Quantitative Performance Evaluation Research Articles

Quantitative Performance Evaluation of Commonly Used Colormaps for Image Display in Myocardial Perfusion Imaging: Analysis based on Perceptual Metrics.

To quantitatively evaluate the performance of the most used colormaps in image display using perceptual metrics and to what extent these measures are congruent with the true intensity or uptake of pixels at different levels of defect severity in simulated cardiac images. Six colormaps, labeled "Gray", "Thermal", "Cool", "CEqual", "Siemens" and "S Pet" extracted from FIJI ImageJ software are included. Colormap data are converted from the red, green, blue color space to CIELAB. Perceptual metrics for measuring "color difference" were calculated, including difference (ΔE76) and "speed". The pairwise color difference in every two levels or entries is visualized in a 2-dimensional "heatmap distance matrix" for each colormap. Curves are plotted for each colormap and compared. In addition, to apply this technique to clinical images, simulated short-axis cardiac slices with incremental defect severity (10% grading) were employed. The circumferential profile curves of true pixel intensity, lightness or luminance, and color difference are plotted simultaneously for each defect severity to visualize the concordance of the three curves in various colormaps. In 0% defect, all the curves are at the highest level, except for "s pet", in that the lightness is not at its maximum value. In the phantom with 10% defect (or 90% of maximum value), discrepancies among curves appear. In "Siemens", the ΔE76 drops sharply. In "Siemens" colormap, the ΔE76 drops sharply. In 80% defect, ΔE76 curve, in "gray" colormap drops more slowly than other curves of other colormaps. In "s pet", lightness curve rises paradoxically, although the count intensity and ΔE76 curve match. In 70% defect, again, the curves are in good agreement in "thermal", "Siemens" and "cequal". However, a consistent lag exists in "gray". Up to 50% defect, curves maintain their expected pattern, but in defects more severe than 40%, lightness and ΔE76 curves in "cool" and "cequal" rise paradoxically, and in "thermal", they start to slow down in descent. In "Siemens", falling pattern of the three curves continues. For "s pet" colormap, an erratic pattern of lightness and ΔE76 curves exists. Of 6 colormaps investigated for estimating defect severity, "grayscale" is less favorable than others and "thermal" performs slightly better. "S pet" or rainbow, which is used traditionally by many practitioners, is strongly discouraged. The "Siemens" colormap suffers from decreased discriminating power in the range of mild to moderate/severe. In contrast, the "cool" and "cequal" colormaps outperform the other colormaps employed in this study to some extent, although they have some shortcomings.

Correction: 'Evaluation of Kato-Katz and multiplex quantitative polymerase chain reaction performance for clinical helminth infections in Thailand using a latent class analysis' (2023), by Rotejanaprasert et al.

Correction: 'Evaluation of Kato-Katz and multiplex quantitative polymerase chain reaction performance for clinical helminth infections in Thailand using a latent class analysis' (2023), by Rotejanaprasert et al.

A study on theoretical predictive model and experimental findings of melt‐electrospinning process

AbstractThe current study extends a previous work published by the authors in which an analytical predictive model is proposed to simulate the melt‐electrospinning process. The analytical model is specifically designed to predict the various behaviors of the melt‐electrospun fiber under different material and processing conditions. A brief discussion of this model is presented here to establish context and help the reader capture the modeling philosophy employed. The current study complements the previous work by focusing on the experimental aspects of the research. Correlations between the independent process parameters and the topological attributes of the melt‐electrospun fibers are investigated and compared with findings from the theoretical model. The effects of changes in the process parameters on average fiber diameters and the collection diameter are experimentally analyzed using the design of experiments (DOE) techniques. Toward this end, polylactic acid (PLA) is melt‐electrospun at different treatment levels of the processing parameters in a controlled environment. Two regression‐based models—one for predicting the collection diameter and the other for the fiber diameter—are derived from the DOE data for benchmarking and quantitative evaluation of the predictive performance of the theoretical model. The theoretical model is run based on the same treatment levels as the experiment. The elastic parameter values used in the theoretical simulation are extracted from rheological tests. Comparison between the simulated and the observed fiber characteristics revealed that the collector diameter predictions by the theoretical model exhibited approximately a 16.7% difference compared to 24.2% for the average fiber diameter. Finally, a discussion is presented on the challenges and potential factors contributing to the observed differences. Overall, given the identified challenges and gaps in material characterization, the results of the theoretical predictive model are encouraging.

Impact-based flood forecasting in the Greater Horn of Africa

Abstract. Every year Africa is hit by extreme floods which, combined with high levels of vulnerability and increasing population exposure, often result in humanitarian crises and population displacement. Impact-based forecasting and early warning for natural hazards is recognized as a step forward in disaster risk reduction, thanks to its focus on people, livelihoods, and assets at risk. Yet, the majority of the African population is not covered by any sort of early warning system. This article describes the setup and the methodological approach of Flood-PROOFS East Africa, an impact-based riverine flood forecasting and early warning system for the Greater Horn of Africa (GHA), with a forecast range of 5 d. The system is based on a modeling cascade relying on distributed hydrological simulations forced by ensemble weather forecasts, link to inundation maps for specific return period, and application of a risk assessment framework to estimate population and assets exposed to upcoming floods. The system is operational and supports the African Union Commission and the Disaster Operation Center of the Intergovernmental Authority on Development (IGAD) in the daily monitoring and early warning from hydro-meteorological disasters in eastern Africa. Results show a first evaluation of the hydrological reanalysis at 78 river gauging stations and a semi-quantitative assessment of the impact forecasts for the catastrophic floods in Sudan and in the Nile River basin in summer 2020. More extensive quantitative evaluation of the system performance is envisaged to provide its users with information on the model reliability in forecasting extreme events and their impacts.

Resolving the Java Representation Exposure Problem with an AST-Based Deep Copy and Flexible Alias Ownership System

Encapsulation is a critical factor in object-oriented programming languages and design patterns. Nevertheless, programs written in languages like Java may encounter broken encapsulation due to the lack of sufficient supply for ownership and immutability. As a result, this paper introduces SlimeJava, an ownership system extension based on abstract syntax trees and annotation utilization of Java that aims to help programmers prevent representation exposure. We show the features of the proposal with a motivating example using the Memento pattern. We then discuss how the utilization of annotations realizes ownership and why it is effective in avoiding representation exposure issues by comparing it with existing approaches. In the end, a quantitative performance evaluation was conducted to prove that SlimeJava does not cause a substantial overhead in execution time compared to native Java.

Suitability Analysis of Ground Motion Prediction Equations for Western and Central Himalayas and Indo-Gangetic Plains

ABSTRACT Ground motion prediction equations (GMPEs) play a key role in seismic hazard assessment (SHA). Considering the seismo-tectonic, geophysical, and geotectonic characteristics of a target region, all the GMPEs may not be suitable in predicting the observed ground motion effectively. With a fairly large number of published GMPEs, the selection and ranking of suitable GMPEs for the design of logic trees in SHA for a particular target region have become a necessity of late. This paper presents a detailed quantitative evaluation of performance of 16 GMPEs against recorded ground motion data in two target regions, characterized by distinct seismo-tectonic, geophysical, and geotectonical nature. The dataset comprises 465 three-component spectral accelerograms corresponding to 122 earthquake events. The suitability of a GMPE is tested by two widely accepted data-driven statistical methods, namely, likelihood (LH) and log-likelihood (LLH) method. Different suites of GMPEs are shown suitable for different periods of interest. The results will be useful to scientists and engineers for microzonation and estimation of seismic design parameters for the design of earthquake-resistant structures in these regions.

DITTO: A Visual Digital Twin for Interventions and Temporal Treatment Outcomes in Head and Neck Cancer.

Digital twin models are of high interest to Head and Neck Cancer (HNC) oncologists, who have to navigate a series of complex treatment decisions that weigh the efficacy of tumor control against toxicity and mortality risks. Evaluating individual risk profiles necessitates a deeper understanding of the interplay between different factors such as patient health, spatial tumor location and spread, and risk of subsequent toxicities that can not be adequately captured through simple heuristics. To support clinicians in better understanding tradeoffs when deciding on treatment courses, we developed DITTO, a digital-twin and visual computing system that allows clinicians to analyze detailed risk profiles for each patient, and decide on a treatment plan. DITTO relies on a sequential Deep Reinforcement Learning digital twin (DT) to deliver personalized risk of both long-term and short-term disease outcome and toxicity risk for HNC patients. Based on a participatory collaborative design alongside oncologists, we also implement several visual explainability methods to promote clinical trust and encourage healthy skepticism when using our system. We evaluate the efficacy of DITTO through quantitative evaluation of performance and case studies with qualitative feedback. Finally, we discuss design lessons for developing clinical visual XAI applications for clinical end users.

HealthPrism: A Visual Analytics System for Exploring Children's Physical and Mental Health Profiles with Multimodal Data.

The correlation between children's personal and family characteristics (e.g., demographics and socioeconomic status) and their physical and mental health status has been extensively studied across various research domains, such as public health, medicine, and data science. Such studies can provide insights into the underlying factors affecting children's health and aid in the development of targeted interventions to improve their health outcomes. However, with the availability of multiple data sources, including context data (i.e., the background information of children) and motion data (i.e., sensor data measuring activities of children), new challenges have arisen due to the large-scale, heterogeneous, and multimodal nature of the data. Existing statistical hypothesis-based and learning model-based approaches have been inadequate for comprehensively analyzing the complex correlation between multimodal features and multi-dimensional health outcomes due to the limited information revealed. In this work, we first distill a set of design requirements from multiple levels through conducting a literature review and iteratively interviewing 11 experts from multiple domains (e.g., public health and medicine). Then, we propose HealthPrism, an interactive visual and analytics system for assisting researchers in exploring the importance and influence of various context and motion features on children's health status from multi-levelperspectives. Within HealthPrism, a multimodal learning model with a gate mechanism is proposed for health profiling and cross-modality feature importance comparison. A set of visualization components is designed for experts to explore and understand multimodal data freely. We demonstrate the effectiveness and usability of HealthPrism through quantitative evaluation of the model performance, case studies, and expert interviews in associated domains.

The Role of Artificial Intelligence in Enhancing Energy Management in Microgrid Systems

This look at investigates the effect of artificial intelligence (AI) on microgrid overall performance thru a quantitative evaluation of strength performance, reliability, and real-time optimization metrics. Two hypothetical microgrid systems, System A and System B, are examined, revealing that AI-pushed power management in System A outcomes in superior effects in comparison to System B. Descriptive facts exhibit that System A reveals higher strength efficiency (85.2%), increased reliability indices, and more advantageous actual-time adaptability, showcasing the capacity advantages of AI integration. These findings align with current literature, emphasizing the transformative position of AI in optimizing decentralized energy structures. The look at indicates that making an investment in AI technologies for microgrid power management holds promise for attaining sustainability and resilience. Future research need to consciousness on empirical research with actual-global facts to validate these findings.

Quantitative comparison and performance evaluation of deep learning-based object detection models on edge computing devices

Low latency in the detection of objects in images is a fundamental aspect to provide an immediate response in different application scenarios, and to carry out the relevant actions. Deep learning-based models offer better results than traditional object detection techniques. As a result of this increased accuracy, the computational cost of these models is often very high. For this reason, several edge computing devices have emerged that perform inference quickly and efficiently due to the incorporation of hardware accelerators. In this work, different devices are evaluated using object detection algorithms based on deep learning. For this purpose, YOLOv3, YOLOv5 and YOLOX, with all their variants, have been used on an NVIDIA Jetson Nano, an NVIDIA Jetson AGX Xavier and a Google Coral Dev Board. For the evaluation of the models and devices, one of the most widespread datasets, MS COCO, is used. In addition, they have been evaluated using twenty different input sizes and three frameworks (Pytorch, TensorRT and Tensorflow Lite). From the data obtained, data can be extrapolated to other models such as YOLOv8. Additionally, the FPS/Power Consumption and FPS/Cost ratios are analyzed, as well as their feasibility in a real use scenario. As a result of this work, valuable recommendations are provided for projects where this technology is to be applied.

Generalizable stereo depth estimation with masked image modelling.

Generalizable and accurate stereo depth estimation is vital for 3D reconstruction, especially in surgery. Supervised learning methods obtain best performance however, limited ground truth data for surgical scenes limits generalizability. Self-supervised methods don't need ground truth, but suffer from scale ambiguity and incorrect disparity prediction due to inconsistency of photometric loss. This work proposes a two-phase training procedure that is generalizable and retains the high performance of supervised methods. It entails: (1) performing self-supervised representation learning of left and right views via masked image modelling (MIM) to learn generalizable semantic stereo features (2) utilizing the MIM pre-trained model to learn robust depth representation via supervised learning for disparity estimation on synthetic data only. To improve stereo representations learnt via MIM, perceptual loss terms are introduced, which improve the model's stereo representations learnt by explicitly encouraging the learning of higher scene-level features. Qualitative and quantitative performance evaluation on surgical and natural scenes shows that the approach achieves sub-millimetre accuracy and lowest errors respectively, setting a new state-of-the-art. Despite not training on surgical nor natural scene data for disparityestimation.

Unveiling the Potential of Infrared Thermography in Quantitative Investigation of Potential-Induced Degradation in Crystalline Silicon PV Module

Potential-induced degradation-shunting (PID-s) is a severe degradation mechanism in photovoltaic (PV) cells that significantly impacts module performance. Regular monitoring and quantitative assessment of PID-s are crucial for ensuring long-term reliability of PV systems. Current-voltage (I-V) characteristics and electroluminescence (EL) imaging are commonly used for quantitative performance evaluation of PID-s affected PV modules. However, conducting I-V measurements is time-consuming when performed across large PV installations, while EL imaging has limitations for severely PID-s affected cells with no EL emission. This article proposes the use of inverse infrared (IRINV) thermography as an alternative investigation technique for PID-s in a PV module. IRINV imaging is fast and also effectively maps the severely PID-s affected cells in a PV module. This article unveils the potential of IRINV thermography in quantitative investigation of PID-s in crystalline silicon PV modules. The module level investigations present insights into the correlations between cell temperature and power output under different imaging conditions using Pearson correlation. Results indicate that steady-state operation with medium input current provides the most suitable condition for quantitative PID-s investigation. Furthermore, cell level analysis of temperature distribution and its variation with PID-s progression has been investigated using histogram and kernel density estimation (KDE) statistical tools, revealing distinct patterns as PID-s progresses. A PID-s severity index is proposed based on KDE, providing a quantitative measure of PID-s severity in cells within a PV module. This work provides valuable insights into the use of IRINV thermography as an alternative technique for assessment of PID-s in PV module inspection.

Predicting River Discharge in the Niger River Basin: A Deep Learning Approach

Across West Africa, the River Niger is a major source of freshwater. In addition, the river system also provides services such as aquaculture, transportation, and hydropower. The river network plays a critical role in the hydropolitics and hydroeconomics of the region. Therefore, River Niger is integral to the development of West Africa, hence, there is a need to ensure that the river’s ecosystem is a healthy one. In light of the changing climate and its associated threats such as droughts and floods, constant monitoring and measurements of the the river’s flow system cannot be overemphasized. This study investigates temporal variations in annual river discharge characteristics at eight stations (Koulikoro, Dioila, Kirango, Douna, Mopti, Dire, Ansongo, and Niamey) in the Niger River basin, presenting detailed quantitative findings. Analyzing discharge data of River Niger from 1950 to 1990, the minimum discharge measures (minimum and 10th percentile) exhibit a consistent decreasing trend post-1960, persisting into the 1990s at several stations. Central tendency measures (mean and 50th percentile) also consistently reduced since 1950, with near-zero median values observed in Diola and Douna. Recovery in mean discharge is evident in Ansongo after 1980. Extreme values (maximum and 90th percentile) show decreasing trends across all stations, with some locations exhibiting a slight recovery after 1980. The decreasing trend in annual minimum, mean, and maximum values has implications for water resources, affecting hydroelectric generation, fish farming, and dry season irrigation. Machine learning algorithms (MLAs) are deployed to predict the prediction of monthly river discharge, with LSTM identified as the best-performing model overall. However, model performance varies across locations, with TCN excelling in Diola but underperforming in Koulikoro. This study emphasizes the chaotic nature of time series data and external drivers limiting the long-term predictive capabilities of MLAs. Quantitative evaluation of MLA performance reveals specific strengths and weaknesses at each station. This study underscores the importance of predicting the 10th percentile of annual river discharge for water resource planning. Models exhibit diverse performance across basins, emphasizing the need for tailored approaches. Further analysis considers measures of central tendencies, predicting the 50th percentile (Q50) and mean discharge values. TCN emerges as the best model for Q50 prediction, showcasing superior performance over other models. Additionally, the study delves into predicting high and low extreme discharges, crucial for understanding potential flood events and preparing for meteorological and hydrological droughts. This study concludes by emphasizing the necessity for location-specific studies in the River Niger basin to facilitate an enhanced integrated river management system.

Experimental study on the quantitative evaluation of the thermal stability performance and heat insulation characteristics of fire‐fighting foam

AbstractFoam extinguishing agents are crucial for the suppression of flammable liquid fires. Their thermal stability performance and heat insulation characteristics are critical indicators to evaluate the efficiency of the fire‐fighting foam. There have been some studies focused on exploring the behavior of fire‐fighting foams exposed to radiant heating. However, the decay mechanisms and heat transfer behaviors of the foam at the micro‐scale are still unclear and require further clarification. Therefore, in this study, the volume reduction coefficient, falling time of foam column height, and the temperature profiles of the foam layer under the thermal radiation environment of different conditions are discussed. The results indicate that the high temperature generated by the radiative heat flux will accelerate the collapse rate of the foam layer. The stability of the foam structure will be seriously damaged. There is a relationship between heat radiation intensity and foam attenuation coefficient. The empirical model for reflecting the fire‐fighting foam collapse process under the fire environment with high heat radiation flux is modified. Moreover, the average collapse rate and temperature difference gradient are used to characterize the thermal stability performance and heat insulation characteristics of the foam. Analysis of the micro‐scale foam structure parameters from the foam scans has revealed that the thermal stability performance and heat insulation characteristics of the foam are stronger when the surface tension of the foam is within the range of 17.4–20.4 mN/m.

Bicycle Data-Driven Application Framework: A Dutch Case Study on Machine Learning-Based Bicycle Delay Estimation at Signalized Intersections Using Nationwide Sparse GPS Data.

Data-driven approaches are helpful for quantitative justification and performance evaluation. The Netherlands has made notable strides in establishing a national protocol for bicycle traffic counting and collecting GPS cycling data through initiatives such as the Talking Bikes program. This article addresses the need for a generic framework to harness cycling data and extract relevant insights. Specifically, it focuses on the application of estimating average bicycle delays at signalized intersections, as this is an essential variable in assessing the performance of the transportation system. This study evaluates machine learning (ML)-based approaches using GPS cycling data. The dataset provides comprehensive yet incomplete information regarding one million bicycle rides annually across The Netherlands. These ML models, including random forest, k-nearest neighbor, support vector regression, extreme gradient boosting, and neural networks, are developed to estimate bicycle delays. The study demonstrates the feasibility of estimating bicycle delays using sparse GPS cycling data combined with publicly accessible information, such as weather information and intersection complexity, leveraging the burden of understanding local traffic conditions. It emphasizes the potential of data-driven approaches to inform traffic management, bicycle policy, and infrastructure development.

Assessment of Postural Control and Proprioception Using the Delos Postural Proprioceptive System

Background. Proper postural control and proprioception are essential for maintaining balance, preventing falls, and performing daily activities. Traditional methods of assessing postural control and proprioceptive function often lack objectivity and precision. The DPPS offers a promising solution by providing a quantitative and comprehensive evaluation of postural stability and proprioceptive performance.
 Methods. The DPPS utilises advanced sensor technology and computerised analysis to measure various parameters related to postural control, including stability indexes, weight distribution, sway patterns, and proprioceptive responses. It incorporates both static and dynamic tests, allowing for a comprehensive assessment of postural control in different conditions.
 Results. Numerous studies have demonstrated the effectiveness of the DPPS in evaluating postural control and proprioception in various populations, such as athletes, older adults, individuals with balance disorders, and patients undergoing rehabilitation. The DPPS has shown good reliability and validity, with its measurements correlating well with other established assessment tools. It has also proven to be sensitive in detecting changes in postural control and proprioceptive function over time.
 Conclusions. The DPPS represents a valuable tool for objective assessment and monitoring of postural control and proprioceptive function. Its ability to provide quantitative data in a reliable and sensitive manner makes it an attractive option for researchers, clinicians, and rehabilitation professionals. By enhancing our understanding of postural control and proprioception, the DPPS can contribute to the development of targeted interventions and personalised treatment strategies aimed at improving balance, preventing falls, and optimising functional performance.
 Keywords: Delos Postural Proprioceptive System, Postural control, Proprioception, Assessment, Balance.

Experimental investigation on comprehensive thermal-hydraulic performance of supercritical CO2 in a NACA 0020 airfoil fin printed circuit heat exchanger

Printed circuit heat exchanger (PCHE) is a promising compact heat exchanger for supercritical CO2 power conversion systems. Among the channel shapes of PCHE, an airfoil fin channel was considered for the high thermal-hydraulic performance of the PCHE. However, quantitative experimental evaluation of the thermal-hydraulic performance between the optimized airfoil fin and the wavy channel (commercially utilized channel shape) has not been performed. Therefore, in this study, to compare the performance according to channel shape, experiments on the optimized arrangement of airfoil fin PCHE (NACA 0020 airfoil fin with non-dimensional arrangement parameters, staggered number 1.0, horizontal number 2.0, and vertical number 1.8) and wavy PCHE (120° bending angle) were conducted. The experimental conditions of supercritical CO2 were temperatures ranging from 70°C (inlet) to 40.5°C (outlet) with pressures varying from 7.5 to 8.5 MPa. To reflect the real etched shape of airfoil fin channel, the real shape of airfoil fin was measured by using optical 3-Dimensional scanner. And hydraulic diameter was calculated based on measured information of real etched airfoil fin shape. The pressure drop of the airfoil fin PCHE was one-fifth of that of the wavy PCHE with the same total heat transfer rate per unit volume. New empirical correlations for the friction factor and Nusselt number of the airfoil fin PCHE were proposed based on the experimental data for the precooler design in the supercritical CO2 Brayton cycle. A friction factor correlation with maximum deviations of ±5.04 % and Nusselt number correlation with maximum deviations of ±14.0 % were proposed.

Quantitative evaluation of collector flotation performance II: The creation of a collector property index based on molecular structure

The evaluation and prediction of flotation performance is a key step in the development of high-performance collectors for the efficient flotation separation of low-grade complex ores. In our previous work, we have created the flotation index (FI) for a comprehensive and standardized evaluation of collector flotation performance. However, as an experiment-based index, FI does not possess the quantitative predictive power that is fundamental to rational flotation reagent design. In this work, the collector properties were obtained by quantum chemistry (QC) calculation. By ranking the importance of the QC properties, machine learning (ML) aids the selection of key properties representing different aspects of a collector (polar, non-polar and overall polarity). The key properties were incorporated into a theory-based collector property index (CPI). Based on a quantitative structure–activity relationship (QSAR), the bridge between theory and experiment were established by searching for the relationship CPI and FI, enabling convenient evaluation and prediction of collector flotation performance. The accuracy of our QSAR model was verified using the galena-pyrite separation system. Our QSAR model demonstrates transferability in that it can predict collectors with diverse skeletons or types, pushing the limit of most conventional QSAR models, which only apply to collectors with similar skeletons or types. This work provides an alternative pathway for the rational design and performance prediction of flotation surfactants.

EPCO-15. DETECTING HISTOLOGIC & CLINICAL GLIOBLASTOMA PATTERNS OF PROGNOSTIC RELEVANCE

Abstract BACKGROUND Glioblastoma is the most common adult primary brain tumor. Limited improvement in patient prospects over the last 20 years reflects our limited disease understanding. Robust prognostic patient stratification from whole slide images (WSI) & clinical data using interpretable computational methods could improve disease understanding and patient management. MATERIAL AND METHODS The TCGA-GBM and TCGA-LGG data collections were reclassified according to the 2021 WHO classification criteria, identifying 188 glioblastoma (IDH-wt, Gr.4) cases and stratifying them to short- (< 9 months) and long-survivors ( > 13 months). H&E-stained WSI were comprehensively curated to remove artifactual content. An interpretability mechanism based on weakly-supervised attention-based multiple-instance-learning was employed. The prognostic relevance of associated clinical data was also evaluated both independently and integratively, using XGBoost and ‘SHapely Additive exPlanations’. Late fusion amalgamates the advantage of WSI patterns with clinical features by averaging the output probabilities of two distinct models at the decision level. Quantitative performance evaluation was based on a 10-fold monte carlo cross-validation, with training (80%), validation (10%), and test (10%) sets. RESULTS Our imaging, clinical, and integrative multi-modal models revealed AUCTest of 0.68/0.72/0.75 respectively. The imaging analysis revealed correctly predicted long-survivors associated with histologically malignant areas, including necrosis, hypercellularity, atypia, infiltration, and proliferation. Correctly classified short survivors highlighted influence by aggressive infiltrative features, including leptomeningeal involvement. The clinical analysis revealed age, mutation count, and TERT promoter status as the top three features contributing to stratification, in the XGBoost model. CONCLUSION Our findings support data-driven histology & data interpretation of algorithmic decisions contribute to further understanding of glioblastoma by identifying morphology patterns & clinical patterns of prognostic relevance. Integration of multiple modalities can be a powerful approach for improving the performance of AI models.

Wind tunnel testing of a Formula Student vehicle for checking CFD simulation trends

The aerodynamic performance analysis of Formula Student racecars has been mostly done by teams with CFD tools, for time and cost savings, that often lack proper validation. To address this, the FST Lisboa team performed a detailed wind tunnel (WT) test campaign, using a one-third scale model, under different configurations, including variable ride heights, bullhorn appendix, and rear wing flap settings, also replicated in CFD. The simulations used RANS with the SST k-omega turbulence model, with a 13.7 million polyhedral mesh for the test chamber region domain. Both experimental and numerical errors were estimated from the instrumentation and mesh convergence analysis, respectively. Comparisons were made between WT and CFD both in terms of local flow, using tufts for flow visualization, and global flow, using lift, drag, and pitching moment coefficients. Overall, the numerical streamlines agreed very well with the orientations of the tufts in experiments, but some discrepancies were found in regions of cross-flow and high-frequency unsteadiness, mainly caused by limitations of the visualization technique. The gamma transition model in CFD was abandoned as it could not replicate the WT observations. In terms of aerodynamic coefficients, a strong correlation was found between WT and CFD. The parametric studies revealed that the simulations captured the experimental sensitivity to each car setting parameter studied but the uncertainties did not enable a full quantitative evaluation of the aerodynamic performance. The drag reduction system significantly impacted the aerodynamic balance of the racecar, while the current bullhorn design proved to be ineffective. The ride height increase led to higher downforce, mostly due to the higher pitch angle of the vehicle, with negligible variation of the aerodynamic balance. This work validated the team CFD studies, building confidence in that trends estimated in numerical parametric studies are likely to be translated to the real prototype performance.