Variable Importance Research Articles

518. ARTIFICIAL INTELLIGENCE MODELS VS CLINICIAN PERSPECTIVES OF IMPORTANT FACTORS IN OESOPHAGEAL CANCER DECISION-MAKING

Abstract Background Decision-making for oesophageal cancer (OC) management remains multifaceted and nuanced in the face of a traditionally complex patient demographic. The evolving movement of artificial intelligence (AI) offers the potential of an imminent paradigm shift in traditional multidisciplinary team (MDT) workflow with Machine Learning models proving capable of encapsulating much of that decision-making framework. Clinician buy-in however stems from trust that such models accurately represent the perceptions and considerations of the MDT in its current form. Methods A mixed methodology study combining a national survey of United Kingdom OC MDT clinicians (sent to the British Society of Gastroenterologists, Association of Upper Gastrointestinal Surgery and UK& Ireland Oesophagogastric Cancer group) was conducted and compared with a Machine Learning (ML) based Random Forests model of treatment-decisions from a leading UK tertiary referral centre. The factors influencing OC treatment decisions from a human perspective were contrasted with variable importance from the AI approach. Human perceptions of barriers and attractions of AI-driven support tools were also explored to understand how human-AI collaboration could be facilitated in future. Results The National Survey which generated 67 responses found age and gender to play a more significant role in the ML model than they did in the clinicians' conscious decision-making process. Additionally, thematic analysis identified a variety of important factors, including molecular tests, symptoms, social circumstances, and nutritional status. The prospect of utilising AI-based decision support in the future received generally positive feedback, although opinions varied widely. Challenges to its adoption include perceptions of clinician superiority, the uniqueness of patient cases, the need for transparency and safeguards in the model, the requirements for data input, and the demand for proven effectiveness. Conclusions Comparison with the model highlighted demographic factors which appear to heavily influence treatment but were consciously deemed less important by clinicians. While overall, the response was positive towards an ML tool, some respondents felt very unwilling to use one. This could limit its adoption by MDTs unless decision-tools are developed to address such concerns around the barriers identified.

Spatial lipidomic profiles of atherosclerotic plaques: A mass spectrometry imaging study

Lipids contribute to atherosclerotic cardiovascular disease but their roles are not fully understood. Spatial lipid composition of atherosclerotic plaques was compared between species focusing on aortic plaques from New Zealand White rabbits and carotid plaques from humans (n = 3), using matrix-assisted laser desorption/ionization mass spectrometry imaging. Histologically discriminant lipids within plaque features (neointima and media in rabbits, and lipid-necrotic core and fibrous cap/tissue in humans) included sphingomyelins, phosphatidylcholines, and cholesteryl esters. There were 67 differential lipids between rabbit plaque features and 199 differential lipids in human, each with variable importance in projection score ≥1.0 and p < 0.05. The lipid profile of plaques in the rabbit model closely mimicked that of human plaques and two key pathways (impact value ≥ 0.1), sphingolipid and glycerophospholipid metabolism, were disrupted by atherosclerosis in both species. Thus, mass spectrometry imaging of spatial biomarkers offers valuable insights into atherosclerosis.

Pooling and winsorizing machine learning forecasts to predict stock returns with high-dimensional data

We evaluate US market return predictability using a novel data set of several hundred ag- gregated firm-level characteristics. We apply LASSO, Elastic Net, Random Forest, Neural Net, Extreme Gradient Boosting, and Light Gradient Boosting Machine methods and find these models experience large prediction errors that lead to forecast failures. However, winsorizing and pooling machine learning model forecasts provides consistent out-of-sample predictability. To assess robustness, we apply machine learning methods to high-dimensional data for Canada, China, Germany and the UK as well as the Goyal–Welch data. All machine learning models we consider, except for the ensemble pooled methods, fail to significantly predict returns across our samples, highlighting the importance of pooling, evaluating additional economies, and the fragility of individual machine learning methods. Our results shed light on the sparsity versus density debate as the degree of sparsity and variable importance evolves over time.

Prediction of perceived annoyance caused by an electric drone noise through its technical, operational, and psychoacoustic parameters.

Electric drones serve diverse functions, including delivery and surveillance. Nonetheless, they encounter significant challenges due to their annoying noise emissions. To address this issue, a sound database was created from experiments conducted in a hover-test-bench and real flights operated indoors. These experiments involved a wide range of parameter variations and operational conditions. A global digital user study involving 578 participants was conducted to assess drone noise annoyance. Furthermore, correlations between annoyance levels, psychoacoustic metrics, sociocultural factors, and technical/operational parameters were analyzed. The effects of implementing acoustic optimization modifications on the drone's performance were quantified with a conceptual design tool. The findings indicate that reducing the levels of loudness, sharpness, tonality, and roughness or fluctuation strength led to an improvement in annoyance. Differences in variable importance of psychoacoustic metrics dependent on the specific model were found. Sociocultural factors did not affect annoyance. Technical and operational parameters impacted annoyance, especially when reducing blade tip speed. A 20% reduction in tip speed showed potential through tool application as it maintained acceptable drone performance while beneficially targeting annoyance. A multi-disciplinary optimization is recommended to maintain operational efficiency. Last, psychoacoustic metrics were validated as an effective measure to evaluate a design solution.

Aboveground Biomass Mapping in SemiArid Forests by Integrating Airborne LiDAR with Sentinel-1 and Sentinel-2 Time-Series Data

Aboveground biomass (AGB) is a vital indicator for studying carbon sinks in forest ecosystems. Semiarid forests harbor substantial carbon storage but received little attention due to the high spatial–temporal heterogeneity that complicates the modeling of AGB in this environment. This study assessed the performance of different data sources (annual monthly time-series radar was Sentinel-1 [S1]; annual monthly time series optical was Sentinel-2 [S2]; and single-temporal airborne light detection and ranging [LiDAR]) and seven prediction approaches to map AGB in the semiarid forests on the border between Gansu and Qinghai Provinces in China. Five experiments were conducted using different data configurations from synthetic aperture radar backscatter, multispectral reflectance, LiDAR point cloud, and their derivatives (polarimetric combination indices, texture information, vegetation indices, biophysical features, and tree height- and canopy-related indices). The results showed that S2 acquired better prediction (coefficient of determination [R2]: 0.62–0.75; root mean square error [RMSE]: 30.08–38.83 Mg/ha) than S1 (R2: 0.24–0.45; RMSE: 47.36–56.51 Mg/ha). However, their integration further improved the results (R2: 0.65–0.78; RMSE: 28.68–35.92 Mg/ha). The addition of single-temporal LiDAR highlighted its structural importance in semiarid forests. The best mapping accuracy was achieved by XGBoost, with the metrics from the S2 and S1 time series and the LiDAR-based canopy height information being combined (R2: 0.87; RMSE: 21.63 Mg/ha; relative RMSE: 14.45%). Images obtained during the dry season were effective for AGB prediction. Tree-based models generally outperformed other models in semiarid forests. Sequential variable importance analysis indicated that the most important S1 metric to estimate AGB was the polarimetric combination indices sum, and the S2 metrics were associated with red-edge spectral regions. Meanwhile, the most important LiDAR metrics were related to height percentiles. Our methodology advocates for an economical, extensive, and precise AGB retrieval tailored for semiarid forests.

Determinants and prediction of hypertension among Chinese middle-aged and elderly adults with diabetes: A machine learning approach

ObjectiveMultimorbidity, particularly diabetes combined with hypertension (DCH), is a significant public health concern. Currently, there is a gap in research utilizing machine learning (ML) algorithms to predict hypertension risk in Chinese middle-aged and elderly diabetic patients, and gender differences in DCH comorbidity patterns remain unclear. We aimed to use ML algorithms to predict DCH and identify its determinants among middle-aged and elderly diabetic patients in China. Study designCross-sectional study. MethodsData were collected on 2775 adults with diabetes aged ≥45 years from the 2015 China Health and Retirement Longitudinal Study. We employed nine ML algorithms to develop prediction models for DCH. The performance of these models was evaluated using the area under the curve (AUC). Additionally, we conducted variable importance analysis to identify key determinants. ResultsOur results showed that the best prediction models for the overall population, men, and women were extreme gradient boosting (AUC = 0.728), light gradient boosting machine (AUC = 0.734), and random forest (AUC = 0.737), respectively. Age, waist circumference, body mass index, creatinine level, triglycerides, taking Western medicine, high-density lipoprotein cholesterol, blood urea nitrogen, total cholesterol, low-density lipoprotein cholesterol, and sleep disorders were identified as common important predictors by all three populations. ConclusionsML algorithms showed accurate predictive capabilities for DCH. Overall, non-linear ML models outperformed traditional logistic regression for predicting DCH. DCH predictions exhibited variations in predictors and model accuracy by gender. These findings could help identify DCH early and inform the development of personalized intervention strategies.

Imaging Spectroscopy‐Based Estimation of Aboveground Biomass in Louisiana's Coastal Wetlands: Toward Consistent Spectroscopic Retrievals Across Atmospheric States

AbstractDeveloping accurate landscape‐scale aboveground biomass (AGB) maps is critical to understanding coastal deltaic wetland resilience, as AGB influences stability and elevation dynamics in herbaceous wetlands. Here we used AVIRIS‐NG imaging spectrometer (or “hyperspectral”) data from NASA's 2021 Delta‐X mission in coastal Louisiana to map seasonal changes in herbaceous AGB across two deltaic basins with contrasting sediment delivery and hydrologic regimes: the Atchafalaya (active) and Terrebonne (inactive). We assessed the impact of atmospheric effects on our retrievals, as high water vapor content in August 2021 caused significant noise in the 880–1,000 and 1,080–1,200 nm near‐infrared (NIR) wavelengths. We hypothesized that correcting these wavelengths with our conditional Gaussian interpolation algorithm would improve AGB estimates due to their association with plant canopy water content. We empirically assessed the performance of the corrected spectra on AGB estimates using Partial Least Squares Regression (PLSR), finding that the corrected NIR bands attained high variable importance and reduced estimation errors. Our Random Forest regression approach based on the corrected spectra attained equivalent error metrics via leave‐one‐out‐cross‐validation as the PLSR models (R2 = 0.43, mean absolute error = 257.3 g/m2) while greatly improving the AGB maps' visual quality, having better captured variability while reducing noise and discontinuities in AGB estimates across flightlines. The maps show differing seasonal growth, with the Atchafalaya and Terrebonne Basins' AGB increasing from means of 4.3–9.4 and 4.6–8.9 Mg/ha, respectively. We demonstrated that imaging spectroscopy can be applied to assess herbaceous biomass stocks, growth patterns, and resilience in coastal ecosystems.

Integration of Patient Reported Quality-of-life Data into Risk Assessment in Heart Failure

Optimal management of outpatients with heart failure (HF) requires serially updating the estimates of their risk for adverse clinical outcomes to guide treatment. Patient-reported outcomes (PROs) are becoming increasingly used in clinical care. The purpose of this study was to determine whether the inclusion of PROs can improve the risk prediction for HF hospitalization and death in ambulatory patients with HF. We included consecutive patients with HF with reduced ejection fraction (HFrEF) and HF with preserved EF (HFpEF) seen in a HF clinic between 2015 and 2019 who completed PROs as part of routine care. Cox regression with a least absolute shrinkage and selection operator regularization and gradient boosting machine analyses were used to estimate risk for a combined outcome of HF hospitalization, heart transplant, left ventricular assist device implantation, or death. The performance of the prediction models was evaluated with the time-dependent concordance index (Cτ). Among 1165 patients with HFrEF (mean age 59.1 ± 16.1, 68% male), the median follow-up was 487 days. Among 456 patients with HFpEF (mean age 64.2 ± 16.0 years, 55% male) the median follow-up was 494 days. Gradient boosting regression that included PROs had the best prediction performance - Cτ 0.73 for patients with HFrEF and 0.74 in patients with HFpEF, and showed very good stratification of risk by time to event analysis by quintile of risk. The Kansas City Cardiomyopathy Questionnaire overall summary score, visual analogue scale and Patient Reported Outcomes Measurement Information System dimensions of satisfaction with social roles and physical function had high variable importance measure in the models. PROs improve risk prediction in both HFrEF and HFpEF, independent of traditional clinical factors. Routine assessment of PROs and leveraging the comprehensive data in the electronic health record in routine clinical care could help more accurately assess risk and support the intensification of treatment in patients with HF.

Effect of hot-air drying processing on the volatile organic compounds and maillard precursors of Dictyophora Rubrovalvata based on GC-IMS, HPLC and LC-MS

The dynamic changes in volatile organic compounds (VOCs), reducing sugars, and amino acids of Dictyophora rubrovalvata (DR) at various drying temperatures were analyzed using GC-IMS, HPLC, and LC-MS. Orthogonal partial least squares discriminant analysis (OPLS-DA) combined with VOCs indicated that drying temperature of 80 °C was optimal. Variable importance in the projection (VIP) and relative odor activity value (ROAV) were employed to identify 22 key VOCs. The findings suggested that esters played a predominant role among the VOCs. Pearson correlation analysis revealed that serine (Ser), glutamine (Gln), lysine (Lys), alanine (Ala), threonine (Thr), glutamic acid (Glu), asparagine (Asn), ribose, and glucose were closely associated with the formation of esters, aldehydes, ketones, pyrimidines, and pyrazines. In conclusion, this study laid a foundational theory for elucidating the characteristics aroma substances and their production pathways, providing a valuable reference for analysing the flavor characteristics of DR.

Effect of weight loss interventions on metabolomic signatures in obese children with insulin resistance

The obesity epidemic among children has become a major public health issue, and the presence of childhood insulin resistance (IR) has been demonstrated prior to the onset of type 2 diabetes mellitus. However, it is unclear whether the metabolomic signature is associated with weight loss interventions in obese children with IR. Thirty-six obese children with IR were selected from the weight loss camp (Shenzhen Sunshine Xing Yada health Technology Co., LTD). Clinical parameters were collected before and after weight loss intervention. Targeted metabolomics of plasma samples was performed by ultra-performance liquid chromatography coupled to the tandem mass spectrometry, and principal component analysis, variable importance in projection, and orthogonal partial least squares discriminant analysis were used to obtain the differentially expressed metabolites. Pathway analysis was conducted with the Homo sapiens (HSA) sets in the Kyoto Encyclopedia of Genes and Genomes. We used machine learning algorithms to obtain the potential biomarkers and Spearman correlation analysis to clarify the association between potential biomarkers and clinical parameters. We found that clinical parameters and metabolite clusters were significantly changed in obese children with IR before and after weight loss intervention. Mechanistically, weight loss intervention significantly changed 61 metabolites in obese children with IR. Furthermore, 12 pathways were significantly changed. Moreover, the machine learning algorithm found 6 important potential biomarkers. In addition, these potential biomarkers were strongly associated with major clinical parameters. These data indicate different metabolomic profiles in obese children with IR after weight loss intervention, providing insights into the clinical parameters and metabolite mechanisms involved in weight loss programs.

Differences in flavor characteristics of rounds of Sauce-based Baijiu based on GC-IMS, GC-MS, and electronic tongue

The volatile flavor and metabolic pathway differences of rounds of Sauce-based Baijiu (RSBs, R-1∼R-7) were analyzed using HS-GC-IMS and HS-SPME-GC-MS. Firstly, 18 (R-1, R-2, R-4, R-6, and R-7), 17 (R-3), and 19 (R-5) characteristic flavor compounds were identified based on GC-IMS by the relative odor activity value (ROAV) analysis. Secondly, a total of 751 metabolites were identified using GC-MS, and the differential metabolites were screened for pathway enrichment analysis by variable importance in projection (VIP) > 1 combined with fold change (FC) ≥ 2 or ≤ 0.5. The results demonstrated that unsaturated fatty acid metabolism, fatty acid metabolism, terpene metabolism, and secondary metabolite biosynthesis pathways were the most significant metabolic pathways influencing the flavor of basic baijiu, which primarily resulted in alterations in the composition and concentration of metabolites, including fatty acids and terpenes. Furthermore, RSBs demonstrated distinctive taste attributes, the addition of R-1 was found to markedly influence the astringency of baijiu (p < 0.05), while no notable impact was observed on the acidity or aftertaste. This study elucidates the flavor characteristics of RSBs, which provides a reference for the flavor quality regulation and blending of baijiu.

Identifying Key Markers for Monofloral (Eucalyptus, Rosemary, and Orange Blossom) and Multifloral Honey Differentiation in the Spanish Market by UHPLC-Q-Orbitrap-High-Resolution Mass Spectrometry Fingerprinting and Chemometrics.

Honey differentiation based on the botanical origin is crucial to guarantee product authenticity, especially considering the increasing number of fraud cases. This study assessed the metabolomic differences arising from various botanical origins in honey products sold in Spanish markets, focusing on two goals: (1) discrimination within monofloral samples (eucalyptus, rosemary, and orange blossom honey) and (2) differentiation between multifloral vs. monofloral honey samples. An omics strategy based on ultra-high-performance liquid chromatography coupled with quadrupole-Orbitrap-high-resolution mass spectrometry (UHPLC-Q-Orbitrap-HRMS) was applied for the reliable identification of specific honey markers selected by orthogonal partial least squares discriminant analysis (OPLS-DA) (R2Y = 0.929-0.981 and Q2 = 0.868-0.952), followed by the variable importance in projection (VIP) approach. Key amino acid, alkaloid, and trisaccharide markers were identified to distinguish between honey samples. Some Amadori compounds were highlighted as eucalyptus honey markers, suggesting their potential use for honey aging and botanical origin differentiation. L-phenylalanine and raffinose were markers of rosemary honey. Four markers (e.g., trigonelline, L-isoleucine, and N-(1-deoxy-1-fructosyl)isoleucine) were found in higher levels in multifloral samples, indicating a greater availability of amino acids, potentially increasing the Maillard reaction. This research is the first to address the botanical origin's impact on honey by identifying novel markers not previously described.

Model-agnostic variable importance for predictive uncertainty: an entropy-based approach

In order to trust the predictions of a machine learning algorithm, it is necessary to understand the factors that contribute to those predictions. In the case of probabilistic and uncertainty-aware models, it is necessary to understand not only the reasons for the predictions themselves, but also the reasons for the model’s level of confidence in those predictions. In this paper, we show how existing methods in explainability can be extended to uncertainty-aware models and how such extensions can be used to understand the sources of uncertainty in a model’s predictive distribution. In particular, by adapting permutation feature importance, partial dependence plots, and individual conditional expectation plots, we demonstrate that novel insights into model behaviour may be obtained and that these methods can be used to measure the impact of features on both the entropy of the predictive distribution and the log-likelihood of the ground truth labels under that distribution. With experiments using both synthetic and real-world data, we demonstrate the utility of these approaches to understand both the sources of uncertainty and their impact on model performance.

Genotypic variation in blueberry flower morphology and nectar reward content affects pollinator attraction in a diverse breeding population

BackgroundPollination is crucial to obtaining optimal blueberry yield and fruit quality. Despite substantial investments in seasonal beekeeping services, blueberry producers consistently report suboptimal pollinator visitation and fruit set in some cultivars. Flower morphology and floral rewards are among the key factors that have shown to contribute to pollinator attraction, however little is known about their relative importance for improving yield in the context of plant breeding. Clarifying the relationships between flower morphology, nectar reward content, pollinator recruitment, and pollination outcomes, as well as their genetic components, can inform breeding priorities for enhancing blueberry production. In the present study, we measured ten flower and nectar traits and indices of successful pollination, including fruit set, seed count, and fruit weight in 38 southern highbush blueberry genotypes. Additionally, we assessed pollinator visitation frequency and foraging behavior over two growing seasons. Several statistical models were tested to optimize the prediction of pollinator visitation and pollination success, including partial least squares, BayesB, ridge-regression, and random forest.ResultsRandom forest models obtained high predictive abilities for pollinator visitation frequency, with values of 0.54, 0.52, and 0.66 for honey bee, bumble bee, and total pollinator visits, respectively. The BayesB model provided the most consistent prediction of fruit set, fruit weight, and seed set, with predictive abilities of 0.07, -0.08, and 0.42, respectively. Variable importance analysis revealed that genotypic differences in nectar volume had the greatest impact on honey bee and bumble bee visitation, although preferences for flower morphological traits varied depending on the foraging task. Flower density was a major driving factor attracting nectar-foraging honey bees and bumble bees, while pollen-foraging bumble bees were most influenced by flower accessibility, specifically corolla length and the length-to-width ratio.ConclusionsHoney bees comprised the majority of pollinator visits, and were primarily influenced by nectar volume and flower density. Corolla length and the length-to-width ratio were also identified as the main predictors of fruit set, fruit weight, seed count, as well as pollen-foraging bumble bee visits, suggesting that these bees and their foraging preferences may play a pivotal role in fruit production. Moderate to high narrow-sense heritability values (ranging from 0.30 to 0.77) were obtained for all floral traits, indicating that selective breeding efforts may enhance cultivar attractiveness to pollinators.

Improving genomic prediction of rhizomania resistance in sugar beet (Beta vulgaris L.) by implementing epistatic effects and feature selection

Background Rhizomania counts as the most important disease in sugar beet Beta vulgaris L. for which no plant protection is available, leaving plant breeding as the only defence strategy at the moment. Five resistance genes have been detected on the same chromosome and further studies suggested that these might be different alleles at two resistance clusters. Nevertheless, it was postulated that rhizomania resistance might be a quantitative trait with multiple unknown minor resistance genes. Here, we present a first attempt at genomic prediction of rhizomania resistance in a population that carries resistances at the two known resistance clusters. The sugar beet population was genotyped using single nucleotide polymorphism (SNP) markers. Methods First, genomic prediction was performed using all SNPs. Next, we calculated the variable importance for each SNP using machine learning and performed genomic prediction by including the SNPs incrementally in the prediction model based on their variable importance. Using this method, we selected the optimal number of SNPs that maximised the prediction accuracy. Furthermore, we performed genomic prediction with SNP pairs. We also performed feature selection with SNP pairs using the information about the variable importance of the single SNPs. Results From the four methods under investigation, the latter led to the highest prediction accuracy. These results lead to the conclusion that more than the two known resistance clusters are involved in rhizomania resistance and that genetic interactions affect rhizomania resistance. Finally, we have analysed which SNPs were repeatedly detected in the feature selection process and discovered four SNPs, two of which are located on chromosomes that were previously not associated with rhizomania resistance.

Analyzing the differences and correlations between key metabolites and dominant microorganisms in different regions of Daqu based on off-target metabolomics and high-throughput sequencing

Daqu is usually produced in an open environment, which makes its quality unstable. The microbial community of Daqu largely determines its quality. Therefore, in order to improve the fermentation characteristics of Daqu, samples were collected from Jinsha County (MT1), Xishui County (MT2), and Maotai Town (MT3) in Guizhou Province to explore the microbial diversity of Daqu and its impact on Daqu's metabolites.Off-target metabolomics was used to detect metabolites, and high-throughput sequencing was used to detect microorganisms. Metabolomics results revealed that MT1 and MT2 had the highest relative fatty acid content, whereas MT3 had the highest organooxygen compound content. Principal component analysis and partial least squares discriminant analysis revealed significant differences in the metabolites among the three groups, followed by the identification of 33 differential metabolites (key metabolites) filtered using the criteria of variable importance in projection >1 and p < 0.001. According to the microbiological results, Proteobacteria was the dominant bacteria phylum in three samples. Gammaproteobacteria was the dominant class in MT1(26.84 %) and MT2(36.54 %), MT3 is Alphaproteobacteria(25.66 %). And Caulobacteraceae was the dominant family per the abundance analysis, MTI was 24.32 %, MT2 and MT3 were 33.71 % and 24.40 % respectively. Three samples dominant fungi phylum were Ascomycota, and dominant fungi family were Thermoascaceae. Pseudomonas showed a significant positive connection with various fatty acyls, according to correlation analyses between dominant microorganisms (genus level) and key metabolites. Fatty acyls and Thermomyces showed a positive correlation, but Thermoascus had the reverse relation. These findings offer a theoretical framework for future studies on the impact of metabolites on Baijiu quality during fermentation.

Machine learning insights: probing the variable importance of ex-ante information

PurposeEmpirical examinations of initial public offering (IPO) initial returns often rely heavily on linear regression models. However, these models can prove inefficient owing to their susceptibility to outliers, a common occurrence in IPO data. This study introduces a machine learning method, known as random forest, to address issues that linear regression may struggle to resolve.Design/methodology/approachThe study’s sample comprises 352 fixed-priced IPOs from the year 2004 until 2021. A unique aspect of this research is its application of the random forest method. The accuracy of random forest in comparison to other methods is evaluated. The findings indicate that the random forest model significantly outperforms other methods in all of the evaluated aspects.FindingsThe variable importance measure indicates that investors’ demand, divergence of opinion among investors and offer price are the most crucial predictors of IPO initial returns. These determinants hold particular significance due to the widespread use of the fixed-price method in Malaysia, as this method amplifies the information asymmetry in the IPO market.Originality/valueTo the best of the authors’ knowledge, this study is among the pioneering works in Malaysian literature to apply the random forest method to address the constraints of conventional linear regression models. This is achieved by considering a more extensive array of factors and acknowledging the influence of outliers. Additionally, this study adds value to Malaysian literature by ranking and identifying the ex-ante information that best signals the issuing firm’s quality. This contribution facilitates prospective investors’ decision-making processes and provides issuing firms with effective means to communicate their value and quality to the IPO market.

A Novel Traffic Speed Prediction for Road Weaving Sections: Incorporating Traffic Flow Characteristics

Weaving sections on roads are crucial areas with high concentrations of mandatory lane changes, which can increase the likelihood of traffic accidents. Speed is a key factor in determining traffic safety, and the development of an accurate speed prediction method is essential for improving safety in weaving sections. While current methods are effective in predicting speeds in straightforward situations, they face challenges in more complex scenarios such as weaving sections. This study presents a refined traffic speed prediction approach specifically designed for weaving sections in order to tackle the aforementioned issue. Initially, novel variables were formulated to capture the unique traffic flow attributes present in weaving sections, which distinguish them from standard road segments. Subsequently, supplementary empirical variables that are known to impact speed were incorporated. We conducted a variable importance assessment to ascertain the extent and direction of each variable’s contribution. Lastly, variables with significant positive effects were chosen as inputs for three machine learning algorithms: Random Forest (RF), Backpropagation Neural Network optimised with Genetic Algorithm (BPNN-GA) and Support Vector Regression (SVR). This method was evaluated using aerial footage from five distinct weaving sections in China, maintaining an approximately 3 km/h prediction error. In addition, the study also finds that the speed distribution in weaving sections is negatively correlated with the number of lane-changing. Vehicles experience deceleration at both on-ramp and off-ramp, with a more significant deceleration occurring at the on-ramp. Speed is significantly affected by short length, number of lanes and proportion of large vehicles. The proposed method can be embedded into intelligent traffic systems for safe speeds of autonomous vehicles in weaving sections. Reconstructing spatiotemporal patterns of traffic congestion, predicting traffic accidents and implementing active traffic management strategies in weaving sections could be investigated in the future.

Effect of fatty acid composition on the volatile compounds of pasteurized milk during low-temperature storage

The change in milk fat during storage greatly influences its flavor. This study investigates the effect of fatty acid composition on milk flavor by analyzing volatile compounds in pasteurized whole milk (PWM) and pasteurized skim milk (PSM) during storage at 4 °C. 33 types of volatile compounds were detected and the content of ketones was highest, followed by esters and aldehydes. Based on variable importance in projection and relative odor activity value, 2-hexenal dimer, acetic acid ethyl ester dimer, acetic acid ethyl ester, and butanal were identified as the key differential volatile compounds. These compounds were found in higher concentrations in PWM than in PSM, indicating a close relationship with the changes in the fatty acid composition of milk fat. Among 11 fatty acids detected in PWM, the content of saturated fatty acids (SFA) and polyunsaturated fatty acids (PUFA) decreased by 0.69 % and 49.1 %, respectively, while the content of monounsaturated fatty acids increased by 46.8 % during 15 days storage, which suggests that the oxidation of SFA and PUFA contributed more to the volatile compound formation. Correlation analysis between fatty acid composition and volatile compounds found that fatty acid C18:2 and C16:0 were strongly associated for 2-hexenal, acetic acid ethyl ester, and butanal. These fatty acids were mainly derived from neutral lipids or phospholipids. These findings provide a new perspective for the formation pathway of milk flavor.

Determination of Possible Biomarkers for Predicting Well-Differentiated Thyroid Cancer Recurrence by Different Ensemble Machine Learning Methods

Objective: Well-differentiated thyroid cancer (WDTC) is the most common thyroid malignancy and although it is curable, the risk of recurrence is high. In this study, classification algorithms based on clinicopathologic features of WDTC patients were used to determine the possible of recurrence in WDTC and to evaluate potential predictive factors, and possible biomarkers based on the optimal model were identified. Method: In this study, open access data on 383 patients with WDTC, 108 with recurrence and 275 without recurrence, were used. In order to predict recurrence in WDTC patients, features were selected using recursive feature elimination variable selection method among features and classification was performed with two ensemble learning methods (Random Forest, Adaboost). Results: Two different ensemble learning models used to classify recurrence in WDTC were Random Forest with an accuracy of 0.957, sensitivity of 0.889, specificity of 0.978, positive predictive value of 0.923, negative predictive value of 0.967, Matthews correlation coefficient of 0.878, G-mean of 0.945, F1-score of 0.906, and accuracy of 0.940, sensitivity of 0.889, specificity of 0.955, positive predictive value of 0.857, negative predictive value of 0.966, Matthews correlation coefficient of 0.833, G-mean of 0.910, F1-score of 0.873. Conclusion: According to variable importance based on the Random Forest, the 5 possible clinical biomarkers for predicting WDTC recurrence are Response, Risk, Node, Tumor, and age. In the light of these findings, patient management and treatment planning can be organized.