex alternative that may achieve comparable dosimetric performance. In this retrospective study, dosimetric quality, deliverability, and plan complexity of VMAT and OptDCA were compared for single-isocenter SRS of multiple brain metastases. Materials and Methods: Thirty patients previously treated with VMAT were randomly selected and replanned using OptDCA with identical beam arrangements. Plan quality was evaluated using the Paddick conformity index, gradient index, target coverage, MUs, and brain V12Gy and V20Gy. Deliverability was assessed using gamma passing rates, and plan complexity was quantified using multiple complexity metrics. Results: VMAT achieved a slightly higher CI (0.72 vs. 0.71) but required a higher number of MUs (5376 vs. 4820), while no significant differences were observed in GI or target coverage. OptDCA demonstrated significantly higher GPR (median 96.95% vs. 91.1%) and consistently lower plan complexity. Significant correlations were observed between GPR and several complexity metrics for both techniques. Conclusion: Overall, OptDCA provides comparable plan quality to VMAT, while offering improved deliverability and reduced complexity, making it a viable alternative technique.

Purpose Volumetric modulated arc therapy (VMAT) is gaining popularity for stereotactic treatment of lung lesions for medically inoperable patients. Due to multiple beamlets in delivery of highly modulated VMAT plans, there are dose delivery uncertainties associated with small‐field dosimetry error and interplay effects with small lesions. We describe and compare a clinically useful dynamic conformal arc (DCA)‐based VMAT (d‐VMAT) technique for lung SBRT using flattening filter free (FFF) beams to minimize these effects. Materials and Methods Ten solitary early‐stage I‐II non‐small‐cell lung cancer (NSCLC) patients were treated with a single dose of 30 Gy using 3–6 non‐coplanar VMAT arcs (clinical VMAT) with 6X‐FFF beams in our clinic. These clinically treated plans were re‐optimized using a novel d‐VMAT planning technique. For comparison, d‐VMAT plans were recalculated using DCA with user‐controlled field aperture shape before VMAT optimization. Identical beam geometry, dose calculation algorithm, grid size, and planning objectives were used. The clinical VMAT and d‐VMAT plans were compared via RTOG‐0915 protocol compliances for conformity, gradient indices, and dose to organs at risk (OAR). Additionally, treatment delivery efficiency and accuracy were recorded. Results All plans met RTOG‐0915 requirements. Comparing with clinical VMAT, d‐VMAT plans gave similar target coverage with better target conformity, tighter radiosurgical dose distribution with lower gradient indices, and dose to OAR. Lower total number of monitor units and small beam modulation factor reduced beam‐on time by 1.75 min (P < 0.001), on average (maximum up to 2.52 min). Beam delivery accuracy was improved by 2%, on average (P < 0.05) and maximum up to 6% in some cases for d‐VMAT plans. Conclusion This simple d‐VMAT technique provided excellent plan quality, reduced intermediate dose‐spillage, and dose to OAR while providing faster treatment delivery by significantly reducing beam‐on time. This novel treatment planning approach will improve patient compliance along with potentially reducing intrafraction motion error. Moreover, with less MLC modulation through the target, d‐VMAT could potentially minimize small‐field dosimetry errors and MLC interplay effects. If available, d‐VMAT planning approach is recommended for future clinical lung SBRT plan optimization.