Smart Healthcare System Design. Группа авторов

Чтение книги онлайн.

      1

      Machine Learning Technologies in IoT EEG-Based Healthcare Prediction

       Karthikeyan M.P.1*, Krishnaveni K.2 and Muthumani N.3

      ¹Department of Computer Science, PPG College of Arts and Science, Coimbatore, India

      ²Department of Computer Science, Sri Ramasamy Naidu Memorial College, Sattur, India

       ³ PPG College of Arts and Science, Coimbatore, India

       Abstract

      The classification of medical data is the demanding challenge to be addressed among all research issues since it provides a larger business value in any analytics environment. Medical data classification is a mechanism that labels data enabling economical and effective performance in valuable analysis. Proposed research has indicated that the quality of the features may cause a backlash to the classification performance. Also squeezing the classification model with entire raw features can create a bottleneck to the classification performance. Thus, there is necessity for selecting appropriate features for training the classifier. In this proposed, a system is proposed that can use multiple channel real-time EEG signals to predict the onset of an epileptic seizure. The system is given a select number of EEG channels as input and reports back the corresponding epileptic seizure state at every second and the Hybrid Artificial Neural Network with Support Vector Machine (HANNSVM) based classifications are done as a simulation of real-time dynamic predictions and are dependent upon past predictions that were made. As a result, the sensitivity must be controlled such that seizures aren’t predicted more often than they actually occur. Statistical analysis of accuracy values and computational time portrays that the proposed schemes provide compromising results over existent methods.

      Keywords: Computer aided diagnosis, K-nearest neighbor, artificial neural network, electroencephalography, Internet of Things, support vector machine, brain modeling feature exraction

1.1 Introduction

      IoT (Internet of Things) is utilized as a part of a great deal of medical uses. A portion of the uses of Internet of Things are savvy stopping, shrewd home, brilliant city, keen condition, mechanical spots, horticulture fields and wellbeing observing procedure [38]. One such application in medicinal services to screen the patient’s wellbeing status by means of Internet of Things makes therapeutic gear more effective by permitting ongoing checking of patient’s wellbeing, in which sensor get information of patient’s and decreases the human blunder. The Internet of Things in the therapeutic field draws out the answer for compelling continuous checking of rationally impaired individual at diminished cost and furthermore lessens the exchange off between tolerant result and infection administration [33]. So far we have seen the wellbeing observing framework which gathers data of fundamental parameters, for example, heartbeat, temperature, circulatory strain and development parameters. The medical data stored in cloud in the form of huge dataset, need to analyze and predict the diseases based on IoT data is very important [1, 37].

The progress of science has driven every individual to mine and consume medical data for analyses of business, customer, bank account, medical, etc. made privacy break or intrusion also in most circumstances. The IoT-based medical data is all over in the pattern of text, number, images and videos [35, 36]. This type of data continues to grow bigger, thereby organizing these data as a necessary process. The collected enormous data should produce logical use unless it would be waste of time, effort and storage. The action of grabbing or collection of huge data is called datafication. Clinical data can be used effectively as it is datafied. The organizing of data alone cannot make useful but should identify what can be performed by its use. Optimal processing power, analytical capabilities and skills are needed for squeezing essential information from medical data. The data mining features are shown in Figure 1.1.

      Medical data is of various types, formats and shapes which are brought together from various sources. Data Analytics is the action of studying and extracting big data which can yield functional and business knowledge in a remarkable form. The behavior of business is reconstituted in different ways by big data analytics [15]. Approaches like information technology, statistics, quantitative methods and various methods are used by medical analytics to deliver results. Data mining analytics is divided into three main types. They are descriptive analytics, predictive analytics and prescriptive analytics. The traditional database systems are not sufficient to progress huge data characteristics (elements) [2].

Figure 1.1 Data mining features.

      1.1.1 Descriptive Analytics

      Descriptive analytical type is the best accepted one being the basis for uplifted analytical models. It benefits leaders, researchers, planners, etc. to build a guideline for forthcoming activities by reviewing the database to determine knowledge on current or past medical data proceedings [16]. This model does a detailed review of data to expose particulars like operation costs, cause for false steps and frequency of events. Descriptive analytics assists locating the root cause of the issue. Descriptive analytics also deal with it Proposed modal EEG classification [4].

      1.1.2 Analytical Methods

      The different analytical methods of data mining are

      1.1.3 Predictive Analysis

      The probable questions in predictive analysis are