Search

Multiobjective optimization approaches have allowed the improvement of technical features in industrial processes, focusing on more accurate approaches for solving complex engineering problems and support decision-making. This paper proposes a hybrid approach to optimize the 3D printing technology parameters, integrating the design of experiments and multiobjective optimization methods, as an alternative to classical parametrization design used in machining processes. Alongside the approach, a multiobjective differential evolution with uniform spherical pruning (usp-MODE) algorithm is proposed to serve as an optimization tool. The parametrization design problem considered in this research has the following three objectives: to minimize both surface roughness and dimensional accuracy while maximizing the mechanical resistance of the prototype. A benchmark with non-dominated sorting genetic algorithm II (NSGA-II) and with the classical sp-MODE is used to evaluate the performance of the proposed algorithm. With the increasing complexity of engineering problems and advances in 3D printing technology, this study demonstrates the applicability of the proposed hybrid approach, finding optimal combinations for the machining process among conflicting objectives regardless of the number of decision variables and goals involved. To measure the performance and to compare the results of metaheuristics used in this study, three Pareto comparison metrics have been utilized to evaluate both the convergence and diversity of the obtained Pareto approximations for each algorithm: hyper-volume (H), g-Indicator (G), and inverted generational distance. To all of them, ups-MODE outperformed, with significant figures, the results reached by NSGA-II and sp-MODE algorithms.

BACKGROUND Leukaemias comprise a heterogenous group of haematological malignancies. In CONCORD-3, we analysed data for children (aged 0-14 years) and adults (aged 15-99 years) diagnosed with a haematological malignancy during 2000-14 in 61 countries. Here, we aimed to examine worldwide trends in survival from leukaemia, by age and morphology, in young patients (aged 0-24 years). METHODS We analysed data from 258 population-based cancer registries in 61 countries participating in CONCORD-3 that submitted data on patients diagnosed with leukaemia. We grouped patients by age as children (0-14 years), adolescents (15-19 years), and young adults (20-24 years). We categorised leukaemia subtypes according to the International Classification of Childhood Cancer (ICCC-3), updated with International Classification of Diseases for Oncology, third edition (ICD-O-3) codes. We estimated 5-year net survival by age and morphology, with 95% CIs, using the non-parametric Pohar-Perme estimator. To control for background mortality, we used life tables by country or region, single year of age, single calendar year and sex, and, where possible, by race or ethnicity. All-age survival estimates were standardised to the marginal distribution of young people with leukaemia included in the analysis. FINDINGS 164 563 young people were included in this analysis: 121 328 (73·7%) children, 22 963 (14·0%) adolescents, and 20 272 (12·3%) young adults. In 2010-14, the most common subtypes were lymphoid leukaemia (28 205 [68·2%] patients) and acute myeloid leukaemia (7863 [19·0%] patients). Age-standardised 5-year net survival in children, adolescents, and young adults for all leukaemias combined during 2010-14 varied widely, ranging from 46% in Mexico to more than 85% in Canada, Cyprus, Belgium, Denmark, Finland, and Australia. Individuals with lymphoid leukaemia had better age-standardised survival (from 43% in Ecuador to ≥80% in parts of Europe, North America, Oceania, and Asia) than those with acute myeloid leukaemia (from 32% in Peru to ≥70% in most high-income countries in Europe, North America, and Oceania). Throughout 2000-14, survival from all leukaemias combined remained consistently higher for children than adolescents and young adults, and minimal improvement was seen for adolescents and young adults in most countries. INTERPRETATION This study offers the first worldwide picture of population-based survival from leukaemia in children, adolescents, and young adults. Adolescents and young adults diagnosed with leukaemia continue to have lower survival than children. Trends in survival from leukaemia for adolescents and young adults are important indicators of the quality of cancer management in this age group. FUNDING Children with Cancer UK, the Institut National du Cancer, La Ligue Contre le Cancer, Centers for Disease Control and Prevention, Swiss Re, Swiss Cancer Research foundation, Swiss Cancer League, Rossy Family Foundation, US National Cancer Institute, and the American Cancer Society.

The generation of electric energy by photovoltaic (PV) panels depends on many parameters, one of them is the sun’s angle of incidence. By using solar active trackers, it is possible to maximize generation capacity through real-time positioning. However, if the engines that update the position of the panels use more energy than the difference in efficiency, the solar tracker system becomes ineffective. In this way, a time series forecasting method can be assumed to determine the generation capacity in a pre-established horizon prediction to evaluate if a position update would provide efficient results. Among a wide range of algorithms that can be used in forecasting, this work considered a Neuro-Fuzzy Inference System due to its combined advantages such as smoothness property from Fuzzy systems and adaptability property from neural networks structures. Focusing on time series forecasting, this article presents a model and evaluates the solar prediction capacity using the Wavelet Neuro-Fuzzy algorithm, where Wavelets were included in the model for feature extraction. In this sense, this paper aims to evaluate whether it is possible to obtain reasonable accuracy using a hybrid model for electric power generation forecasting considering solar trackers. The main contributions of this work are related to the efficiency improvement of PV panels. By assuming a hybrid computational model, it is possible to make a forecast and determine if the use of solar tracking is interesting during certain periods. Finally, the proposed model showed promising results when compared to traditional Nonlinear autoregressive model structures.

General anaesthesia is a clinical procedure that involves the continuous monitoring of several parameters for the correct application of anaesthetics and associated drugs. Focusing on the automatic control in anaesthesia, this work presents a multiobjective optimization design of controllers based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to solve the problem of drug delivery for induction of anaesthesia. Five Proportional-Integral-Derivative (PID) controllers in a decentralized scheme were tuned for one specific patient and tested in a total of 24 simulated patients. Acting over the infusions of Propofol, Remifentanil, Atracurium, Dobutamine, and Sodium Nitroprusside the proposed controllers could maintain the controlled variables in a safe range for surgical procedures.

The electricity demand has grown continuously in recent years, raising the necessity to expand generation sources, distribution networks, and equipment efficiency. In addition, it is necessary to attend sustainable development in a conciliated manner. Applications involving the intelligent management of the distributed networks have increased to achieve a balance between growth and sustainability. In this context, this article presents the development of an Intelligent Electric Power Management System (IEPMS) for the economic maximisation of a photovoltaic system applied to a prosumer residential unit without storage in Brazil. Using historical meteorological data and a heuristic to simulate energy use habits, the IEPMS forecasts both generation and demand in 24 hours. From the projections, an optimisation problem was built and solved using the genetic algorithm technique to find the most economical moments for driving loads. This model aims to reach the lowest daily cost of electricity, considering the return (sale) of unused energy to the power distribution company. The validation of the IEPMS considered four usage patterns, integrating 26 scenarios, those composed by the (i) flexibility; (ii) type of tariff; and (iii) hit rates provided by the climate forecasting method proposed for the system. As a result, the IEPMS savings considering the white tariff were 34.72% for one year, assuming full-time external work usage. Additionally, it was possible to identify in all scenarios that the proposed method’s performance was not less than 97%, measured through the relative error among distinct hit rates of the evaluated climatic forecast.

Identifying problems in insulators is a task that requires the experience of the operator. Contaminated insulators generally do not represent a system failure, however, due to higher surface conductivity, they may suffer from electrical discharges and may result in irreversible failures. The identification of possible failures in inspections can help to forecast faults to improve reliability in the power grid. Based on this need, this article presents a study on fault prediction in distribution insulators, through a laboratory evaluation in a contaminated insulator, where 13.8 kV (root mean square) was applied considering an ultrasound detector connected to a computer for data set acquisition. In the sequence, a time series prediction, using a hybrid deep learning technique defined as wavelet long short-term memory (LSTM), was performed. The hybrid LSTM was proposed considering feature extraction through the wavelet energy coefficient. Finally, for a complete evaluation, deeper LSTM layers were included, and both the training method and the hardware configuration were evaluated. The wavelet LSTM algorithm showed interesting accuracy results when compared to classic prediction algorithms like the non-linear autoregressive exogenous model.

Breast cancer (BC) risk factors have been differentially associated with BC subtypes, but quantification is still undefined. Therefore, we compared selected risk factors with BC subtypes, using a case-case approach. We retrieved 1321 invasive female BCs from the Piedmont Cancer Registry. Through record linkage of clinical records, we obtained data on estrogen (Er) and progesterone (Pr) receptors, Ki67 and HER2+ status, BC family history, breast imaging reporting and data system (BI-RADS) density, reproductive risk factors and education. We defined BC subtypes as follows : luminal A (Er+ and/or Pr+ , HER2− , low Ki67), luminal BH- (Er+ and/or Pr + , HER2− , Ki67 high), luminal BH+ (Er+ and/or Pr + , HER2+), HER2+ (Er − , Pr − , HER2+), ) and triple negative (Er − , Pr − , HER2−). Using a multinomial regression model, we estimated the odds ratios (ORs) for selected BC risk factors considering luminal A as reference. For triple negative, the OR for BC family history was 1.83 (95% confidence interval (CI) 1.13–2.97). Compared to BI-RADS 1, for triple negative, the OR for BI-RADS 2 was 0.56 (95% CI 0.27–1.14) and for BI-RADS 3–4 was 0.37 (95% CI 0.15–0.88); for luminal BH +, the OR for BI-RADS 2 was 2.36 (95% CI 1.08–5.11). For triple negative, the OR for high education was 1.78 (95% CI 1.03–3.07), and for late menarche, the OR was 1.69 (95% CI 1.02–2.81). For luminal BH + , the OR for parous women was 0.56 (95% CI 0.34–0.92). This study supported BC etiologic heterogeneity across subtypes, particularly for triple negative.

The classification of distinct problems of insulators in the distribution networks is a task that requires operator's experience. The applications of techniques to automate the inspection of electrical systems with the objective of detecting faults in insulators have shown to be reasonable alternatives to improve reliability in power grid. In this paper, based on the development of an experimental setup, signals are acquired considering three distinct faults in insulators. In this case, 13.8 kV (rms) is applied in drilled, contaminated, and good insulators considering an ultrasound detector connected to a computer. In the sequence, a multiclass classification method is proposed considering the ensemble of classifiers. The method considers the association of five distinct techniques, Bottom-Up segmentation, wavelet energy coefficient, principal component analysis, and particle swarm optimization associated with ensemble extreme learning machine (EN-ELM). Named optimized ensemble extreme learning machine, the present approach outperforms the original EN-ELM method. Finally, results show significant increase in robustness and faster training procedure when compared to classical approaches.

Genome-wide association studies (GWAS) and candidate pathway studies have identified low-penetrant genetic variants associated with cutaneous melanoma. We investigated the association of melanoma-risk variants with primary melanoma tumor prognostic characteristics and melanoma-specific survival. The Genes, Environment, and Melanoma Study enrolled 3285 European origin participants with incident invasive primary melanoma. For each of 47 melanoma-risk single nucleotide polymorphisms (SNPs), we used linear and logistic regression modeling to estimate, respectively, the per allele mean changes in log of Breslow thickness and odds ratios for presence of ulceration, mitoses, and tumor-infiltrating lymphocytes (TILs). We also used Cox proportional hazards regression modeling to estimate the per allele hazard ratios for melanoma-specific survival. Passing the false discovery threshold (p = 0.0026) were associations of IRF4 rs12203592 and CCND1 rs1485993 with log of Breslow thickness, and association of TERT rs2242652 with presence of mitoses. IRF4 rs12203592 also had nominal associations (p < 0.05) with presence of mitoses and melanoma-specific survival, as well as a borderline association (p = 0.07) with ulceration. CCND1 rs1485993 also had a borderline association with presence of mitoses (p = 0.06). MX2 rs45430 had nominal associations with log of Breslow thickness, presence of mitoses, and melanoma-specific survival. Our study indicates that further research investigating the associations of these genetic variants with underlying biologic pathways related to tumor progression is warranted.

Multiobjective optimization approaches have allowed the improvement of technical features in industrial processes, focusing on more accurate approaches for solving complex engineering problems and support decision-making. This paper proposes a hybrid approach to optimize the 3D printing technology parameters, integrating the design of experiments and multiobjective optimization methods, as an alternative to classical parametrization design used in machining processes. Alongside the approach, a multiobjective differential evolution with uniform spherical pruning (usp-MODE) algorithm is proposed to serve as an optimization tool. The parametrization design problem considered in this research has the following three objectives: to minimize both surface roughness and dimensional accuracy while maximizing the mechanical resistance of the prototype. A benchmark with non-dominated sorting genetic algorithm II (NSGA-II) and with the classical sp-MODE is used to evaluate the performance of the proposed algorithm. With the increasing complexity of engineering problems and advances in 3D printing technology, this study demonstrates the applicability of the proposed hybrid approach, finding optimal combinations for the machining process among conflicting objectives regardless of the number of decision variables and goals involved. To measure the performance and to compare the results of metaheuristics used in this study, three Pareto comparison metrics have been utilized to evaluate both the convergence and diversity of the obtained Pareto approximations for each algorithm: hyper-volume (H), g-Indicator (G), and inverted generational distance (IGD). To all of them, ups-MODE outperformed, with significant figures, the results reached by NSGA-II and sp-MODE algorithms.

The digital transformation of society, strengthened by the social isolation resulting from the COVID-19 pandemic, boosted sales and returns of products in e-commerce. In this sense, reverse logistics in e-commerce (RLec) has become essential to meet environmental legislation and consumer expectations, which evaluate exchange policies on new purchases. In this sense, this article presents a systematic review of the literature and content analysis, from 2009 to 2019, to identify methods of decision making and applications in RLec. Thus, 261 publications were selected, of which 92 met the search criteria related to reverse logistics and only 7 applied to e-commerce. In view of this, the main applications involved network design (26%), remanufacturing (21%) and outsourcing (16%), aiming at reducing costs and identifying barriers in reverse operations. Finally, artificial intelligence for decision making was identified as a competitive differential in reducing the complexity and subjectivity of LRec problems.

Nearly 1.24 million people die every year in traffic accidents. More than 20% of the number of deaths are pedestrians. Traffic crashes will become, in 2030, the fifth leading cause of deaths worldwide. A considerable amount of these accidents occurs due to either imprudence or lack of dexterity of the driver. More than 70% of traffic accidents are attributed to human error, such as perception errors as distance judgment error or simply by lack of attention. Because of the problem aforementioned, this work proposes the development and discussion of a pedestrian recognition system using an automotive radar of 79 GHz. The main idea is to early detect pedestrians using the micro Doppler characteristics of the human body in near to crash situations (0–15 m). Based on these requirements, at first this work presents the improvement of velocity resolution of the radar system with the objective of extracting the micro Doppler characteristics of the objects. The improvement of velocity resolution was reached assuming the application of multi-objective optimization techniques, in this case Genetic Algorithm (GA) and Random Search (RS) were adopted, enhancing the quality of the radar signal. Based on radar measures of velocity and range, a machine learning approach was considered in order to classify objects detected by the radar. The Support Vector Machine (SVM) method was assumed to distinguish pedestrians and non-pedestrians objects. Four different SVM based models were considered in this work, aiming the improvement of both classification performance and speed, comparing three different kernel functions. As a result, it was possible to verify the advantages of the multi-objective optimization to extract the micro Doppler effects based on radar signals. Moreover, the optimization reached the velocity resolution of 0.12 m/s. To conclude, we showed that a model involving a polynomial kernel for the SVM reported better result in terms of accuracy (99.5%), confirming the promising perspectives of vehicles embedded radar-based safety systems.

BACKGROUND: Genome-wide association studies have reported that genetic variation at ANRIL (CDKN2B-AS1) is associated with risk of several chronic diseases including coronary artery disease, coronary artery calcification, myocardial infarction, and type 2 diabetes mellitus. ANRIL is located at the CDKN2A/B locus, which encodes multiple melanoma tumor suppressors. We investigated the association of these variants with melanoma prognostic characteristics. METHODS: The Genes, Environment, and Melanoma Study enrolled 3,285 European origin participants with incident invasive primary melanoma. For each of ten disease-associated single-nucleotide polymorphisms (SNPs) at or near ANRIL, we used linear and logistic regression modeling to estimate, respectively, the per allele mean changes in log of Breslow thickness and odds ratios (ORs) for presence of ulceration and tumor-infiltrating lymphocytes (TILs). We also assessed effect modification by tumor NRAS/BRAF mutational status. RESULTS: Rs518394, rs10965215, and rs564398 passed false discovery and were each associated (P ≤ 0.005) with TILs, although only rs564398 was independently associated (P = 0.0005) with TILs. Stratified by NRAS/BRAF mutational status, rs564398*A was significantly positively associated with TILs among NRAS/BRAF mutant, but not wildtype, cases. We did not find SNP associations with Breslow thickness or ulceration. CONCLUSIONS: ANRIL rs564398 was associated with TIL presence in primary melanomas, and this association may be limited to NRAS/BRAF mutant cases. IMPACT: Pathways related to ANRIL variants warrant exploration in relationship to TILs in melanoma, especially given the impact of TILs on immunotherapy and survival.

Na análise biomecânica de nadadores, o uso de câmeras vem contribuindo para a melhora de rendimento esportivo. Porém grande parte dos vídeos são gravados de forma manual. Este trabalho tem como objetivo realizar uma análise comparativa entre os algoritmos de aprendizado profundo, YOLO v4 e YOLO v4 Tiny, para detecção de nadadores, visando o projeto de um veículo capaz de seguir, de forma autônoma, os atletas ao longo da piscina. O treinamento dos algoritmos foi realizado com a base de dados Video Diver Dataset (VDD-C), que contém imagens anotadas de mergulhadores em piscinas e na costa de Barbados no Caribe. Para comparação entre os algoritmos, levou-se em consideração qualidade e tempo de detecção, baseando-se nas métricas de Average Intersection Over Union (IoU), Recall, Average Precision (AP) e tempo de latência. O algoritmo YOLO v4, apresentou superioridade na qualidade de detecção, atingindo 71,38% de Average IoU. Porém a YOLO v4 Tiny, apresentou um tempo de latência 6,93 vezes menor, mostrando ser uma opção para sistemas embarcados e aplicações em tempo real.

Clonal hematopoiesis of indeterminate potential (CHIP) is associated with increased risk of cancers and inflammation-related diseases. This phenomenon becomes common in persons aged ≥80 years, in whom the implications of CHIP are not well defined. We performed a mutational screening in 1794 persons aged ≥80 years and investigated the relationships between CHIP and associated pathologies. Mutations were observed in one-third of persons aged ≥80 years and were associated with reduced survival. Mutations in JAK2 and splicing genes, multiple mutations (DNMT3A, TET2, and ASXL1 with additional genetic lesions), and variant allele frequency ≥0.096 had positive predictive value for myeloid neoplasms. Combining mutation profiles with abnormalities in red blood cell indices improved the ability of myeloid neoplasm prediction. On this basis, we defined a predictive model that identifies 3 risk groups with different probabilities of developing myeloid neoplasms. Mutations in DNMT3A, TET2, ASXL1, or JAK2 were associated with coronary heart disease and rheumatoid arthritis. Cytopenia was common in persons aged ≥80 years, with the underlying cause remaining unexplained in 30% of cases. Among individuals with unexplained cytopenia, the presence of highly specific mutation patterns was associated with myelodysplastic-like phenotype and a probability of survival comparable to that of myeloid neoplasms. Accordingly, 7.5% of subjects aged ≥80 years with cytopenia had presumptive evidence of myeloid neoplasm. In summary, specific mutational patterns define different risk of developing myeloid neoplasms vs inflammatory-associated diseases in persons aged ≥80 years. In individuals with unexplained cytopenia, mutational status may identify those subjects with presumptive evidence of myeloid neoplasms.

Objectives Cigarette smoking is related to higher levels of circulating androgens, but its association with androgen receptor (AR) status is still unaddressed. Methods We analysed, with a case-only approach, smoking habits according to AR status in 112 cases of invasive female breast cancer, from the Piedmont Cancer Registry. We used a multivariate logistic regression model to estimate the odds ratio (OR) and the corresponding confidence interval (CI). Results The OR of AR-positive breast cancer (versus AR-negative) for ever smokers (versus never) was 2.85 (95% CI 1.02-7.96). Conclusion Smoking is related to AR-positive breast cancer.

The steep increase in incidence of cutaneous malignant melanoma in white populations mainly applies to thin lesions with good survival suggesting overdiagnosis. Based on population-based cancer registries (CRs), we have investigated changes in aggressive melanoma, selecting only cases who died within 1 or 3 years after diagnosis in 11 European countries between 1995 and 2012. Trends in fatal cases were analysed by period of diagnosis, sex, tumour thickness, histologic subtype of the lesion, tumour site and CR with a multivariate generalised linear mixed effects model, where geographical area was considered as a random effect. We collected data on 123 360 invasive melanomas, with 5133 fatal cases at 1 year (4%) and 12 330 (10%) at 3 years. The number of fatal cases showed a 16% decrease at 1 year and 8% at 3 years between the first (1995-2000) and the last (2007-2012) period. The highest proportion of fatal cases was seen for men, older age (≥65 years), thick lesions (>1 mm), nodular melanoma, melanoma on the trunk and for poorly documented cases, lacking information about thickness and histologic subtype. The mixed-effects model showed a remarkable variability among European countries. The majority of registries showed a decreasing trend in fatal cases, but a few registries showed an opposite pattern. Trends in fatal melanoma cases, highlighting real changes in risk not related to overdiagnosis, showed a decrease in most European countries, with a few exceptions. Stronger efforts for early detection could lead to a more efficient treatment of melanoma in general.

The availability of diverse data has increased the demand for expertise in algorithmic trading strategies. Reinforcement learning has shown interesting applicability in a wide range of tasks, especially in some challenging problems as trading, where slow model convergence, inference speed, and reduced model accuracy appear as barriers in this type of application. In this paper, we propose the transformation of time series into images considering a transfer learning based on a semi-supervised model with deep Q learning agents, where labels were generated by an evolutionary algorithm to improve both training speed and performance measures.

A utilização de manutenção preditiva na indústria auxilia na diminuição de custos de produção, pois reduz imprevistos ao reparar equipamentos durante períodos ociosos de funcionamento. Diversos métodos de aprendizado de máquina têm sido empregados para detecção de falhas em processos de soldagem. Neste estudo, os algoritmos Árvores de Decisão, k-ésimo Vizinho mais Próximo, Naive Bayes, Máquina de Vetores de Suporte e Boosting Adaptativo foram avaliados levando em conta a detecção de falhas no processo de soldagem a ponto por resistência na indústria automotiva. Neste sentido, o objetivo deste trabalho foi analisar o desempenho desses algoritmos na detecção de expulsão em soldas utilizando dados, não rotulados, provenientes de um controlador proprietário. Os resultados obtidos apresentam a acurácia de cada modelo levando em conta o conjunto de dados selecionado, indicando que técnicas de aprendizado de máquina podem ser utilizadas para a detecção de falhas no processo de soldagem. Por fim,o método k-ésimo Vizinho mais Próximo se mostrou como o método mais preciso avaliado neste estudo.

A grande maioria dos sistemas comerciais de condicionamento de ar não consideram a questão do conforto térmico dos ocupantes e, devido às diferentes condições climáticas existentes no Brasil e à preocupação com a eficiência energética do setor de construção, torna-se evidente a necessidade de mecanismos distintos de controle para melhorar as condições de conforto térmico. Este trabalho está focado na otimização de controladores Proporcional-Integral (PI) com foco no conforto térmico. Inicialmente, é apresentado um procedimento de identificação do sistema, com base no método dos mínimos quadrados, para reproduzir o comportamento térmico de um ambiente interno. O modelo foi validado considerando um conjunto de dados obtido a partir de um software completo de construção e simulação de energia. Na sequência, três métodos distintos de otimização metaheurística: Evolução Diferencial, Otimização de Enxame de Partículas e Algoritmo Genético foram adaptados para o ajuste do PI, com foco no conforto térmico. Os resultados mostraram que a comparação entre todas as estratégias de ajuste indicou que resultados semelhantes foram obtidos com relação aos métodos de otimização em termos de conforto térmico.

The Proportional-Integral-Derivative (PID) controller is widely used in industrial applications. In order to evaluate the performance of distinct types of controllers, several benchmark systems are available in the specialized literature. In this work, an alternative approach to tune PID controllers based on the Competition Over Resources (COR) metaheuristic optimization algorithm and TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) method is proposed to solve a vapor compression refrigeration system problem, which is a benchmark control challenge. The proposed technique considered the normalization of multiple cost functions to simplify the selection of two PIDs’ parameters, reducing the need of understanding the influences of cost functions variations in complex problems. The proposed controllers outperformed the base line controllers, providing a performance index 68.39% better than the original approach. Finally, the proposed method was compared with other tuning methods.

BRAF and NRAS mutations arise early in melanoma development, but their associations with low-penetrance melanoma susceptibility loci remain unknown. In the Genes, Environment and Melanoma Study, 1,223 European-origin participants had their incident invasive primary melanomas screened for BRAF/NRAS mutations and germline DNA genotyped for 47 single-nucleotide polymorphisms identified as low-penetrant melanoma-risk variants. We used multinomial logistic regression to simultaneously examine each single-nucleotide polymorphism's relationship to BRAF V600E, BRAF V600K, BRAF other, and NRAS+ relative to BRAF-/NRAS- melanoma adjusted for study features. IRF4 rs12203592*T was associated with BRAF V600E (odds ratio [OR] = 0.59, 95% confidence interval [CI] = 0.43-0.79) and V600K (OR = 0.65, 95% CI = 0.41-1.03), but not BRAF other or NRAS+ melanoma. A global test of etiologic heterogeneity (Pglobal = 0.001) passed false discovery (Pglobal = 0.0026). PLA2G6 rs132985*T was associated with BRAF V600E (OR = 1.32, 95% CI = 1.05-1.67) and BRAF other (OR = 1.82, 95% CI = 1.11-2.98), but not BRAF V600K or NRAS+ melanoma. The test for etiologic heterogeneity (Pglobal) was 0.005. The IRF4 rs12203592 associations were slightly attenuated after adjustment for melanoma-risk phenotypes. The PLA2G6 rs132985 associations were independent of phenotypes. IRF4 and PLA2G6 inherited genotypes may influence melanoma BRAF/NRAS subtype development.

Brazilian agribusiness is responsible for almost 25% of the country gross domestic product, and companies from this economic sector may have strategies to control their actions in a competitive market. In this way, models to properly predict variations in the price of products and services could be one of the keys to the success in agribusiness. Consistent models are being adopted by companies as part of a decision making process when important choices are based on short or long-term forecasting. This work aims to evaluate Wavelet Neural Networks (WNNs) performance combined with five optimization techniques in order to obtain the best time series forecasting by considering two case studies in the agribusiness sector. The first one adopts the soybean sack price and the second deals with the demand problem of a distinct groups of products from a food company, where nonlinear trends are the main characteristic on both time series. The optimization techniques adopted in this work are: Differential Evolution, Artificial Bee Colony, Glowworm Swarm Optimization, Gravitational Search Algorithm, and Imperialist Competitive Algorithm. Those were evaluated by considering short-term and long-term forecasting, and a prediction horizon of 30 days ahead was considered for the soybean sack price case, while 12 months ahead was selected for the products demand case. The performance of the optimization techniques in training the WNN were compared to the well-established Backpropagation algorithm and Extreme Learning Machine (ELM) assuming accuracy measures. In long-term forecasting, which is considered more difficult than the short-term case due to the error accumulation, the best combinations in terms of precision was reached by distinct methods according to each case, showing the importance of testing different training strategies. This work also showed that the prediction horizon significantly affected the performance of each optimization method in different ways, and the potential of assuming optimization in WNN learning process.

Background We analysed trends in incidence for in situ and invasive melanoma in some European countries during the period 1995–2012, stratifying for lesion thickness. Material and methods Individual anonymised data from population-based European cancer registries (CRs) were collected and combined in a common database, including information on age, sex, year of diagnosis, histological type, tumour location, behaviour (invasive, in situ) and lesion thickness. Mortality data were retrieved from the publicly available World Health Organization database. Results Our database covered a population of over 117 million inhabitants and included about 415,000 skin lesions, recorded by 18 European CRs (7 of them with national coverage). During the 1995–2012 period, we observed a statistically significant increase in incidence for both invasive (average annual percent change (AAPC) 4.0% men; 3.0% women) and in situ (AAPC 7.7% men; 6.2% women) cases. Discussion The increase in invasive lesions seemed mainly driven by thin melanomas (AAPC 10% men; 8.3% women). The incidence of thick melanomas also increased, although more slowly in recent years. Correction for lesions of unknown thickness enhanced the differences between thin and thick cases and flattened the trends. Incidence trends varied considerably across registries, but only Netherlands presented a marked increase above the boundaries of a funnel plot that weighted estimates by their precision. Mortality from invasive melanoma has continued to increase in Norway, Iceland (but only for elder people), the Netherlands and Slovenia.

Genetic predisposition to cutaneous malignant melanoma (CMM) involves highly penetrant predisposing genes and low and intermediate penetrant predisposing alleles. However, the missing heritability in (CMM) is still high. For such and in order to identify new genetic factors for CMM, we conducted an exome sequencing study in high-risk CMM patients. Two rounds of exome sequencing were successively performed in 33 and 27 high-risk patients. We focused on genes carrying rare nonsense, frameshift, and splice variants (allelic frequency1%) that were present in both series of exomes. An extension study was then conducted in a large cohort (1 079 CMM patients and 1 230 Caucasian ethnically matched healthy controls), and the inactivating variants frequency was compared between groups using two-sided Fisher exact test. Two TP53AIP1 truncating mutations were identified in four patients: a frameshift c.63_64insG, p.Q22Afs*81 in two patients from the same family and in the proband of a second family; and a nonsense mutation c.95 C A, p.Ser32Stop in a patient with multiple CMMs. In all patients, TP53AIP1 truncating variants were strongly associated with CMM risk (two-sided Fisher exact test = 0.004, OR = 3.3[1.3-8.5]). Additionally, we showed that TP53AIP1 mRNA was strongly down-regulated throughout different phases of melanoma progression. TP53AIP1 gene is a TP53 target which plays a key role by inducting apoptosis in response to UV-induced DNA damage. Constitutional mutations of TP53AIP1 had previously been involved in susceptibility to prostate cancer. Our results show that constitutional truncating TP53AIP1 mutations predispose to CMM in the French population. Replication studies in other populations should be performed.