Machine Learning Applications for Classification Emergency and Non-Emergency Patients

Emergency departments of hospitals are busy. In recent years, patient arrivals have significantly risen at emergency departments in Turkey like other countries in the world. The main important features of emergency services are uninterrupted service, providing services in a short time, and priority to emergency patients. However, patients who do not need immediate treatment can sometimes apply to this department due to several reasons like working time and short waiting time. This situation can reduce efficiency and effectiveness at emergency departments. On the other hand, computers solve complex classification problems by using machine learning methods. The methods have a wide range of applications, such as computational biology and computer vision. Therefore, classification of emergency and non-emergency patients is vital to increase productivity of the department. This chapter tries to find the best classifier for detection of emergency patients by utilizing a data set.

Computational Intelligence Approach for Classification of Diabetes Mellitus Using Decision Tree

Diabetes mellitus (DM) is a group of metabolic disorders with one common manifestation: elevated blood sugar or hyperglycemia. The diagnosis of diabetes is the most crucial point due to chronic hyperglycemia. This chapter improves the performance of the Classification and Regression Trees (CART) algorithm because the accurate classification of diabetes depends on the algorithm efficiency. Authors use the accuracy rate for the objective function in the prediction process by Genetic Algorithm (GA). The proposed GA-CART algorithm provides the best performance at 96.05%.

A Fuzzy Decision-Making Model for the Key Performance Indicators of Hospital Service Quality Evaluation

One of the major concerns of the healthcare industry throughout the world is to provide better hospital service quality. Management and delivery of hospital healthcare services are achieved in a competitive environment in Turkey. For this reason, to make better decisions, the services provided by the public and private hospitals should be monitored and evaluated according to the viewpoint of medical stakeholders. This chapter presents a cause-and-effect, decision-making model in evaluating hospital service quality criteria. Since the decision-making process involves the vagueness of human judgments, a combination of fuzzy sets and decision-making trial and evaluation laboratory (DEMATEL) is used. Results of the study demonstrate that medical equipment level of the hospital, the attitude of nurses and medical staff to patients, pharmacists' advice on medicine preservation, medical staff with professional abilities, outpatient waiting time for medical treatment, and number and quality of the bathrooms available have more impact on the entire hospital service quality. In conclusion, the proposed approach will contribute to better providing of healthcare services at a higher quality level.

Deep Learning in the Healthcare Industry

Artificial Neural networks (ANN) are composed of nodes that are joint to each other through weighted connections. Deep learning, as an extension of ANN, is a neural network model, but composed of different categories of layers: input layer, hidden layers, and output layers. Input data is fed into the first (input) layer. But the main process of the neural network models is done within the hidden layers, ranging from a single hidden layer to multiple ones. Depending on the type of model, the structure of the hidden layers is different. Depending on the type of input data, different models are applied. For example, for image data, convolutional neural networks are the most appropriate. On the other hand, for text or sequential and time series data, recurrent neural networks or long short-term memory models are the better choices. This chapter summarizes the state-of-the-art deep learning methods applied to the healthcare industry.

Modeling and Forecasting the Daily Number of Emergency Department Visits Using Hybrid Models

In this study, linear regression and neural network-based hybrid models are developed for modelling the daily ED visits. Month and week of the year, day of the week, and period of the day, are used as input variables of the linear regression model. Generated forecasts and the residuals are further processed through a multilayer perceptron model to improve the performance of forecasting. To obtain forecasts for daily number of patient visits, aggregation is used where the obtained periodical forecasts are summed up. By comparing the performances of models in generating periodical and daily forecasts, this chapter not only shows that hybrid model improves the forecasting performance significantly, but also aggregation fits well in practice.

A Survey and Analysis to Find Dental Implant Issues in India

Presently, dental implantations are the ideal solution and best substitute option for missing teeth. Dental implants determined the replacements of root and non-working teeth after the loss of natural teeth. Strengthening dental materials helps to fix the loss teeth. For teeth with root disease and fault in bone density, dental implantation is very essential. While bridge work and dentures decline to reclaim the suitable chewing condition and turn the problem of missing teeth, dental implants trace the solution of missing teeth. For more progress in dental-concerned technology, an extensive growth in dental implants has been noticed during the last several years. So there is a large need of excellent, quality dental materials. The selection of the implant materials, their generation system, manufacturing method, age-long durability, bio compatibility with medical principle, and professional dental field of study are therefore essential.

Smart Cataract Detector

Eyes react to light. Cataract, a common eye disease, clouds the lens of the eye, which decreases vision and can cause blindness. It can be in one or both eyes. The most major cause of cataract is ageing. Cataract also can be found in children and can occur due to eye injuries and inflammation. As mobile technology emerges to a greater extent, designing an application would help people perform a self check-up without meeting the doctors in person, thus saving time and cost. The pre-diagnosis of cataracts is costly for poor people. Time taken by the machine to detect cataract is greater, so the doctor can't attend as many patients. A cheap, reliable mobile application that helps diagnose cataract would cut cost and simplify diagnosis. This chapter proposes the development of smart phone application to detect the presence of cataract using image processing techniques.

Detection of Chronic Disease in Primary Care Using Artificial Intelligence Techniques

Chronic diseases are the leading causes of death and disability worldwide. By 2020, it is expected to increase to 73% of all deaths and 60% of global burden of disease associated with chronic diseases. For all these reasons, early diagnosis and treatment of chronic diseases is very important. Machine learning is an application of artificial intelligence that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning is the development of computer programs that can access data and use it to learn for themselves. The learning process starts by searching for patterns in the examples, experiences, or observations. It will make faster and better decisions in the future based on all these. The primary purpose in machine learning is to allow computers to learn automatically without human help and affect. Considering all the reasons above, this chapter finds the most appropriate artificial intelligence technique for the early detection of chronic diseases.

A Maximal Covering Location Model for the Organ Transplant Organization in Turkey

Transplantation of the organs is one of the principal treatment techniques for the people who have critical organ issues. One of the main steps of this treatment is getting the organ as quickly as possible to its recipient from a potential donor. To provide the aforementioned treatment, the spatial distributions of the organ transplant centers which maximize the donors/patients coverage become essential. In this chapter, the organ transplant process of Turkey which has 99 transplant centers (hospitals) in 81 cities is considered. The main goal of this study is to maximize the donor/patient coverage within the system constraints. As a potential demand source, 923 districts are determined using geographic information system (GIS). With the application of maximal covering location model, the number of covered donors/patients is maximized considering the cold ischemia time of organs. Optimal results are obtained for different scenarios based on heart, liver, and kidney.

NARX Neural Networks Model for Forecasting Daily Patient Arrivals in the Emergency Department

Regarding measuring of service quality at the emergency departments (ED), essential parameters are length of stay (LOS) and waiting times. Patient arrivals, which is related to LOS and waiting times, is hard to forecast and is affected by many parameters. Therefore, authors employed a Nonlinear Autoregressive Exogenous (NARX) model for forecasting of ED arrivals. NARX models are used extensively in many applications that show non-linear and dynamic behavior, but as far as authors know, the NARX method has not yet been used in the forecast of ED arrivals before. In this study, calendar and climatic variables are defined as input parameters. Patient Arrivals is defined as output parameter. A commercial software, MATLAB, was used to train and test the data set. To find the best network architecture Levenberg-Marquardt (LM) and Bayesian Regularization (BR) algorithms, different lags, and number of neurons were tested. R-squared and mean square error (MSE) are used to evaluate the accuracy of the tested networks.