Breast cancer stands as one of the world’s most perilous and formidable diseases,having recently surpassed lung cancer as the most prevalent cancer type.This disease arises when cells in the breast undergo unregulate...Breast cancer stands as one of the world’s most perilous and formidable diseases,having recently surpassed lung cancer as the most prevalent cancer type.This disease arises when cells in the breast undergo unregulated proliferation,resulting in the formation of a tumor that has the capacity to invade surrounding tissues.It is not confined to a specific gender;both men and women can be diagnosed with breast cancer,although it is more frequently observed in women.Early detection is pivotal in mitigating its mortality rate.The key to curbing its mortality lies in early detection.However,it is crucial to explain the black-box machine learning algorithms in this field to gain the trust of medical professionals and patients.In this study,we experimented with various machine learning models to predict breast cancer using the Wisconsin Breast Cancer Dataset(WBCD)dataset.We applied Random Forest,XGBoost,Support Vector Machine(SVM),Multi-Layer Perceptron(MLP),and Gradient Boost classifiers,with the Random Forest model outperforming the others.A comparison analysis between the two methods was done after performing hyperparameter tuning on each method.The analysis showed that the random forest performs better and yields the highest result with 99.46%accuracy.After performance evaluation,two Explainable Artificial Intelligence(XAI)methods,SHapley Additive exPlanations(SHAP)and Local Interpretable Model-Agnostic Explanations(LIME),have been utilized to explain the random forest machine learning model.展开更多
Analytical investigation on a combined heat and mass transfer of air flow near a continuously moving infinite plate with a constant heat sink is performed in the presence of a uniform magnetic field. To observe the th...Analytical investigation on a combined heat and mass transfer of air flow near a continuously moving infinite plate with a constant heat sink is performed in the presence of a uniform magnetic field. To observe the thermal radiation and Soret effect on the flow, thermal radiation and thermal diffusion term are added in energy and concentration equations. A flow of model is established by employing the well known boundary layer approximations. In order to obtain non-dimensional system of equations, a similarity transformation is applied on the flow model. Perturbation technique is used as main tool for the analytical approach. The numerical values of flow variables are computed by a FORTRAN program. The obtain numerical values of fluid velocity, temperature and species concentration are drawn for the different values of various parameters. To observe the effects of various parameters on the flow variables, the results are discussed in detailed with the help of graph.展开更多
Disasters may occur at any time and place without little to no presage in advance.With the development of surveillance and forecasting systems,it is now possible to forebode the most life-threatening and formidable di...Disasters may occur at any time and place without little to no presage in advance.With the development of surveillance and forecasting systems,it is now possible to forebode the most life-threatening and formidable disasters.However,forestfires are among the ones that are still hard to anticipate beforehand,and the technologies to detect and plot their possible courses are still in development.Unmanned Aerial Vehicle(UAV)image-basedfire detection systems can be a viable solution to this problem.However,these automatic systems use advanced deep learning and image processing algorithms at their core and can be tuned to provide accurate outcomes.Therefore,this article proposed a forestfire detection method based on a Convolutional Neural Network(CNN)architecture using a newfire detection dataset.Notably,our method also uses separable convolution layers(requiring less computational resources)for immediatefire detection and typical convolution layers.Thus,making it suitable for real-time applications.Consequently,after being trained on the dataset,experimental results show that the method can identify forestfires within images with a 97.63%accuracy,98.00%F1 Score,and 80%Kappa.Hence,if deployed in practical circumstances,this identification method can be used as an assistive tool to detectfire outbreaks,allowing the authorities to respond quickly and deploy preventive measures to minimize damage.展开更多
Colorization is the practice of adding appropriate chromatic values to monochrome photographs or videos.A real-valued luminance image can be mapped to a three-dimensional color image.However,it is a severely ill-defin...Colorization is the practice of adding appropriate chromatic values to monochrome photographs or videos.A real-valued luminance image can be mapped to a three-dimensional color image.However,it is a severely ill-defined problem and not has a single solution.In this paper,an encoder-decoder Convolutional Neural Network(CNN)model is used for colorizing gray images where the encoder is a Densely Connected Convolutional Network(DenseNet)and the decoder is a conventional CNN.The DenseNet extracts image features from gray images and the conventional CNN outputs a^(*)b^(*)color channels.Due to a large number of desaturated color components compared to saturated color components in the training images,the saturated color components have a strong tendency towards desaturated color components in the predicted a^(*)b^(*)channel.To solve the problems,we rebalance the predicted a^(*)b^(*)color channel by smoothing every subregion individually using the average filter.2 stage k-means clustering technique is applied to divide the subregions.Then we apply Gamma transformation in the entire a^(*)b^(*)channel to saturate the image.We compare our proposed method with several existing methods.From the experimental results,we see that our proposed method has made some notable improvements over the existing methods and color representation of gray-scale images by our proposed method is more plausible to visualize.Additionally,our suggested approach beats other approaches in terms of Peak Signal-to-Noise Ratio(PSNR),Structural Similarity Index Measure(SSIM)and Histogram.展开更多
One of the most dangerous diseases that affect people worldwide is lung cancer.The survival rate is minimal,because of the complexity in identifying lung cancer at developed stages.Henceforth,earlier detection of lung...One of the most dangerous diseases that affect people worldwide is lung cancer.The survival rate is minimal,because of the complexity in identifying lung cancer at developed stages.Henceforth,earlier detection of lung cancer is significant.Several Machine Learning(ML)approaches have been modeled for lung cancer recognition with the advent of Artificial Intelligence.However,small-scale datasets and deprived generalizability to recognize unknown data are considered challenges in lung cancer detection.This work proposes an advanced deep learning model,named Generative Adversarial Network-Attention Gated Network(GA-AGN),which is the integration of Generative Adversarial Network(GAN)and Attention Gated Network(AGN).Initially,the chest CT scan images are subjected to the pre-processing phase,where image resizing and normalization are used to preprocess the images.Then,the data augmentation is performed using the GAN model that is trained by Elk Herd Optimizer(EHO).Subsequently,lung cancer detection is done by means of GA-AGN model.Ultimately analysis is performed by using three measures,like accuracy,sensitivity as well as specificity with values of 0.938,0.948 and 0.927.The overall analysis states that the proposed model attained better outcomes than the conventional models.展开更多
Ischemic heart disease(IHD)is one of the leading causes of death worldwide.However,different geographic regions show different variations of the risk factors of this disease based on the different lifestyles of people...Ischemic heart disease(IHD)is one of the leading causes of death worldwide.However,different geographic regions show different variations of the risk factors of this disease based on the different lifestyles of people.This study examines the current IHD condition in southern Bangladesh,a Southeast Asian middle-income country.The main approach to this research is an Al-based proposal of a reduced set of the greatest impact clinical traits that may cause IHD.This approach attempts to reduce IHD morbidity and mortality by early detection of risk factors using the reduced set of clinical data.Demographic,diagnostic,and symptomatic features were considered for analysing this clinical data.Data pre-processing utilizes several machine learning techniques to select significant features and make meaningful interpretations.A proposed voting mechanism ranked the selected 138 features by their impact factor.In this regard,diverse patterns in correlations with variables,including age,sex,career,family history,obesity,etc.,were calculated and explained in terms of voting scores.Among the 138 risk factors,three labels were categorized:high-risk,medium-risk,and low-risk features;19 features were regarded as high,25 were medium,and 94 were considered low impactful features.This research's technological methodology and practical goals provide an innovative and resilient framework for addressing IHD,especially in less developed cities and townships of Bangladesh,where the general population's socioeconomic conditions are often unexpected.The data collection,pre-processing,and use of this study's complete and comprehensive IHD patient dataset is another innovative addition.We believe that other relevant research initiatives will benefit from this work.展开更多
基金supported by the Researchers Supporting Project(RSPD2024R846),King Saud University,Riyadh,Saudi Arabia.
文摘Breast cancer stands as one of the world’s most perilous and formidable diseases,having recently surpassed lung cancer as the most prevalent cancer type.This disease arises when cells in the breast undergo unregulated proliferation,resulting in the formation of a tumor that has the capacity to invade surrounding tissues.It is not confined to a specific gender;both men and women can be diagnosed with breast cancer,although it is more frequently observed in women.Early detection is pivotal in mitigating its mortality rate.The key to curbing its mortality lies in early detection.However,it is crucial to explain the black-box machine learning algorithms in this field to gain the trust of medical professionals and patients.In this study,we experimented with various machine learning models to predict breast cancer using the Wisconsin Breast Cancer Dataset(WBCD)dataset.We applied Random Forest,XGBoost,Support Vector Machine(SVM),Multi-Layer Perceptron(MLP),and Gradient Boost classifiers,with the Random Forest model outperforming the others.A comparison analysis between the two methods was done after performing hyperparameter tuning on each method.The analysis showed that the random forest performs better and yields the highest result with 99.46%accuracy.After performance evaluation,two Explainable Artificial Intelligence(XAI)methods,SHapley Additive exPlanations(SHAP)and Local Interpretable Model-Agnostic Explanations(LIME),have been utilized to explain the random forest machine learning model.
文摘Analytical investigation on a combined heat and mass transfer of air flow near a continuously moving infinite plate with a constant heat sink is performed in the presence of a uniform magnetic field. To observe the thermal radiation and Soret effect on the flow, thermal radiation and thermal diffusion term are added in energy and concentration equations. A flow of model is established by employing the well known boundary layer approximations. In order to obtain non-dimensional system of equations, a similarity transformation is applied on the flow model. Perturbation technique is used as main tool for the analytical approach. The numerical values of flow variables are computed by a FORTRAN program. The obtain numerical values of fluid velocity, temperature and species concentration are drawn for the different values of various parameters. To observe the effects of various parameters on the flow variables, the results are discussed in detailed with the help of graph.
文摘Disasters may occur at any time and place without little to no presage in advance.With the development of surveillance and forecasting systems,it is now possible to forebode the most life-threatening and formidable disasters.However,forestfires are among the ones that are still hard to anticipate beforehand,and the technologies to detect and plot their possible courses are still in development.Unmanned Aerial Vehicle(UAV)image-basedfire detection systems can be a viable solution to this problem.However,these automatic systems use advanced deep learning and image processing algorithms at their core and can be tuned to provide accurate outcomes.Therefore,this article proposed a forestfire detection method based on a Convolutional Neural Network(CNN)architecture using a newfire detection dataset.Notably,our method also uses separable convolution layers(requiring less computational resources)for immediatefire detection and typical convolution layers.Thus,making it suitable for real-time applications.Consequently,after being trained on the dataset,experimental results show that the method can identify forestfires within images with a 97.63%accuracy,98.00%F1 Score,and 80%Kappa.Hence,if deployed in practical circumstances,this identification method can be used as an assistive tool to detectfire outbreaks,allowing the authorities to respond quickly and deploy preventive measures to minimize damage.
基金Taif University Researchers Supporting Project Number(TURSP-2020/10),Taif University,Taif,Saudi Arabia.
文摘Colorization is the practice of adding appropriate chromatic values to monochrome photographs or videos.A real-valued luminance image can be mapped to a three-dimensional color image.However,it is a severely ill-defined problem and not has a single solution.In this paper,an encoder-decoder Convolutional Neural Network(CNN)model is used for colorizing gray images where the encoder is a Densely Connected Convolutional Network(DenseNet)and the decoder is a conventional CNN.The DenseNet extracts image features from gray images and the conventional CNN outputs a^(*)b^(*)color channels.Due to a large number of desaturated color components compared to saturated color components in the training images,the saturated color components have a strong tendency towards desaturated color components in the predicted a^(*)b^(*)channel.To solve the problems,we rebalance the predicted a^(*)b^(*)color channel by smoothing every subregion individually using the average filter.2 stage k-means clustering technique is applied to divide the subregions.Then we apply Gamma transformation in the entire a^(*)b^(*)channel to saturate the image.We compare our proposed method with several existing methods.From the experimental results,we see that our proposed method has made some notable improvements over the existing methods and color representation of gray-scale images by our proposed method is more plausible to visualize.Additionally,our suggested approach beats other approaches in terms of Peak Signal-to-Noise Ratio(PSNR),Structural Similarity Index Measure(SSIM)and Histogram.
文摘One of the most dangerous diseases that affect people worldwide is lung cancer.The survival rate is minimal,because of the complexity in identifying lung cancer at developed stages.Henceforth,earlier detection of lung cancer is significant.Several Machine Learning(ML)approaches have been modeled for lung cancer recognition with the advent of Artificial Intelligence.However,small-scale datasets and deprived generalizability to recognize unknown data are considered challenges in lung cancer detection.This work proposes an advanced deep learning model,named Generative Adversarial Network-Attention Gated Network(GA-AGN),which is the integration of Generative Adversarial Network(GAN)and Attention Gated Network(AGN).Initially,the chest CT scan images are subjected to the pre-processing phase,where image resizing and normalization are used to preprocess the images.Then,the data augmentation is performed using the GAN model that is trained by Elk Herd Optimizer(EHO).Subsequently,lung cancer detection is done by means of GA-AGN model.Ultimately analysis is performed by using three measures,like accuracy,sensitivity as well as specificity with values of 0.938,0.948 and 0.927.The overall analysis states that the proposed model attained better outcomes than the conventional models.
文摘Ischemic heart disease(IHD)is one of the leading causes of death worldwide.However,different geographic regions show different variations of the risk factors of this disease based on the different lifestyles of people.This study examines the current IHD condition in southern Bangladesh,a Southeast Asian middle-income country.The main approach to this research is an Al-based proposal of a reduced set of the greatest impact clinical traits that may cause IHD.This approach attempts to reduce IHD morbidity and mortality by early detection of risk factors using the reduced set of clinical data.Demographic,diagnostic,and symptomatic features were considered for analysing this clinical data.Data pre-processing utilizes several machine learning techniques to select significant features and make meaningful interpretations.A proposed voting mechanism ranked the selected 138 features by their impact factor.In this regard,diverse patterns in correlations with variables,including age,sex,career,family history,obesity,etc.,were calculated and explained in terms of voting scores.Among the 138 risk factors,three labels were categorized:high-risk,medium-risk,and low-risk features;19 features were regarded as high,25 were medium,and 94 were considered low impactful features.This research's technological methodology and practical goals provide an innovative and resilient framework for addressing IHD,especially in less developed cities and townships of Bangladesh,where the general population's socioeconomic conditions are often unexpected.The data collection,pre-processing,and use of this study's complete and comprehensive IHD patient dataset is another innovative addition.We believe that other relevant research initiatives will benefit from this work.
