Statistical Significance Tests Research Articles

XL1R-Net: Explainable AI-driven improved [formula omitted]-regularized deep neural architecture for NSCLC biomarker identification

Background and Objective:Non-small cell lung cancer (NSCLC) exhibits intrinsic molecular heterogeneity, primarily driven by the mutation of specific biomarkers. Identification of these biomarkers would assist not only in distinguishing NSCLC into its major subtypes — Adenocarcinoma and Squamous Cell Carcinoma, but also in developing targeted therapy. Medical practitioners use one or more types of omic data to identify these biomarkers, copy number variation (CNV) being one such type. CNV provides a measure of genomic instability, which is considered a hallmark of carcinoma. However, the CNV data has not received much attention for biomarker identification. This paper aims to identify biomarkers for NSCLC using CNV data. Methods:An eXplainable AI (XAI)-driven L1-regularized deep learning architecture, XL1R-Net, is proposed that introduces a novel modification of the standard L1-regularized gradient descent algorithm to arrive at an improved deep neural classifier for NSCLC subtyping. Further, XAI-based feature identification has been used to leverage the trained classifier to uncover a set of twenty NCSLC-relevant biomarkers. Results:The identified biomarkers are evaluated based on their classification performance and clinical relevance. Using Multilayer Perceptron (MLP)-based model, a classification accuracy of 84.95% using 10-fold cross-validation is achieved. Moreover, the statistical significance test on the classification performance also revealed the superiority of the MLP model over the competitive machine learning models. Further, the publicly available Drug–Gene Interaction Database reveals twelve of the identified biomarkers as potentially druggable. The K–M Plotter tool was used to verify eighteen of the identified biomarkers with a high probability of predicting NSCLC patients’ likelihood of survival. While nine of the identified biomarkers confirm the recent literature, five find mention in the OncoKB Gene List. Conclusion:A set of seven novel biomarkers that have not been reported in the literature could be investigated for their potential contribution towards NSCLC therapy. Given NSCLC’s genetic diversity, using only one omics data type may not adequately capture the tumor’s complexity. Multiomics data and its integration with other sources will be examined in the future to better understand NSCLC heterogeneity.

Impact Analysis of the Use of IoT Technology and Environmental Monitoring Systems on Waste Reduction in the Food and Beverage Industry in Indonesia

The present research endeavors to examine the effects of IoT technology and environmental monitoring systems on waste reduction within Indonesia's food and beverage sector. Using a sample of 250 participants, a thorough study was carried out while accounting for the size, location, and kind of operation of the organization. Discriminant validity was verified after the measurement model underwent a thorough evaluation for validity and reliability. Tests of statistical significance demonstrated how important IoT technology and environmental monitoring systems are in affecting waste reduction. The model's significant explanatory and predictive power is demonstrated by the R-Square and Q2 values, while the model fit index highlights the estimated model's suitability. The results offer significant perspectives for industry participants and policy officials who seek to enhance sustainability measures in Indonesia's food and beverage industry.

Left atrial appendage occlusion for atrial fibrillation and bleeding diathesis.

Patients with atrial fibrillation(AF) and likelihood of bleeding can undergo left atrial appendage occlusion (LAAO) as an alternative method of stroke prophylaxis. Short-term anti-thrombotic drugs are used postprocedure to offset the risk of device-related thrombus, evidence for this practice is limited. To investigate optimal postimplant antithrombotic strategy in high bleeding-risk patients. Patients with AF and high-risk for both stroke and bleeding undergoing LAAO were advised their perioperative drug therapy by a multidisciplinary physician panel. Those deemed to be at higher risk of bleeding from anti-thrombotic drugs were assigned to minimal treatment with no antithrombotics or Aspirin-alone. The remaining patients received standard care (STG) with a 12 week course of dual-antiplatelets or anticoagulation postimplant. We compared mortality, device-related thrombus, ischemic stroke, and bleeding events during the 90 days postimplant and long-term. Event-free survival was assessed using Kaplan-Meier survival analysis, with logrank testing for statistical significance. Seventy-five patientsunderwent LAAO of whom 63 patients(84%) had a prior serious bleeding event. The 42 patientson minimal treatment were older (74.3 ± 7.7 vs. 71.2 ± 7.2) with higher HASBLED score (3.6 ± 0.9 vs. 3.3 ± 1.2) than the 33 patientshaving standard care. There were no device-related thrombi or strokes in either group at 90 days postprocedure; STG had more bleeding events (5/33 vs. 0/42,p = 0.01) with associated deaths (3/33 vs. 0/42,p = 0.05). During long-term follow-up (median 2.2 years), all patients transitioned onto no antithrombotic drugs (43 patients [61%]) or a single-antiplatelet (29 patients[39%]). There was no evidence of early minimal treatment adversely affecting long-term outcomes. Short-term anti-thrombotic drugs may not be needed after LAAO implant in patients with high bleeding risk and could be harmful. Larger, prospective studies would be warranted to test these findings.

How to Calculate Statistics for Significant Difference Test Using SPSS: Understanding Students Comprehension on the Concept of Steam Engines as Power Plant

A significant difference test is used to evaluate certain treatments on the sample in two different observation periods. One of the commonly used software is SPSS which is used to analyze data which helps researchers in calculating data so that it can be completed quickly. However, there are still many students and researchers who are not experts in calculating data using SPSS software, especially significant difference tests. This article aims to provide a step-by-step guide in calculating data using SPSS for statistical requirements and significant difference tests. To understand the calculations well using SPSS, we demonstrate the requirements tests (i.e., normality and homogeneity tests), parametric significant difference tests (i.e., One Sample t-test, Paired sample t-test, and Independent Sample t-test), and non-parametric (i.e., Wilcoxon test and Mann-Whitney test). We also added and demonstrated the steps for calculating data in the field of education with the variables analyzed being differences in student learning outcomes. We used the data when delivering the steam engine concept to students, showing how statistical calculation can understand students' comprehension. Bibliometric analysis regarding statistics was also added. This paper can be used as a guide in carrying out statistical tests using SPSS software.

A gene network-driven approach to infer novel pathogenicity-associated genes: application to Pseudomonas aeruginosa PAO1.

We present here a new systems-level approach to decipher genetic factors and biological pathways associated with virulence and/or antibiotic treatment of bacterial pathogens. The power of this approach was demonstrated by application to a well-studied pathogen Pseudomonas aeruginosa PAO1. Our gene co-expression network-based approach unraveled known and unknown genes and their networks associated with pathogenicity in P. aeruginosa PAO1. The systems-level investigation of P. aeruginosa PAO1 helped identify putative pathogenicity and resistance-associated genetic factors that could not otherwise be detected by conventional approaches of differential gene expression analysis. The network-based analysis uncovered modules that harbor genes not previously reported by several original studies on P. aeruginosa virulence and resistance. These could potentially act as molecular determinants of P. aeruginosa PAO1 pathogenicity and responses to antibiotics.

Network Modeling of Apoptosis-Related Genes Reveal a High-Risk Cluster in Multiple Myeloma

The plasma cell cancer multiple myeloma (MM) varies significantly in genomic characteristics, response to therapy, and long-term prognosis. Recent studies have shown the complexity of biological systems associated with poor outcomes in MM. One hallmark of cancer pathogenesis, applicable to MM, is the dysregulation of apoptotic cell death. Understanding the relationship between known drivers of MM and critical pathways in the apoptosis gene network is key towards identifying novel therapeutic targets. Recently, there has been a surge of interest in developing network-based approaches to study gene interactions and use machine learning tools to further elucidate useful biological factors. It is well-known that genes operate as network systems; however, interpretable computational methods are still limited. In this work, we combine a set of network analysis tools to provide novel biological insights into how the apoptosis network is dysregulated in MM. We identify a set of genes associated with poor outcomes in MM using both univariate and network-based approaches. Genomic and clinical data, including RNA-sequencing (RNA-seq) and copy number alteration (CNA), for 659 subjects were obtained from the Multiple Myeloma Research Foundation's CoMMpass (IA19) database. The apoptosis gene set is defined using the apoptosis hallmark pathway provided by the gene set enrichment analysis tool. The interactions between the given list of genes are provided via MetaCore. Univariate gene modeling is done using a Cox proportional hazards model. To determine the relationship between genes, we used Ollivier-Ricci curvature (ORC), a measure of network robustness. The ORC value between two genes incorporates information about both the two genes in question and their local neighborhoods, a concept critical for understanding the role a neighborhood of genes has on a given connection. Once edge information is computed, we used a uniform manifold approximation and projection (UMAP) to reduce the complexity of the dataset. The resulting projection is clustered using k-means clustering, and survival analysis of the given clusters is done using the Kaplan-Meier (KM) method. To understand which features are most informative, we used a random forest to predict the cluster labels using the RNA-seq data. After five-fold cross-validation, feature importance was computed using random forest's permutation importance methodology. All statistical significance testing was corrected for multiple comparisons via the Benjamini-Hochberg false discovery rate method with an alpha set to 0.05. Using RNA-seq data of genes associated with apoptosis, we identified nineteen apoptosis genes associated with poor prognosis in MM. Some of these genes, such as ANXA1 and WEE1 are known to have an effect in MM, while others, such as AVPR1A have yet to be studied extensively for their role in MM. When examining CNA data associated with apoptosis, a similar pattern emerged - nineteen genes were associated with poor prognosis. However, between the two key gene sets found by RNA-seq and CNA, only FAS appeared in both lists. FAS, Fas Cell Surface Death Receptor, is critical for triggering apoptosis. Half of the 37 genes identified using RNA-seq and CNA data are directly connected to the TP53 gene. While TP53 mutations are not a marker of the presence of MM itself, TP53 mutations have been shown to be associated with poorer outcomes. After reducing the complexity of the Ollivier-Ricci curvature analysis of the RNA-seq apoptosis network using UMAPs, the data revealed five clusters with KM curves for progression-free survival were significant (p<0.001). To identify the genes most impactful for separating these clusters, we trained a random forest model on the RNA-seq - note, it was the edge robustness values that were clustered - and evaluated its performance for predicted class membership. The model had an overall accuracy of approximately 78% with F1 scores for each class ranging from 61% to 87%. The top three features associated with the model include CTNNB1, RELA, and LEF1. Of note, CTNNB1 and RELA are directly interacted with the TP53 gene. Because these three genes discriminate between all five clusters, their expression levels may play a mediating effect on MM outcomes. Further investigation with external datasets is needed to validate these results.

Competitive Coevolution-Based Improved Phasor Particle Swarm Optimization Algorithm for Solving Continuous Problems

Particle swarm optimization (PSO) is a population-based heuristic algorithm that is widely used for optimization problems. Phasor PSO (PPSO), an extension of PSO, uses the phase angle θ to create a more balanced PSO due to its increased ability to adjust the environment without parameters like the inertia weight w. The PPSO algorithm performs well for small-sized populations but needs improvements for large populations in the case of rapidly growing complex problems and dimensions. This study introduces a competitive coevolution process to enhance the capability of PPSO for global optimization problems. Competitive coevolution disintegrates the problem into multiple sub-problems, and these sub-swarms coevolve for a better solution. The best solution is selected and replaced with the current sub-swarm for the next competition. This process increases population diversity, reduces premature convergence, and increases the memory efficiency of PPSO. Simulation results using PPSO, fuzzy-dominance-based many-objective particle swarm optimization (FMPSO), and improved competitive multi-swarm PPSO (ICPPSO) are generated to assess the convergence power of the proposed algorithm. The experimental results show that ICPPSO achieves a dominating performance. The ICPPSO results for the average fitness show average improvements of 15%, 20%, 30%, and 35% over PPSO and FMPSO. The Wilcoxon statistical significance test also confirms a significant difference in the performance of the ICPPSO, PPSO, and FMPSO algorithms at a 0.05 significance level.

SFKNN-DPC: Standard deviation weighted distance based density peak clustering algorithm

DPC (Clustering by fast search and find of Density Peaks) algorithm and its variations typically employ Euclidean distance, overlooking the diverse contributions of individual feature to similarity and subsequent clustering. To address this limitation, the standard deviation weighted distance is proposed in this paper to enhance the Euclidean distance. This weighted distance takes into account the specific contribution of each feature to the distance (similarity) between data points. By utilizing this weighted distance, the local density ρi and distance δi of point i are defined, thereby capturing the local pattern of point i to the fullest extent possible. Outliers are defined using this innovative weighted distance. The divide and conquer assignment strategy is proposed based on this proposed weighted distance and the semi-supervised learning and the mutual K-nearest neighbor assumption. Consequently, the SFKNN-DPC (Standard deviation weighted distance and Fuzzy weighted K-Nearest Neighbors based Density Peak Clustering) algorithm is proposed, aiming to effectively uncover the hidden clusters within a dataset. Extensive experiments conducted on benchmark datasets demonstrate the superiority of SFKNN-DPC over DPC, its variations, and other benchmark clustering algorithms. Moreover, statistical significance tests indicate that SFKNN-DPC exhibits notable differences when compared to its counterparts.

Multi-Criteria Optimization and RapidPlan for Improved Organs at Risk Sparing in Treatment Planning of Malignant Pleural Mesothelioma

Treatment planning for malignant pleural mesothelioma is complex and time-consuming owing to the large volume of the target as well as its overlap and proximity to critical organs. Knowledge-based planning (KBP) model using RapidPlan (RP) was previously developed and clinically tested at our institution for treatment planning of these cases that had undergone pleurectomy-decortication, and therefore had two intact-lungs. The aim of this work is to investigate if multi-criteria optimization (MCO) can further improve the RP model with respect to organ at risk (OAR) sparing without compromising target coverage. The RP model was trained with clinically accepted plans of 57 patients that used Volumetric Modulated Arc Therapy (VMAT) with 2 partial arcs and 6 MV photons. The dose volume histogram (DVH) estimation model was trained to estimate doses to the heart, ipsilateral lung, total lung, contralateral lung, stomach, esophagus, kidneys and liver. Clinical treatment plans for 12 patients were re-planned using RP as well as a combination of RP and MCO. Application of MCO was after RP and used tradeoff exploration to navigate to the improved dose distribution on a Pareto surface. Selection of a plan that further improved OAR sparing while maintaining coverage constraints of PTV D95 and V95 ≥ 94% was made. Dosimetric parameters for clinical plans (CP), plans using RP as well as plans generated with RP and MCO were all compared for 12 new validation cases. Wilcoxon sign-rank test was used for statistical significance testing. The mean heart dose was reduced from 20.9 Gy ± 3.2 Gy for CP to 17.1 Gy ± 4.2 Gy with RP and further to 13.8 Gy ± 3.3 Gy with RP and MCO. The heart V30 Gy was reduced from 23.7% ± 8% to 18.2% ± 9.4% with RP and further to 13.6% ± 5.7% with RP and MCO. The corresponding results for contralateral lung V5 Gy were 68.6% ± 14.1% with CP, 50% ± 26.3% with RP and 45.3% ± 22.8% with RP and MCO, while the mean esophagus dose was reduced from 26.5 Gy ± 3.1 Gy to 21.5 Gy ± 4.7 Gy with RP and further to 19.1 Gy ± 4.2 Gy with RP and MCO. All these dosimetric improvements were statistically significant (p<0.001). However, improvements with RP and MCO for the total lung mean and V20 Gy, liver mean, stomach mean and kidney V18 Gy were marginal over the RP. Sparing of ipsilateral lung V20 Gy was maintained at ≥ 50 cc on average for all plans. PTV D95 and V95 were both normalized at 94%. Combination of RP and MCO significantly improved sparing of the OARs, especially the heart, contralateral lung and esophagus without compromising coverage or doses to other structures. Since doses to the heart, and contralateral lung are known to correlate with incidence of radiation pneumonitis, it would be prudent to consider planning with both RP and MCO to help determine the optimal treatment plan for the individual patient's anatomy.

Cost per Patient Achieving Treatment Targets and Number Needed to Treat with Tirzepatide Versus Semaglutide 1mg in Patients with Type 2 Diabetes in the United States.

Achieving glycemic control can help reduce complications of type 2 diabetes (T2D). This study compared the pharmacy cost per responder and number needed to treat (NNT) of tirzepatide 5mg, 10mg, and 15mg versus semaglutide 1mg to achieve glycemic, weight loss, and composite treatment endpoints in patients with T2D in the United States. The proportions of patients achieving glycemic, weight loss, and composite treatment endpoints were obtained from the phase 3 SURPASS-2 randomized clinical trial which compared tirzepatide 5mg, 10mg, and 15mg to semaglutide 1mg. Annual pharmacy costs were calculated using 2022 wholesale acquisition costs. Cost per responder and NNT were calculated along with 95% confidence intervals and tests for statistical significance (P ≤ 0.05). Tirzepatide had a lower cost per responder to achieve glycated hemoglobin A1c (HbA1c) endpoints of ≤ 6.5% (10mg and 15mg doses) and < 5.7% (all doses) and weight loss endpoints of ≥ 5% (10mg and 15mg doses), ≥ 10% (all doses), and ≥ 15% (all doses). The cost per responder to achieve HbA1c < 7% (all doses of tirzepatide) and ≤ 6.5% (5mg tirzepatide) were not statistically significantly different between tirzepatide and semaglutide 1mg. The cost per patient to achieve the composite endpoints (HbA1c < 7.0%, ≤ 6.5%, or < 5.7%/weight loss ≥ 10%/no hypoglycemia) was statistically significantly lower for all doses of tirzepatide than for semaglutide 1mg. The NNTs for all doses of tirzepatide were statistically significantly lower than that for semaglutide 1mg to achieve all individual and composite endpoints, with the exception of the 5mg dose for HbA1c < 7.0% and HbA1c ≤ 6.5%, where tirzepatide had numerically lower NNTs that were not statistically significant. Tirzepatide is a novel glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) receptor agonist (RA) that may offer the potential to achieve stringent glycemic goals, weight loss targets, and composite treatment goals at a lower cost per responder compared to semaglutide 1mg among people with T2D.

Dietary intake, anthropometric profile and somatotype in university athletes: Differences between gender and sports practiced

BACKGROUND: Research on energy intake, nutrient intake, and anthropometric measurements in elite athletes is extensive, while it is limited in university athletes (UA), where a specific modality segments it. OBJECTIVE: This study aimed to identify the dietary intake, anthropometric profile, somatotype in UA, and differences between gender and sports modalities practiced. METHODS: A descriptive and observational study was conducted, including UA from Ecuador’s Catholic University of Santiago de Guayaquil (CUSG). Dietary intake was evaluated with a 24-hour recall questionnaire, an anthropometric profile with the ISAK protocol, and the somatotype using the Heath and Carter formula. Normality tests of Shapiro Wilk and Kolmogorov Smirnov were performed; mean and standard deviation (SD) were calculated, and to establish statistical significance, ANOVA and Kruskal-Wallis tests were used. RESULTS: The study population consisted of 29 males and 51 females, with a mean of 21.1 years. Energy intake, macronutrients, micronutrients, and fiber consumption showed no differences regardless of the sport practiced. The mean somatotype of male athletes was categorized as endomesomorphic and females as mesoendomorph. CONCLUSIONS: The data obtained through the anthropometric variables and used to determine the somatotypes confirmed that sexual dimorphism presented differences in size, muscle development, and body fat percentage.

TestDimorph: An R package for analysis of interpopulation sexual dimorphism differences using summary statistics.

The degree of sexual dimorphism in certain traits between males and females differ from one sample to another. Although trait differences by sex are often reported in bioanthropological research, few studies test for statistical significance or make raw data available. TestDimorph is the first R package dedicated to testing and comparing the degree of sexual dimorphism among different samples by leveraging summary statistics. We provide two approaches of analysis of inter-sample differences in degree of sexual dimorphism: univariate and multivariate for two or more samples. The methods follow upon publications primarily from the AJBA. Within-sex size variability between samples is compared using one-way ANOVA followed by control for multiple pairwise comparisons. In addition, we compute the overlapping area between the density functions of two normal distributions from the mixture intersection index or the non-overlapping area using the dissimilarity index as well as Hedges' g with inferential support using the 95% confidence interval. Finally, we use a multivariate analysis of differences in patterning of sexual dimorphism between samples. We demonstrate various results from applying TestDimorph functions to data supplied with the package. The package has many features including functionality for working with summary statistics, simulating data from summary statistics, and the extraction of summary statistics from raw data, so that the entire analysis can be performed through the package.

Metastatic site patterns by intrinsic subtype and HER2DX in early HER2-positive breast cancer.

Even with contemporary treatment strategies, more than 10% of HER2-positive early stage breast cancer patients may experience distant metastasis as first event during follow-up. Tools for predicting unique patterns of metastatic spread are needed to plan personalized surveillance. We evaluated how molecular heterogeneity affects the pattern of distant relapse in HER2-positive breast cancer. A total of 677 HER2-positive stage I-III breast cancer patients from ShortHER trial, Cher-LOB trial, and 2 institutional cohorts were included. PAM50 molecular subtypes and research-based HER2DX scores were evaluated. The cumulative incidence of distant relapse as the first event (any site and site specific) was evaluated using competing risk analysis. Median follow-up was 8.4 years. Tests of statistical significance are 2-sided. Stage III and high HER2DX risk score identified patients at the highest risk of distant relapse as first event (10-year incidence 24.5% and 19.7%, respectively). Intrinsic molecular subtypes were associated with specific patterns of metastatic spread: compared with other subtypes, HER2-enriched tumors were more prone to develop brain metastases (10-year incidence 3.8% vs 0.6%, P = .005), basal-like tumors were associated with an increased risk of lung metastases (10-year incidence 11.1% vs 2.6%, P = .001), and luminal tumors developed more frequently bone-only metastases (10-year incidence 5.1% vs 2.0%, P = .042). When added to stage or HER2DX risk score in competing risk regression models, intrinsic subtype maintained an independent association with site-specific metastases. The integration of intrinsic molecular subtypes with stage or HER2DX risk score predicts site-specific metastatic risk in HER2-positive breast cancer, with potential implications for personalized surveillance and clinical trials aimed at preventing site-specific recurrence.

Memahami Dinamika Harga Saham melalui Kinerja Keuangan dan Skala Perusahaan, Dengan Peran Kebijakan Dividen sebagai Variabel Pengendali pada Perusahaan Industri Dasar dan Kimia di Indonesia

The dynamics of stock prices is a very important aspect of the capital market. The price of a company's stock is influenced by many factors, including the company's financial performance and the size of the company. In addition, dividend policy is also considered as an important variable in controlling stock prices. However, there is not yet a strong consensus about the extent of the role of dividend policy in moderating the relationship between financial performance and company scale and stock prices. This study aims to understand in depth the dynamics of stock prices through analysis of financial performance and company scale, with particular emphasis on the role of dividend policy as a controlling variable. We aim to identify the extent to which financial performance and firm size affect stock prices, and to what extent dividend policy can moderate this relationship. This study uses a quantitative approach with regression analysis as the main method. Historical data on financial performance, company scale, dividend policy and company stock price will be analyzed for the relevant period. In addition, descriptive statistical analysis and significance test will be used to examine the relationship between these variables. The results of the study show that simultaneously financial performance, company scale, and dividend policy have an effect on stock prices with a contribution value of 72% is a significant finding. It provides a solid understanding of the complexity of the factors that influence stock prices in Chemical and Basic Industry companies in Indonesia. In addition, the finding that dividend policy can act as a moderating factor in increasing stock prices also has important implications. This indicates that company management has an effective tool in managing market perceptions of their stock prices through dividend policy. When used wisely, dividend policy can be a strategic instrument to increase the value of the company in the eyes of investors.

Prediction Variability to Identify Reduced AI Performance in Cancer Diagnosis at MRI and CT.

Background A priori identification of patients at risk of artificial intelligence (AI) failure in diagnosing cancer would contribute to the safer clinical integration of diagnostic algorithms. Purpose To evaluate AI prediction variability as an uncertainty quantification (UQ) metric for identifying cases at risk of AI failure in diagnosing cancer at MRI and CT across different cancer types, data sets, and algorithms. Materials and Methods Multicenter data sets and publicly available AI algorithms from three previous studies that evaluated detection of pancreatic cancer on contrast-enhanced CT images, detection of prostate cancer on MRI scans, and prediction of pulmonary nodule malignancy on low-dose CT images were analyzed retrospectively. Each task's algorithm was extended to generate an uncertainty score based on ensemble prediction variability. AI accuracy percentage and partial area under the receiver operating characteristic curve (pAUC) were compared between certain and uncertain patient groups in a range of percentile thresholds (10%-90%) for the uncertainty score using permutation tests for statistical significance. The pulmonary nodule malignancy prediction algorithm was compared with 11 clinical readers for the certain group (CG) and uncertain group (UG). Results In total, 18 022 images were used for training and 838 images were used for testing. AI diagnostic accuracy was higher for the cases in the CG across all tasks (P < .001). At an 80% threshold of certain predictions, accuracy in the CG was 21%-29% higher than in the UG and 4%-6% higher than in the overall test data sets. The lesion-level pAUC in the CG was 0.25-0.39 higher than in the UG and 0.05-0.08 higher than in the overall test data sets (P < .001). For pulmonary nodule malignancy prediction, accuracy of AI was on par with clinicians for cases in the CG (AI results vs clinician results, 80% [95% CI: 76, 85] vs 78% [95% CI: 70, 87]; P = .07) but worse for cases in the UG (AI results vs clinician results, 50% [95% CI: 37, 64] vs 68% [95% CI: 60, 76]; P < .001). Conclusion An AI-prediction UQ metric consistently identified reduced performance of AI in cancer diagnosis. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Babyn in this issue.

Prediction of drowsiness using EEG signals in young Indonesian drivers

Indonesia is among the countries with the highest accident rates in the world. Fatigue and drowsiness are among the main causes of the increased risks of accidents in the road transport sector. Sleep-related factors (quality and quantity, time of day) and work-related factors significantly affect the development of fatigue. The EEG signal indicator is often referred to as the gold standard for measuring fatigue and drowsiness. However, previous studies focused primarily on the trends of EEG signals under certain conditions but overlooking the development of drowsiness indicators based on EEG signals. Furthermore, existing studies still do not agree on what parameters in the EEG signal indicator are best at detecting drowsiness. Thus, this study aims to design an EEG signal-based drowsiness indicator under simulated driving conditions. Drowsy drivers were monitored through EEG signal indicators and subjective assessments. The methods used in this study include statistical significance tests, logistic regression, and support vector machine. The results showed that sleep deprivation had a significant effect on increasing alpha, beta, and theta waves. In addition, driving duration significantly increased the theta power and all EEG ratios and decreased the beta power in the alert group. The ratio of (θ + α)/β and θ/β in the SD group also showed a considerable increase in the end of driving. Furthermore, sleep status and driving duration both influenced subjective sleepiness. EEG signals combined with sleep status and driving duration factors generated acceptable model accuracies (77.1% and 90.2% in training and testing, respectively), with 90.5% sensitivity and 90% specificity in data test. Support vector machine showed better classification than that of logistics regression, with the linear kernel as the best classifier. Theta power had the highest effect in the model compared with other EEG signals.

An evolutionary functional link artificial neural network for assessment of compressive strength of concrete structures

Compressive strength (CS) has been considered as the utmost critical parameter while designing concrete structures. Usually, it is determined through laboratory tests, which are expensive, time consuming, and requires consumptions of materials. Therefore, correct prediction of CS before actual placement of concrete is highly desirable. The relationship among the constituent materials that forms concrete structures is highly nonlinear and necessitates application of intelligent methods. Though a few such methods like artificial neural network (ANN) based models are available in the literature; their performance in the context is limited to certain extent and they have their own merits and demerits. Hence, to address some of the limitations of ANN (non-higher order Neural Network) based models, this contribution proposed a hybrid model in which a flat network i.e., a type of higher order neural network- functional link artificial neural network (FLANN) is used as the base structure and genetic algorithm (GA) is employed to find out the optimal FLANN parameters (i.e., GA + FLANN). The training process comprises selection of connection weights, bias, as well as optimal number of basis functions of FLANN by GA rather fixing them earlier. Thus, an optimal FLANN is crafted on fly from exploitation of training data. The proposed model is used to assess the CS of concrete cements from datasets available in the literature considering samples of curing age at 3, 7, 14, 28, 56, and 91 days. A rolling window is used for input selection and five evaluation metrics are used for performance evaluation. On an average, the GA + FLANN obtained 0.477052 MAPE (mean absolute percentage of error), 0.598067 ARV (average relative variance), 0.593862 UT (U of Theil’s statistic), 0.551452 NMSE (normalized mean squared error), and 0.206135 SD (standard deviation) values which are the lowest compared to others. The superiority of the model is established through comparative studies and statistical significance tests.

An Objective Evaluation Method for Driver/Passenger Acceptance of an Autonomous Driving System for Lane Changes

Reliable assessment methods of driver acceptance are needed due to increased interest in high levels of autonomous driving systems. Subjective evaluation methods have mostly been utilized to assess the acceptance of newly developed advanced driver assistance systems because acceptance varies depending on the individual. In this paper, an objective evaluation methodology of driver acceptance for an autonomous driving system was proposed based on objective measurable parameters in the case of automatic lane change situations. To this end, a massive driver–vehicle interaction database was utilized, constructed by a specially designed experimental program. The experiment was carried out with 19 selected drivers (9 experts and 10 novices), supposed as an autonomous driving system. The database consisted of not only various measurable parameters on control commands, vehicle behaviors, and relations with other vehicles but also subjective acceptances. To interpret the driver acceptance, objective parameter sets were derived by two different methods: a statistical significance test and an acceptance sensitivity analysis. Then, a modeling method based on stochastic estimation to evaluate driver acceptance was suggested as an objective evaluation method for the driver acceptance of an automatic lane change system. The data set of the expert drivers was only used for the acceptance evaluation modeling; the other data sets of the novice drivers were used for verifications for the suggested model. The estimation accuracies of the two different models using a significance test and sensitivity analysis were 90.2% and 99.5%, respectively. This objective method for acceptance evaluation can not only be expanded to other functions of an autonomous driving system but also to an entirely autonomous driving vehicle.

Digital image colour analysis to evaluate the moisture sensitivity of hot-mix asphalt mixes

Moisture in hot-mix asphalt (HMA) can cause serious damage such as cracks and ruts if left untreated. This is due to the separation of binder and aggregates caused by moisture-induced stripping. Existing approaches for assessing moisture damage in asphalt are time consuming, destructive and expensive to implement. In contrast, digital image analysis provides a simple, inexpensive and non-destructive method for identifying moisture-sensitive mixes. Colour, as an essential characteristic of an image, effectively conveys information about its content. This study seeks to explore the potential of analysing digital image colour in the L*a*b* colour space to assess the moisture sensitivity of HMA after moisture-induced stress tester conditioning. Laboratory-compacted specimens of loose asphalt mixtures from the field were studied by image processing followed by statistical tests of significance. It is found that two parameters – namely, the ‘average colour value for the L* channel’ and the ‘standard deviation value for the b* channel’ – can be used to distinguish between moisture-sensitive and non-sensitive mixes. The findings of this study suggest that future research may find colour to be a helpful criterion for characterising asphalt materials using digital image processing.

Data anomaly detection with automatic feature selection and deep learning

As increasing sensor-based structural health monitoring (SHM) systems are implemented on civil infrastructures, sensor data reliability plays a crucial role in the assessment of operational performance of bridges. Sensors are heavily exposed to harsh environmental conditions during their operations and inevitably lead to possible unstable performance or failure. Thus, to accurately identify faulty sensors is a prerequisite to processing and analyzing the collected data for assessment purpose. Recently, researchers adopted the convolutional neural network (CNN) approach to identify faulty sensors, focusing on image features. Such approach may overlook some important detailed signal features and the time series approach may still be needed. However, algorithms based on time series tend to be time consuming because of the lengthy and high dimensional dataset. This may be effectively resolved using an automatic feature selection technique, namely Tsfresh, as proposed in this paper to select highly relevant signal features based on statistical tests of significance. A deep learning technique based on fully convolutional network (FCN) can then be efficiently employed for anomaly classification. The algorithm is validated using a dataset collected from a real cable-stayed bridge and results show that the proposed method significantly reduces the training time for the neural network, albeit with high classification accuracy.