jean pierre Namahoro - Academia.edu (original) (raw)
Papers by jean pierre Namahoro
Science Journal of Applied Mathematics and Statistics, 2019
The standard incubator used to monitor the development of preterm infants, with much attention fo... more The standard incubator used to monitor the development of preterm infants, with much attention for random optimization can interrupt the three main parameters (oxygen, environmental temperature, and humidity) responsible for preterm growth. The artificial neural network (ANN) has been recently proposed as a novel technique to control those parameters to provide a better and stabilized environment in an incubator. Unfortunately, this novel technique cannot continuously provide and indicate the update challenge of preterm growth. The objective of this paper is to apply a Markov multi-state growth process incorporates with multilayer feed-forward artificial neural network as an improved methodology to continuously control and provide an update of preterm growth in an incubator. The exchangeable Markov growth process, transition graph, and artificial neural network discussed on and applied in the designed incubator as methodology in paper and then make a joint density function of Markov multi-states growth process through multi-steps designed Algorithm to get the theoretical results. The updated measurements (weight, height, and head-perimeter) associated with controlled parameters used as input to the threshold logic unit (TLU) of ANN and then distinguish whether the growth process is abnormal or normal at each state. The summarized algorithm and multilayer feed-forward ANN utilized the panel data collected at Murunda hospital in Rwanda as input to show the application of improved methodology proposed in this paper, specifically, multi-state growth process of preterm infants across gender. As results, the continuous exchangeability of the growth process at each state has updated and may show abnormal or normal of growth process, and then sensors may notify these change through the joint density function of Markov multi-states growth process. Thus, improved methodology can increase the security and minimize time consumption in continuous monitoring growth process in an advanced way in time this idea has been implemented.
Applied and Computational Mathematics, 2020
This study focused on exploiting machine learning algorithms for classifying and predicting injur... more This study focused on exploiting machine learning algorithms for classifying and predicting injury severity of vehicle crashes in Yemen. The primary objective is to assess the contribution of the leading causes of injury severity. The selected machine learning algorithms compared with traditional statistical methods. The filtrated second data collected within two months (August-October 2015) from the two main hospitals included 156 injured patients of vehicle crashes reported from 128 locations. The data classified into three categories of injury severity: Severe, Serious, and Minor. It balanced using a synthetic minority oversampling technique (SMOTE). Multinomial logit model (MNL) compared with five machine learning classifiers: Naïve Bayes (NB), J48 Decision Tree, Random Forest (RF), Support Vector Machine (SVM), and Multilayer Perceptron (MLP). The results showed that most of machine learning-based algorithms performed well in predicting and classifying the severity of the traffic injury. Out of five classifiers, RF is the best classifier with 94.84% of accuracy. The characteristics of road type, total injured person, crash type, road user, transport way to the emergency department (ED), and accident action were the most critical factors in the severity of the traffic injury. Enhancing strategies for using roadway facilities may improve the safety of road users and regulations.
Science Journal of Applied Mathematics and Statistics, 2019
The standard incubator used to monitor the development of preterm infants, with much attention fo... more The standard incubator used to monitor the development of preterm infants, with much attention for random optimization can interrupt the three main parameters (oxygen, environmental temperature, and humidity) responsible for preterm growth. The artificial neural network (ANN) has been recently proposed as a novel technique to control those parameters to provide a better and stabilized environment in an incubator. Unfortunately, this novel technique cannot continuously provide and indicate the update challenge of preterm growth. The objective of this paper is to apply a Markov multi-state growth process incorporates with multilayer feed-forward artificial neural network as an improved methodology to continuously control and provide an update of preterm growth in an incubator. The exchangeable Markov growth process, transition graph, and artificial neural network discussed on and applied in the designed incubator as methodology in paper and then make a joint density function of Markov multi-states growth process through multi-steps designed Algorithm to get the theoretical results. The updated measurements (weight, height, and head-perimeter) associated with controlled parameters used as input to the threshold logic unit (TLU) of ANN and then distinguish whether the growth process is abnormal or normal at each state. The summarized algorithm and multilayer feed-forward ANN utilized the panel data collected at Murunda hospital in Rwanda as input to show the application of improved methodology proposed in this paper, specifically, multi-state growth process of preterm infants across gender. As results, the continuous exchangeability of the growth process at each state has updated and may show abnormal or normal of growth process, and then sensors may notify these change through the joint density function of Markov multi-states growth process. Thus, improved methodology can increase the security and minimize time consumption in continuous monitoring growth process in an advanced way in time this idea has been implemented.
Applied and Computational Mathematics, 2020
This study focused on exploiting machine learning algorithms for classifying and predicting injur... more This study focused on exploiting machine learning algorithms for classifying and predicting injury severity of vehicle crashes in Yemen. The primary objective is to assess the contribution of the leading causes of injury severity. The selected machine learning algorithms compared with traditional statistical methods. The filtrated second data collected within two months (August-October 2015) from the two main hospitals included 156 injured patients of vehicle crashes reported from 128 locations. The data classified into three categories of injury severity: Severe, Serious, and Minor. It balanced using a synthetic minority oversampling technique (SMOTE). Multinomial logit model (MNL) compared with five machine learning classifiers: Naïve Bayes (NB), J48 Decision Tree, Random Forest (RF), Support Vector Machine (SVM), and Multilayer Perceptron (MLP). The results showed that most of machine learning-based algorithms performed well in predicting and classifying the severity of the traffic injury. Out of five classifiers, RF is the best classifier with 94.84% of accuracy. The characteristics of road type, total injured person, crash type, road user, transport way to the emergency department (ED), and accident action were the most critical factors in the severity of the traffic injury. Enhancing strategies for using roadway facilities may improve the safety of road users and regulations.