Search results for: Newton interpolation

Authors: Maryam Fallahpoor, Biswajeet Pradhan

Abstract:

Background: Artificial intelligence (AI) serves as a valuable tool in mitigating the scarcity of human resources required for the evaluation and categorization of vast quantities of medical imaging data. When AI operates with optimal precision, it minimizes the demand for human interpretations and, thereby, reduces the burden on radiologists. Among various AI approaches, deep learning (DL) stands out as it obviates the need for feature extraction, a process that can impede classification, especially with intricate datasets. The advent of DL models has ushered in a new era in medical imaging, particularly in the context of COVID-19 detection. Traditional 2D imaging techniques exhibit limitations when applied to volumetric data, such as Computed Tomography (CT) scans. Medical images predominantly exist in one of two formats: neuroimaging informatics technology initiative (NIfTI) and digital imaging and communications in medicine (DICOM). Purpose: This study aims to employ DL for the classification of COVID-19-infected pulmonary patients and normal cases based on 3D CT scans while investigating the impact of image format. Material and Methods: The dataset used for model training and testing consisted of 1245 patients from IranMehr Hospital. All scans shared a matrix size of 512 × 512, although they exhibited varying slice numbers. Consequently, after loading the DICOM CT scans, image resampling and interpolation were performed to standardize the slice count. All images underwent cropping and resampling, resulting in uniform dimensions of 128 × 128 × 60. Resolution uniformity was achieved through resampling to 1 mm × 1 mm × 1 mm, and image intensities were confined to the range of (−1000, 400) Hounsfield units (HU). For classification purposes, positive pulmonary COVID-19 involvement was designated as 1, while normal images were assigned a value of 0. Subsequently, a U-net-based lung segmentation module was applied to obtain 3D segmented lung regions. The pre-processing stage included normalization, zero-centering, and shuffling. Four distinct 3D CNN models (ResNet152, ResNet50, DensNet169, and DensNet201) were employed in this study. Results: The findings revealed that the segmentation technique yielded superior results for DICOM images, which could be attributed to the potential loss of information during the conversion of original DICOM images to NIFTI format. Notably, ResNet152 and ResNet50 exhibited the highest accuracy at 90.0%, and the same models achieved the best F1 score at 87%. ResNet152 also secured the highest Area under the Curve (AUC) at 0.932. Regarding sensitivity and specificity, DensNet201 achieved the highest values at 93% and 96%, respectively. Conclusion: This study underscores the capacity of deep learning to classify COVID-19 pulmonary involvement using real 3D hospital data. The results underscore the significance of employing DICOM format 3D CT images alongside appropriate pre-processing techniques when training DL models for COVID-19 detection. This approach enhances the accuracy and reliability of diagnostic systems for COVID-19 detection.

Keywords: deep learning, COVID-19 detection, NIFTI format, DICOM format

Procedia PDF Downloads 44

Authors: Dhanuj M. Gandikota

Abstract:

Sensor-derived three-dimensional (3D) point clouds of trees are invaluable in remote sensing analysis for the accurate measurement of key structural metrics, bio-inventory values, spatial planning/visualization, and ecological modeling. Machine learning (ML) holds the potential in addressing the restrictive tradeoffs in cost, spatial coverage, resolution, and information gain that exist in current point cloud sensing methods. Terrestrial laser scanning (TLS) remains the highest fidelity source of both canopy and below-canopy structural features, but usage is limited in both coverage and cost, requiring manual deployment to map out large, forested areas. While aerial laser scanning (ALS) remains a reliable avenue of LIDAR active remote sensing, ALS is also cost-restrictive in deployment methods. Space-borne photogrammetry from high-resolution satellite constellations is an avenue of passive remote sensing with promising viability in research for the accurate construction of vegetation 3-D point clouds. It provides both the lowest comparative cost and the largest spatial coverage across remote sensing methods. However, both space-borne photogrammetry and ALS demonstrate technical limitations in the capture of valuable below-canopy point cloud data. Looking to minimize these tradeoffs, we explored a class of powerful ML algorithms called Deep Learning (DL) that show promise in recent research on 3-D point cloud reconstruction and interpolation. Our research details the efficacy of applying these DL techniques to reconstruct accurate below-canopy point clouds from space-borne and aerial remote sensing through learned patterns of tree species fractal symmetry properties and the supplementation of locally sourced bio-inventory metrics. From our dataset, consisting of tree point clouds obtained from TLS, we deconstructed the point clouds of each tree into those that would be obtained through ALS and satellite photogrammetry of varying resolutions. We fed this ALS/satellite point cloud dataset, along with the simulated local bio-inventory metrics, into the DL point cloud reconstruction architectures to generate the full 3-D tree point clouds (the truth values are denoted by the full TLS tree point clouds containing the below-canopy information). Point cloud reconstruction accuracy was validated both through the measurement of error from the original TLS point clouds as well as the error of extraction of key structural metrics, such as crown base height, diameter above root crown, and leaf/wood volume. The results of this research additionally demonstrate the supplemental performance gain of using minimum locally sourced bio-inventory metric information as an input in ML systems to reach specified accuracy thresholds of tree point cloud reconstruction. This research provides insight into methods for the rapid, cost-effective, and accurate construction of below-canopy tree 3-D point clouds, as well as the supported potential of ML and DL to learn complex, unmodeled patterns of fractal tree growth symmetry.

Keywords: deep learning, machine learning, satellite, photogrammetry, aerial laser scanning, terrestrial laser scanning, point cloud, fractal symmetry

Procedia PDF Downloads 62

Authors: Ali Ateş, Ansar B. Mwimbo, Ali H. Abdulkarim

Abstract:

General transport equation has a wide range of application in Fluid Mechanics and Heat Transfer problems. In this equation, generally when φ variable which represents a flow property is used to represent fluid velocity component, general transport equation turns into momentum equations or with its well known name Navier-Stokes equations. In these non-linear differential equations instead of seeking for analytic solutions, preferring numerical solutions is a more frequently used procedure. Finite difference method is a commonly used numerical solution method. In these equations using velocity and pressure gradients instead of stress tensors decreases the number of unknowns. Also, continuity equation, by integrating the system, number of equations is obtained as number of unknowns. In this situation, velocity and pressure components emerge as two important parameters. In the solution of differential equation system, velocities and pressures must be solved together. However, in the considered grid system, when pressure and velocity values are jointly solved for the same nodal points some problems confront us. To overcome this problem, using staggered grid system is a referred solution method. For the computerized solutions of the staggered grid system various algorithms were developed. From these, two most commonly used are SIMPLE and SIMPLER algorithms. In this study Navier-Stokes equations were numerically solved for Newtonian flow, whose mass or gravitational forces were neglected, for incompressible and laminar fluid, as a hydro dynamically fully developed region and in two dimensional cartesian coordinate system. Finite difference method was chosen as the solution method. This is a parametric study in which varying values of velocity components, pressure and Reynolds numbers were used. Differential equations were discritized using central difference and hybrid scheme. The discritized equation system was solved by Gauss-Siedel iteration method. SIMPLE and SIMPLER were used as solution algorithms. The obtained results, were compared for central difference and hybrid as discritization methods. Also, as solution algorithm, SIMPLE algorithm and SIMPLER algorithm were compared to each other. As a result, it was observed that hybrid discritization method gave better results over a larger area. Furthermore, as computer solution algorithm, besides some disadvantages, it can be said that SIMPLER algorithm is more practical and gave result in short time. For this study, a code was developed in DELPHI programming language. The values obtained in a computer program were converted into graphs and discussed. During sketching, the quality of the graph was increased by adding intermediate values to the obtained result values using Lagrange interpolation formula. For the solution of the system, number of grid and node was found as an estimated. At the same time, to indicate that the obtained results are satisfactory enough, by doing independent analysis from the grid (GCI analysis) for coarse, medium and fine grid system solution domain was obtained. It was observed that when graphs and program outputs were compared with similar studies highly satisfactory results were achieved.

Keywords: finite difference method, GCI analysis, numerical solution of the Navier-Stokes equations, SIMPLE and SIMPLER algoritms

Procedia PDF Downloads 362

Authors: Ahmed Elaksher, Islam Omar

Abstract:

Accurate geometric relation algorithms are imperative in Earth and planetary satellite and aerial image processing, particularly for high-resolution images that are used for topographic mapping. Most of these satellites carry push-broom sensors. These sensors are optical scanners equipped with linear arrays of CCDs. These sensors have been deployed on most EOSs. In addition, the LROC is equipped with two push NACs that provide 0.5 meter-scale panchromatic images over a 5 km swath of the Moon. The HiRISE carried by the MRO and the HRSC carried by MEX are examples of push-broom sensor that produces images of the surface of Mars. Sensor models developed in photogrammetry relate image space coordinates in two or more images with the 3D coordinates of ground features. Rigorous sensor models use the actual interior orientation parameters and exterior orientation parameters of the camera, unlike approximate models. In this research, we generate a generic push-broom sensor model to process imageries acquired through linear array cameras and investigate its performance, advantages, and disadvantages in generating topographic models for the Earth, Mars, and the Moon. We also compare and contrast the utilization, effectiveness, and applicability of available photogrammetric techniques and softcopies with the developed model. We start by defining an image reference coordinate system to unify image coordinates from all three arrays. The transformation from an image coordinate system to a reference coordinate system involves a translation and three rotations. For any image point within the linear array, its image reference coordinates, the coordinates of the exposure center of the array in the ground coordinate system at the imaging epoch (t), and the corresponding ground point coordinates are related through the collinearity condition that states that all these three points must be on the same line. The rotation angles for each CCD array at the epoch t are defined and included in the transformation model. The exterior orientation parameters of an image line, i.e., coordinates of exposure station and rotation angles, are computed by a polynomial interpolation function in time (t). The parameter (t) is the time at a certain epoch from a certain orbit position. Depending on the types of observations, coordinates, and parameters may be treated as knowns or unknowns differently in various situations. The unknown coefficients are determined in a bundle adjustment. The orientation process starts by extracting the sensor position and, orientation and raw images from the PDS. The parameters of each image line are then estimated and imported into the push-broom sensor model. We also define tie points between image pairs to aid the bundle adjustment model, determine the refined camera parameters, and generate highly accurate topographic maps. The model was tested on different satellite images such as IKONOS, QuickBird, and WorldView-2, HiRISE. It was found that the accuracy of our model is comparable to those of commercial and open-source software, the computational efficiency of the developed model is high, the model could be used in different environments with various sensors, and the implementation process is much more cost-and effort-consuming.

Keywords: photogrammetry, push-broom sensors, IKONOS, HiRISE, collinearity condition

Procedia PDF Downloads 39

Authors: Nicola Colaninno, Eugenio Morello

Abstract:

The urban environment affects local-to-global climate and, in turn, suffers global warming phenomena, with worrying impacts on human well-being, health, social and economic activities. Physic-morphological features of the built-up space affect urban air temperature, locally, causing the urban environment to be warmer compared to surrounding rural. This occurrence, typically known as the Urban Heat Island (UHI), is normally assessed by means of air temperature from fixed weather stations and/or traverse observations or based on remotely sensed Land Surface Temperatures (LST). The information provided by ground weather stations is key for assessing local air temperature. However, the spatial coverage is normally limited due to low density and uneven distribution of the stations. Although different interpolation techniques such as Inverse Distance Weighting (IDW), Ordinary Kriging (OK), or Multiple Linear Regression (MLR) are used to estimate air temperature from observed points, such an approach may not effectively reflect the real climatic conditions of an interpolated point. Quantifying local UHI for extensive areas based on weather stations’ observations only is not practicable. Alternatively, the use of thermal remote sensing has been widely investigated based on LST. Data from Landsat, ASTER, or MODIS have been extensively used. Indeed, LST has an indirect but significant influence on air temperatures. However, high-resolution near-surface air temperature (NSAT) is currently difficult to retrieve. Here we have experimented Geographically Weighted Regression (GWR) as an effective approach to enable NSAT estimation by accounting for spatial non-stationarity of the phenomenon. The model combines on-site measurements of air temperature, from fixed weather stations and satellite-derived LST. The approach is structured upon two main steps. First, a GWR model has been set to estimate NSAT at low resolution, by combining air temperature from discrete observations retrieved by weather stations (dependent variable) and the LST from satellite observations (predictor). At this step, MODIS data, from Terra satellite, at 1 kilometer of spatial resolution have been employed. Two time periods are considered according to satellite revisit period, i.e. 10:30 am and 9:30 pm. Afterward, the results have been downscaled at 30 meters of spatial resolution by setting a GWR model between the previously retrieved near-surface air temperature (dependent variable), the multispectral information as provided by the Landsat mission, in particular the albedo, and Digital Elevation Model (DEM) from the Shuttle Radar Topography Mission (SRTM), both at 30 meters. Albedo and DEM are now the predictors. The area under investigation is the Metropolitan City of Milan, which covers an area of approximately 1,575 km2 and encompasses a population of over 3 million inhabitants. Both models, low- (1 km) and high-resolution (30 meters), have been validated according to a cross-validation that relies on indicators such as R2, Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). All the employed indicators give evidence of highly efficient models. In addition, an alternative network of weather stations, available for the City of Milano only, has been employed for testing the accuracy of the predicted temperatures, giving and RMSE of 0.6 and 0.7 for daytime and night-time, respectively.

Keywords: urban climate, urban heat island, geographically weighted regression, remote sensing

Procedia PDF Downloads 167

Authors: Meiyan Gao, Zongmin Wang, Haibo Yang, Qiuhua Liang

Abstract:

Under the background of global climate change and rapid development of modern urbanization, the frequency of climate disasters such as extreme precipitation in cities around the world is gradually increasing. In this paper, Hi-PIMS model is used to simulate the "720" flood in Zhengzhou, and the continuous stages of flood resilience are determined with the urban flood stages are divided. The flood resilience curve under the influence of multiple factors were determined and the urban flood toughness was evaluated by combining the results of resilience curves. The flood resilience of urban unit grid was evaluated based on economy, population, road network, hospital distribution and land use type. Firstly, the rainfall data of meteorological stations near Zhengzhou and the remote sensing rainfall data from July 17 to 22, 2021 were collected. The Kriging interpolation method was used to expand the rainfall data of Zhengzhou. According to the rainfall data, the flood process generated by four rainfall events in Zhengzhou was reproduced. Based on the results of the inundation range and inundation depth in different areas, the flood process was divided into four stages: absorption, resistance, overload and recovery based on the once in 50 years rainfall standard. At the same time, based on the levels of slope, GDP, population, hospital affected area, land use type, road network density and other aspects, the resilience curve was applied to evaluate the urban flood resilience of different regional units, and the difference of flood process of different precipitation in "720" rainstorm in Zhengzhou was analyzed. Faced with more than 1,000 years of rainstorm, most areas are quickly entering the stage of overload. The influence levels of factors in different areas are different, some areas with ramps or higher terrain have better resilience, and restore normal social order faster, that is, the recovery stage needs shorter time. Some low-lying areas or special terrain, such as tunnels, will enter the overload stage faster in the case of heavy rainfall. As a result, high levels of flood protection, water level warning systems and faster emergency response are needed in areas with low resilience and high risk. The building density of built-up area, population of densely populated area and road network density all have a certain negative impact on urban flood resistance, and the positive impact of slope on flood resilience is also very obvious. While hospitals can have positive effects on medical treatment, they also have negative effects such as population density and asset density when they encounter floods. The result of a separate comparison of the unit grid of hospitals shows that the resilience of hospitals in the distribution range is low when they encounter floods. Therefore, in addition to improving the flood resistance capacity of cities, through reasonable planning can also increase the flood response capacity of cities. Changes in these influencing factors can further improve urban flood resilience, such as raise design standards and the temporary water storage area when floods occur, train the response speed of emergency personnel and adjust emergency support equipment.

Keywords: urban flood resilience, resilience assessment, hydrodynamic model, resilience curve

Procedia PDF Downloads 18

Authors: H. K. Esfahani, B. Datta

Abstract:

Groundwater is one of the most important natural resources in many parts of the world; however it is widely polluted due to human activities. Currently, effective and reliable groundwater management and remediation strategies are obtained using characterization of groundwater pollution sources, where the measured data in monitoring locations are utilized to estimate the unknown pollutant source location and magnitude. However, accurately identifying characteristics of contaminant sources is a challenging task due to uncertainties in terms of predicting source flux injection, hydro-geological and geo-chemical parameters, and the concentration field measurement. Reactive transport of chemical species in contaminated groundwater systems, especially with multiple species, is a complex and highly non-linear geochemical process. Although sufficient concentration measurement data is essential to accurately identify sources characteristics, available data are often sparse and limited in quantity. Therefore, this inverse problem-solving method for characterizing unknown groundwater pollution sources is often considered ill-posed, complex and non- unique. Different methods have been utilized to identify pollution sources; however, the linked simulation-optimization approach is one effective method to obtain acceptable results under uncertainties in complex real life scenarios. With this approach, the numerical flow and contaminant transport simulation models are externally linked to an optimization algorithm, with the objective of minimizing the difference between measured concentration and estimated pollutant concentration at observation locations. Concentration measurement data are very important to accurately estimate pollution source properties; therefore, optimal design of the monitoring network is essential to gather adequate measured data at desired times and locations. Due to budget and physical restrictions, an efficient and effective approach for groundwater pollutant source characterization is to design an optimal monitoring network, especially when only inadequate and arbitrary concentration measurement data are initially available. In this approach, preliminary concentration observation data are utilized for preliminary source location, magnitude and duration of source activity identification, and these results are utilized for monitoring network design. Further, feedback information from the monitoring network is used as inputs for sequential monitoring network design, to improve the identification of unknown source characteristics. To design an effective monitoring network of observation wells, optimization and interpolation techniques are used. A simulation model should be utilized to accurately describe the aquifer properties in terms of hydro-geochemical parameters and boundary conditions. However, the simulation of the transport processes becomes complex when the pollutants are chemically reactive. Three dimensional transient flow and reactive contaminant transport process is considered. The proposed methodology uses HYDROGEOCHEM 5.0 (HGCH) as the simulation model for flow and transport processes with chemically multiple reactive species. Adaptive Simulated Annealing (ASA) is used as optimization algorithm in linked simulation-optimization methodology to identify the unknown source characteristics. Therefore, the aim of the present study is to develop a methodology to optimally design an effective monitoring network for pollution source characterization with reactive species in polluted aquifers. The performance of the developed methodology will be evaluated for an illustrative polluted aquifer sites, for example an abandoned mine site in Queensland, Australia.

Keywords: monitoring network design, source characterization, chemical reactive transport process, contaminated mine site

Procedia PDF Downloads 209

Authors: Pei Yi Lin

Abstract:

Objective: The objective of this study is to use machine learning methods to build an early prediction classifier model for acute delirium to improve the quality of medical care for intensive care patients. Background: Delirium is a common acute and sudden disturbance of consciousness in critically ill patients. After the occurrence, it is easy to prolong the length of hospital stay and increase medical costs and mortality. In 2021, the incidence of delirium in the intensive care unit of internal medicine was as high as 59.78%, which indirectly prolonged the average length of hospital stay by 8.28 days, and the mortality rate is about 2.22% in the past three years. Therefore, it is expected to build a delirium prediction classifier through big data analysis and machine learning methods to detect delirium early. Method: This study is a retrospective study, using the artificial intelligence big data database to extract the characteristic factors related to delirium in intensive care unit patients and let the machine learn. The study included patients aged over 20 years old who were admitted to the intensive care unit between May 1, 2022, and December 31, 2022, excluding GCS assessment <4 points, admission to ICU for less than 24 hours, and CAM-ICU evaluation. The CAMICU delirium assessment results every 8 hours within 30 days of hospitalization are regarded as an event, and the cumulative data from ICU admission to the prediction time point are extracted to predict the possibility of delirium occurring in the next 8 hours, and collect a total of 63,754 research case data, extract 12 feature selections to train the model, including age, sex, average ICU stay hours, visual and auditory abnormalities, RASS assessment score, APACHE-II Score score, number of invasive catheters indwelling, restraint and sedative and hypnotic drugs. Through feature data cleaning, processing and KNN interpolation method supplementation, a total of 54595 research case events were extracted to provide machine learning model analysis, using the research events from May 01 to November 30, 2022, as the model training data, 80% of which is the training set for model training, and 20% for the internal verification of the verification set, and then from December 01 to December 2022 The CU research event on the 31st is an external verification set data, and finally the model inference and performance evaluation are performed, and then the model has trained again by adjusting the model parameters. Results: In this study, XG Boost, Random Forest, Logistic Regression, and Decision Tree were used to analyze and compare four machine learning models. The average accuracy rate of internal verification was highest in Random Forest (AUC=0.86), and the average accuracy rate of external verification was in Random Forest and XG Boost was the highest, AUC was 0.86, and the average accuracy of cross-validation was the highest in Random Forest (ACC=0.77). Conclusion: Clinically, medical staff usually conduct CAM-ICU assessments at the bedside of critically ill patients in clinical practice, but there is a lack of machine learning classification methods to assist ICU patients in real-time assessment, resulting in the inability to provide more objective and continuous monitoring data to assist Clinical staff can more accurately identify and predict the occurrence of delirium in patients. It is hoped that the development and construction of predictive models through machine learning can predict delirium early and immediately, make clinical decisions at the best time, and cooperate with PADIS delirium care measures to provide individualized non-drug interventional care measures to maintain patient safety, and then Improve the quality of care.

Keywords: critically ill patients, machine learning methods, delirium prediction, classifier model

Procedia PDF Downloads 36