Binary Model Research Articles

A Novel Machine Learning Model and a Web Portal for Predicting the Human Skin Sensitization Effects of Chemical Agents

Skin sensitization is a significant concern for chemical safety assessments. Traditional animal assays often fail to predict human responses accurately, and ethical constraints limit the collection of human data, necessitating a need for reliable in silico models of skin sensitization prediction. This study introduces HuSSPred, an in silico tool based on the Human Predictive Patch Test (HPPT). HuSSPred aims to enhance the reliability of predicting human skin sensitization effects for chemical agents to support their regulatory assessment. We have curated an extensive HPPT database and performed chemical space analysis and grouping. Binary and multiclass QSAR models were developed with Bayesian hyperparameter optimization. Model performance was evaluated via five-fold cross-validation. We performed model validation with reference data from the Defined Approaches for Skin Sensitization (DASS) app. HuSSPred models demonstrated strong predictive performance with CCR ranging from 55 to 88%, sensitivity between 48 and 89%, and specificity between 37 and 92%. The positive predictive value (PPV) ranged from 84 to 97%, versus negative predictive value (NPV) from 22 to 65%, and coverage was between 75 and 93%. Our models exhibited comparable or improved performance compared to existing tools, and the external validation showed the high accuracy and sensitivity of the developed models. HuSSPred provides a reliable, open-access, and ethical alternative to traditional testing for skin sensitization. Its high accuracy and reasonable coverage make it a valuable resource for regulatory assessments, aligning with the 3Rs principles. The publicly accessible HuSSPred web tool offers a user-friendly interface for predicting skin sensitization based on chemical structure.

Pixels relationship analysis for extracting building footprints

Abstract. Currently, remote sensing research is focused on developing an automated algorithm that can compete with empirical methods for mapping the contours of individual buildings. Despite facing numerous challenges related to suboptimal imaging conditions, diverse building architecture, and complex backgrounds, creating such an algorithm is essential for monitoring urban and natural areas, generating 3D city models, managing disasters, and estimating population density. Obtaining the polygonal boundary of a building and extracting a vectorized building mask as output for direct use is one of the current challenges in drawing building outlines. This work provides a comprehensive workflow for building extraction and improves their predicted area by regularization the boundaries of buildings. First, a convolutional neural network is used to train the binary semantic segmentation model and then regularization and vectorization processes are performed. The main difference from existing methods is a new regularization method based on compiling a neighborhood matrix for each point belonging to the “building” class. According to the experimental results, the algorithm shows high efficiency: IOU (intersection over union) = 91.2%, AP (average precision) = 64.1% and AR (average recall) = 75.1%, comparable to leading building boundary extraction methods such as PolyWorld.

Determinants of receiving government support by small and medium-sized enterprises in Russia

Government support measures help companies overcome various restrictions impeding their growth and develop ment, which is of special relevance for small and medium-sized enterprises (SMEs). These measures, therefore, must reach the target companies and ensure efficient allocation of resources. The article reveals key characteristics of SMEs in Russia that increase the probability of them receiving government support. The research methodology is based on public interest theory, public choice theory and the concept of economic freedom. The research method is a two-step regression analysis, in which the binary choice model is first evaluated, and then the LASSO model is used to assess the impact of the textual content of the companies’ websites. The dataset of the study is represented by data on 2,176 Russian metallurgical enterprises for 2020–2023, of which 57% were receiving government support. The results of the first step of evaluation showed that the organization’s age and size meas ured by revenue were the key factors associated with the probability of receiving government support. At the second step, we found that the mention of certain goods or materials on the websites, the enterprise’s location and orientation towards foreign stakeholders were significantly related to the likelihood of getting state support. The study offers tools for analyzing the determi nants of government support for SMEs and highlights the importance of this analysis for more efficient allocation of government resources.

CSTAN: A Deepfake Detection Network with CST Attention for Superior Generalization

With the advancement of deepfake forgery technology, highly realistic fake faces have posed serious security risks to sensor-based facial recognition systems. Recent deepfake detection models mainly use binary classification models based on deep learning. Despite achieving high detection accuracy on intra-datasets, these models lack generalization ability when applied to cross-datasets. We propose a deepfake detection model named Channel-Spatial-Triplet Attention Network (CSTAN), which focuses on the difference between real and fake features, thereby enhancing the generality of the detection model. To enhance the feature-learning ability of the model for image forgery regions, we have designed the Channel-Spatial-Triplet (CST) attention mechanism, which extracts subtle local information by capturing feature channels and the spatial correlation of three different scales. Additionally, we propose a novel feature extraction method, OD-ResNet-34, by embedding ODConv into the feature extraction network to enhance its dynamic adaptability to data features. Trained on the FF++ dataset and tested on the Celeb-DF-v1 and Celeb-DF-v2 datasets, the experimental results show that our model has stronger generalization ability in cross-datasets than similar models.

Predictive accuracy of the logit model to determine factors affecting delivery and postnatal care services utilization in Tanzania

PurposeThis paper examines factors affecting the delivery care (DC) and postnatal care (PNC) services utilization in Tanzania using the predictive accuracy of the binary logit model by including DC as among the predictors of PNC utilization.Design/methodology/approachThe study used secondary data from the Tanzania Demographic and Health Survey and Malaria Indicator Survey (2022) based on a sample of 13,266 women aged 15–49. Binary logit was used to analyze the association between independent variables and each binary dependent variable, followed by a stepwise likelihood ratio test for binary logistic regression analysis to select the most important predictors associated with DC and PNC. Odds ratios were used to predict the likelihood of the occurrence of significant explanatory variables in explaining the outcomes.FindingsThe results showed that the type of place of residence, wealth index, region, highest educational level, husband/partner’s level of education and respondent’s occupation were significant predictors of DC utilization. In contrast, the woman’s age, region, highest educational level, wealth index, DC and husband/partner’s educational level were significant predictors of PNC utilization. Public health interventions are needed to promote and increase the utilization of delivery and postal care services. The targets should be poor women and those residing in rural or remote areas.Originality/valueIncluding DC as a predictor variable (which was not used before) in PNC utilization improved the predictive power of the model and the robustness of the results.

Water status assessment in grapevines using plant electrophysiology

Traditional methods for assessing vine water status, such as the Scholander pressure chamber, are time-consuming, punctual and labour-intensive. The development of alternative methods which are accurate, reliable and can provide real-time information on vine water status is a necessity for farmers all over the world. This study proposes the use of plant electrophysiology as a novel approach for real-time water status assessment in grapevines. We conducted four climate chamber experiments with potted grapevines under different irrigation regimes. Various morphological and physiological assessments were performed in parallel with electrophysiological measurements to correlate classic water status assessment methods with plant electrophysiological signals. Two machine learning approaches based on classification and regression were employed to train the prediction models. Results obtained from both models indicate significant differences in irrigation status between well-watered and water-deficit plants, with the latter showing reduced growth and physiological activity, confirming the water stress status of the plant. While the binary classification model successfully differentiates between well-watered and water-deficit plants, its practical use is limited. Therefore, a regression model was developed to directly predict predawn leaf water potential. To the best of our knowledge, this is the first time that electrical signals are correlated with vine water potential measurements. The findings presented here thus provide a promising new tool for future real-time and remote monitoring of vine water status to manage irrigation and adapt agronomic strategies. Nevertheless, validation and optimisation of the models are still necessary, particularly under field conditions.

Cannabis legalization and changes in cannabis and tobacco/nicotine use and co-use in a national cohort of U.S. adults during 2017–2021

BackgroundLittle is known about whether cannabis legalization impacts cannabis use uptake or has spillover effects on co-use of cannabis and tobacco/nicotine (using both in the past 30 days). We determined associations of cannabis legalization with self-reported (1) current (past 30-day) cannabis use; (2) current (“now”) tobacco/nicotine use (smoking or electronic cigarette use); and (3) current co-use of cannabis and tobacco/nicotine and how prevalence is changing over time. MethodsIn this longitudinal study, a web-based survey was administered to a nationally representative, population-based panel of US adults in 2017, 2020, and 2021. We used weighted unadjusted binomial logistic GEE models to assess changes in prevalence of cannabis, tobacco/nicotine use and co-use and weighted, adjusted binary logistic GEE models to assess associations of cannabis legalization with cannabis, tobacco/nicotine use and co-use. ResultsA total of 9003 participants (age range = 18–94, mean age = 47.9 [±17.4 SD] years; 4696 females [weighted 52.0 %]) completed the survey in 2017; 5979/8529 (70.1 %) in 2020 and 5420/7305 (74.2 %) in 2021 from the original cohort who remained available. Current cannabis use significantly increased +3.3 % between 2017 and 2021, while tobacco/nicotine use significantly declined (−1.9 %); co-use of cannabis and tobacco/nicotine did not change significantly (+0.2 %). Both medical and recreational cannabis legalization was associated with increased current cannabis use; the independent effect of recreational cannabis legalization was 1.13 times larger than medical. There were no statistically significant differences in tobacco/nicotine use and co-use prevalence by legalization status. ConclusionCannabis legalization increases cannabis use but is not associated with changes in tobacco/nicotine use or co-use. Legalization should be coupled with public health efforts.

Joint Observations of Late Universe Probes: Cosmological Parameter Constraints from Gravitational Wave and Type Ia Supernova Data

The growing tensions between the early Universe and the late Universe increasingly highlight the importance of developing precise probes for late cosmology. As significant late-Universe probes, Type Ia supernovae (SNe Ia) and gravitational waves (GWs) can provide measurements of relative and absolute distances, respectively. Their complementary nature is likely to break the degeneracies among cosmological parameters, thereby yielding more precise constraints. In this study, we use 43 gravitational-wave sources from the Third LIGO–Virgo–KAGRA Gravitational-Wave Transient Catalog (GWTC-3) and 1590 light curves from Pantheon+ compilation to constrain the dark energy models, as an attempt to achieve precise late-Universe cosmological constraints. For the dark siren GW event, we estimate the corresponding redshift using the binary black hole redshift distribution model. The combination of GW and SNe Ia data could provide the precision on the Hubble constant H 0 and the present matter density Ω m of approximately 16% and 8% for the ΛCDM model. If we consider the equation of state of dark energy w, the combination sample constrains the precision of w to approximately 24%. Although the combination of GWs and SNe Ia observations effectively breaks degeneracies among various cosmological parameters, yielding more stringent constraints, the precision of these constraints still does not meet the stringent standards required by precision cosmology. However, it is reasonable to anticipate that, in the near future, the joint observations of GWs and SNe Ia will become a powerful tool, particularly in the late Universe, for the precise measurement of cosmological parameters.

Analysis of pedestrian crossing violations at an overpass location by using a logit model

Jaywalking significantly increases accident risks compared to marked crosswalks. Overpasses are often installed to address this, but violations continue. This study examines pedestrian behavior at overpass locations, focusing on 438 individuals recorded on video. Factors like parked vehicles, gender, age, group size, mobile phone distraction, carrying items, the number of waiting pedestrians upon arrival, waiting times, and traffic volume were analyzed. Since jaywalking is more frequent during off-peak hours when traffic is not congested, observations were made at random intervals during these times. A binary logit model was used to identify the factors influencing jaywalking. The study found that points of interest near overpasses significantly affect pedestrian behavior. Additionally, parked vehicles near the overpass reduce crossing distance, encouraging jaywalking. Policy recommendations include implementing parking restrictions and strategically locating overpasses to align with pedestrian movement patterns and nearby points of interest.

Is technological innovation good or bad? An empirical investigation of technology startups

Digitalization has brought several unpredictable changes in the field of technology and innovation management. Despite several studies on how digital innovation stands out in terms of a firm's performance, there is an evident research gap for empirical studies on technology startups from developing countries. Therefore, this research investigates the impact of digitalization in three categories (digital interconnectedness, digitalized value chains and big data analytics) on exploratory and exploitative innovations under the moderation of absorptive capacity in technology startups of three different sizes operating in Pakistan. The study used the binary probit model and the results reveal the impact of digitalization on exploratory and exploitative innovations is heterogenous in startups of different sizes along with a significant moderating effect of absorptive capacity. The findings suggest that technological innovation plays a different role among different sizes of startups under the moderation of absorptive capacity.

KIC 8840638: A Newly Discovered Eclipsing Binary with δ Scuti–Type Oscillations

In this paper, we analyze the light variation of KIC 8840638 using high-precision time-series data from the Kepler mission. The analysis reveals that this target is a new detached eclipsing binary system with a δ Scuti component, rather than a single δ Scuti star as previously known. The frequency analysis of short-cadence data reveals 95 significant frequencies, most of which lie in a frequency range of 23−32 day−1. Among them, seven independent frequencies are detected in the typical frequency range of δ Scuti stars, and they are identified as pressure modes. In addition, a possible large separation value of Δν = 36.5 ± 0.1 μHz is also detected with the Fourier transform (FT) and autocorrelation function (AC) analysis. The orbital frequency f orb (= 0.320008 day−1) and its harmonics are also detected directly in the frequency spectrum. The binary modelings derived from PHOEBE indicate that this binary system is in detached configurations with a mass ratio of q=0.33−0.04+0.06 , an inclination angle of 40.19−2.84+3.96 ​​​​​°. The derived parameters and binary evolutionary model suggest that the primary star is an object on the verge of leaving the main sequence with a temperature of ∼7600 K, while the secondary appears to be a cool component entering the giant branch with a temperature ∼3100 K lower than the primary. Moreover, this system may have undergone a mass ratio reversal, where the more massive star is the gainer component and the less massive one is the donor star.

A comprehensive study of five candidate δ Scuti-type pulsators in detached eclipsing binaries

Context. Pulsating stars in eclipsing binaries (EBs) provide an excellent opportunity to obtain precise, model-independent stellar parameters for studying these oscillations in detail. One of the most common classes of pulsators found in such EBs exhibits δ Scuti-type oscillations. Characterising these pulsators using the precise stellar parameters obtained using EB modelling can help us better understand such stars, and provide strong anchors for asteroseismic studies. Aims. We performed a comprehensive photometric and spectroscopic analysis of candidate pulsators in detached EBs, to add to the sample of such systems with accurately determined absolute parameters. Methods. We performed radial velocity and light curve modelling to estimate the absolute stellar parameters, and detailed spectroscopic modelling to obtain the global metallicity and temperatures. Frequency power spectra were obtained using residuals from binary modelling. Finally, we used isochrones to determine the age of the stars, and compared the estimated physical parameters to the theoretically obtained values. Results. We present a detailed analysis of four candidate δ Scuti-type pulsators in EBs, and update the light curve analysis of the previously studied system TIC 308953703. The masses and radii of components are constrained to a high accuracy, which helps us constrain the age of the systems. We perform a Fourier analysis of the observed oscillations, and try to explain their origin. For TIC 81702112, we report tidal effects causing amplitude variation in the oscillation frequencies over the orbital phase.

The adoption non-adoption dichotomy: Why do smallholder producers dis-adopt improved chicken breeds?

Adopting agricultural technologies is crucial to improve productivity and livelihoods in developing countries. While much research has focused on adoption decisions, understanding dis-adoption, when farmers stop using technology, is equally important. Studies on agricultural technology adoption often treat dis-adopters (those who initially adopted but later discontinued to use) and never-adopters (those who never adopted) as the same, using binary models to analyze farmers' decisions. We argue that a better understanding of these decisions can be achieved by separately analyzing 'never-adoption', 'dis-adoption', and 'adoption.' Using nationally representative data from three African countries, Ethiopia, Nigeria, and Tanzania, we developed a multinomial logit model to analyze the adoption of improved chicken breeds. Our findings show that dis-adopters of improved chicken are different from never-adopters. Factors associated with dis-adoption include gender and education of household heads, access to training and extension services, breeding and culling practices, access to markets, use of complementary inputs, production objectives, landholding size, income diversity, and access to finance. Policies and strategies that aim to enhance sustained adoption and use of improved chicken breeds should promote a bundle of technologies, including tailored training, women empowerment, locally adapted and farmer-preferred chicken breeds, complementary inputs and services, innovative marketing strategies, and delivery models for bundles of technologies.

Spin–Orbit Alignment of Early-type Astrometric Binaries and the Origin of Slow Rotators

The spin–orbit alignment of binary stars traces their formation and accretion history. Previous studies of spin–orbit alignment have been limited to small samples, slowly rotating solar-type stars, and/or wide visual binaries that not surprisingly manifest random spin–orbit orientations. We analyze 917 Gaia astrometric binaries across periods P = 100–3000 days (a = 0.5–5 au) that have B8-F1 IV/V primaries (M 1 = 1.5–3 M ⊙) and measured projected rotational velocities v sin i. The primary stars in face-on orbits exhibit substantially smaller v sin i compared to those in edge-on orbits at the 6σ level, demonstrating significant spin–orbit alignment. The primaries in our astrometric binaries are rotating more slowly than their single-star or wide-binary counterparts and therefore comprise the slow-rotator population in the observed bimodal rotational velocity distribution of early-type stars. We discuss formation models of close binaries where some of the disk angular momentum is transferred to the orbit and/or secondary spin, quenching angular momentum flow to the primary spin. The primaries in astrometric binaries with small mass ratios q = M 2/M 1 < 0.3 possess even smaller v sin i, consistent with model predictions. Meanwhile, astrometric binaries with large eccentricities e > 0.4 do not display spin–orbit alignment or spin reduction. Using a Monte Carlo technique, we measure a spin–orbit alignment fraction of F align = 75% ± 5% and an average spin reduction factor of 〈S align〉 = 0.43 ± 0.04. We conclude that 75% of close A-type binaries likely experienced circumbinary disk accretion and probably formed via disk fragmentation and inward disk migration. The remaining 25%, mostly those with e > 0.4, likely formed via core fragmentation and orbital decay via dynamical friction.

Stripped Helium Star and Compact Object Binaries in Coeval Populations: Predictions Based on Detailed Binary Evolution Models

Massive stars mainly form in close binaries, where their mutual interactions can profoundly alter their evolutionary paths. Evolved binaries consisting of a massive OB-type main-sequence star with a stripped helium star or a compact companion represent a crucial stage in the evolution toward double compact objects, whose mergers are (potentially) detectable via gravitational waves. The recent detection of X-ray-quiet OB+black hole binaries and OB+stripped helium star binaries has set the stage for discovering more of these systems in the near future. In this work, based on 3670 detailed binary-evolution models and using empirical distributions of initial binary parameters, we compute the expected population of such evolved massive binaries in coeval stellar populations, including stars in star clusters and in galaxies with starburst activities, for ages up to 100 Myr. Our results are vividly illustrated in an animation that shows the evolution of these binaries in the color–magnitude diagram over time. We find that the number of OB+black hole binaries peaks around 10 Myr, and OB+neutron star binaries are most abundant at approximately 20 Myr. Both black holes and neutron stars can potentially be found in populations with ages up to 90 Myr. Additionally, we analyze the properties of such binaries at specific ages. We find that OB+helium stars and OB+black hole binaries are likely to be identifiable as single-lined spectroscopic binaries. Our research serves as a guide for future observational efforts to discover such binaries in young star clusters and starburst environments.

Factors Affecting Food Security of Expropriated Peri-Urban Households in Ethiopia: The Case of the East Gojjam Administrative Zone

Food insecurity in peri-urban areas is exacerbated by high living costs, limited access to healthy food, and economic inequality. Despite its growing prevalence due to factors like land loss from urban expansion, food insecurity has received limited attention. In Ethiopia, drastic urbanization creates competition for land between agriculture and urban development, which is becoming extreme in peri-urban areas., This study aims to assess the impact of urban expansion on food security among expropriated peri-urban households in Ethiopia. Using a calorie intake-based food security line, we categorize respondents as food-secure or -insecure. The research analyzed data from 350 expropriated peri-urban households through a structured questionnaire, employing descriptive statistics and binary logistic regression. About 67% of the respondents were classified as food-insecure. The binary logit model identified several significant determinants of food security, including compensation amount and type, household head demographics, agricultural land rent participation, irrigation and credit access, off-farm income, and organic fertilizer use. To address food insecurity among expropriated households, policymakers must prioritize these factors. Government attention and policy consideration are crucial to ensure the well-being of these vulnerable populations.

AGCNAF: predicting disease-gene associations using GCN and multi-head attention to fuse the similarity features

Abstract Association prediction between diseases and genes is a critical step in revealing the molecular mechanisms of diseases and developing drug treatment strategies. With the explosive growth of data in the biomedical field, how to effectively utilize these data for accurate prediction has become a hotspot and challenge in current research. To overcome the limitations of current prediction methods in dealing with complex biological network structures and feature extraction, this study proposes AGCNAF, a method that combines an unsupervised Graph Convolutional Network (GCN) and a multi-head attention mechanism. The metagraph-guided random walk strategy enables AGCNAF to capture local and high-order topological structures in the graph, while GCN is responsible for realizing deep feature extraction of these structures. By incorporating similarity features through the multi-attention mechanism, AGCNAF achieves effective integration of global and local features, which significantly improves the prediction performance. By incorporating similarity features through the multi-attention mechanism, AGCNAF achieves effective integration of global and local features, which significantly improves the prediction performance. By utilizing the machine learning binary classification model for prediction, the experimental results through five-fold cross-validation show that AGCNAF demonstrates significant advantages in prediction performance compared to existing methods, with its AUC and AUPR reaching 0.9686 and 0.9709, respectively, and the AUC up to 0.9812 under specific conditions. &amp;#xD;To verify the practical application value of AGCNAF, this study also conduct case studies on Alzheimer's disease, lung cancer, and breast cancer. The results further confirm the excellent performance of AGCNAF in identifying potential disease-gene associations, which opens up new possibilities for future disease-gene research.

Mildly elevated left ventricular outflow tract gradients are associated with increased short term mortality in inpatients without septal hypertrophy

Abstract Background The significance of left ventricular outflow tract gradients (LVOTG) on the clinical outcomes of hypertrophic cardiomyopathy (HCM) has been well studied. An LVOTG of ≥ 30 mmHg is considered significant. However, mildly elevated LVOTG at rest in individuals without basal septal hypertrophy is sometimes encountered in the echocardiography laboratory and data on its association with survival in this group is lacking. Purpose Here we explore the association between LVOTG and short term all cause mortality in inpatients without basal septal hypertrophy. Methods We retrospectively reviewed 47,866 consecutive inpatients who had an echocardiography study during their hospital stay between March 2010 and February 2021 . After excluding cases with incomplete data (10,882) and patients with basal septal hypertrophy &amp;gt; 12 mm (9,486), we remained with 27,498 patients. Demographic, baseline clinical and echocardiographic characteristics were derived from electronic records. Mortality data was validated against the ministry of health national records. A binary logistic regression model for predicting 30-day mortality according to LVOTG dichotomized at each point was used. Then we divided the cohort into 4 LVOTG groups (0-6, 6-10, 10-20 and &amp;gt; 20 mmHg) and the same regression model was used to compare between the groups. Results The mean age in our cohort was 66 ± 17 years and 54% were male. The mean LVOTG was 4.2 ± 2.4 mmHg (Median 4 (IQR=3-5) and the mean left ventricular ejection fraction was 55 ± 9%. A binary logistic regression model showed that when different LVOTG cut-points were used, even a mildly elevated LVOTG predicted short-term survival in this population (figure 1). We then divided our cohort into 4 groups based on LVOTG. There were 24,703 (90%) patients in group 1, 2549 (9%) in group 2, 201 (1%) in group 3 and 25 (0.1%) individuals in group 4. Patients in groups 1 and 2 were younger, had higher hemoglobin levels, lower C-reactive protein (CRP) levels, mildly lower septal thickness, and lower left ventricular ejection fraction compared to the groups with higher LVOTG (table 1). We then applied the same multivariate binary regression model to predict short term mortality with group 1 as a reference. The hazard ratio (HR) for group 2 was 1.3 (CI 95%: 1-1.6, p=0.03), for group 3 HR=2.8 (CI 95%: 1.6-5.1, p&amp;lt;0.001) and for group 4 HR=2.8 (CI 95%: 0.6-12.8, p=0.18). Conclusions We have shown in a large cohort of inpatients without basal septal hypertrophy that a mildly elevated rest LVOTG (6-10 and 10-20 mmHg) predicted all cause 30-day mortality. Patients in the elevated LVOTG groups were older with lower hemoglobin and higher CRP levels, but LVOTG remained an independent predictor in a multivariate regression model. This could potentially aid early detection of at-risk populations.Mortality at different LVOTG cut-pointsBaseline characteristics

Early plant disease detection by Raman spectroscopy: An open-source software designed for the automation of preprocessing and analysis of spectral dataset

This study introduces a reliable, non-coding software named qREAD-Raman, written in the JavaScript® language, for analyzing and interpreting Raman spectral information. It is designed with a focus on the early detection of diseases in tomato plants (S. lycopersicum) during the asymptomatic stage. The platform integrates a set of machine learning algorithms necessary for the preprocessing consisting of outlier removal, baseline correction, fluorescence removal, smoothing, and normalization. For classification, we applied a Consensus of five different classifiers: Multilayer Perceptron (MLP), Partial Least Squares-Discriminant Analysis (PLS-DA), Linear Discriminant Analysis (LDA), Long Short-Term Memory (LSTM), and K-nearest neighbors (kNN). The experiments were conducted on two bacterial diseases: bacterial canker of tomato induced by Clavibacter michiganesis subsp. michiganensis (Cmm), and the tomato vein-greening associated with Candidatus Liberibacter solanacearum (CLso), a non-culturable bacteria transmitted by Bactericera cockerelli insect. Binary models (Cmm-Healthy and CLso-Healthy) demonstrated excellent classification ability. Asymptomatic Cmm-infected plants were distinguished with an accuracy of 88–95 %, while CLso-infected plants showed an accuracy of 68–77 %. The three-class model (CLso-Cmm-Healthy) exhibited acceptable performance in differentiating between Cmm and CLso, with accuracy rates of 71–83% and 58–67%, respectively. The model's performance highlights differences in the relevant spectral regions associated with the biochemical changes induced by each studied disease. The qREAD-Raman software, implemented for the purpose of this research, was found to be a valuable and comprehensive tool that effectively differentiate diseased tomato plants during their asymptomatic stage.

Impact of baseline thrombocytopenia on outcomes in hospitalizations for acute coronary syndrome and non-elective percutaneous coronary intervention: meta-analysis

Abstract Background Thrombocytopenia's impact on acute coronary syndromes (ACS) outcomes is poorly understood. A systematic review and meta-analysis aim to elucidate its influence on both short-term and long-term clinical endpoints, guiding tailored management strategies for improved patient care. Methods Our meta-analysis is registered on PROSPERO (CRD42024511832). We conducted a systematic search of PubMed and SCOPUS databases up to February 2024, utilizing keywords to identify studies on ACS or non-elective PCI populations with or without baseline thrombocytopenia (TP) reporting on in-hospital and long-term outcomes. Baseline TP was defined as a platelet count less than 150 x 10⁹/L and had to be established before hospitalization or ACS management or intervention. Binary random effects models were used to estimate pooled unadjusted odds ratios (OR) and 95% confidence intervals (CI). Sensitivity analysis was done by the leave-one-out method, and the heterogeneity was assessed with the I² statistics. Results From 2,831 identified studies, 10 were identified as eligible for inclusion in our pooled analysis and included 6,588,688 patients. Thrombocytopenia was associated with higher risks of in-hospital mortality (OR=1.89, 95% CI:1.55-2.30, p&amp;lt;0.01), in-hospital Major adverse cardiovascular events (MACE) (OR=1.99, 95% CI:1.63-2.43, p&amp;lt;0.01), and in-hospital bleeding (OR=2.62, 95% CI:1.89-3.64, p&amp;lt;0.01). Long-term follow-up revealed increased risks of MACE (OR=1.80, 95% CI 1.49-2.18, p&amp;lt;0.01), bleeding (OR=1.63, 95% CI:1.63-2.25, p&amp;lt;0.01), and cardiovascular mortality (OR=2.07, 95% CI:1.76-2.42, p&amp;lt;0.01) with thrombocytopenia. No significant difference was found in in-hospital stroke outcomes and long-term outcomes for stroke and recurrent myocardial infarction between thrombocytopenic and non-thrombocytopenic groups. Conclusion Thrombocytopenia worsens ACS outcomes. Vigilance and tailored management are crucial for mitigating its impact on patient prognosis. Further research is needed for better interventions.Secondary outcomes forest plot