Automatic speech recognition system
The Speech recognition systems are difficult and challenging as the same word can be spelled in many ways also change in accent has a huge impact on the accuracy of speech recognition. Most of the ASR systems in use today are designed to recognize speech in English. The major objective of this research is to design an ASR system which recognizes discrete words from Hindi language & controls some action depending on the provided input. The input voce is captured using a microphone which are then preprocessed using several algorithms like Dynamic Time Wrapping (DTW), Hidden Markov Model (HMM) etc. This paper aims at developing a simplified technique for recognition of speech spoken in the Hindi language by first modeling the system on computer-based design and then deploying it on an embedded system.
Published by: Ishrat Sultana, Nirmaljeet Kaur Pannu
Author: Ishrat Sultana
Paper ID: V4I6-1257
Paper Status: published
Published: November 24, 2018
Coronary heart disease monitoring system based on wireless sensors
An intelligent cardiac auscultation is the process of monitoring the heart beat signals variations of a patient monitoring system for monitoring the patients health condition automatically through sensors based connected networks in Internet of Things (IoT). It detects the critical condition of a patient by processing sensors data and instantly provides push notification to doctors. There is no process of monitoring of a particular cardiac disease which can lead to loss of life due to improper checkups and not following the lifestyle proposed by the doctor. In our proposed system, coronary heart disease monitoring based on wireless sensors are used to monitor cardiac patient for 24/7 without any human intervention using piezoelectric sensors which is used to measure artery thickness by the flow of blood vessels and extract the waveform. The waveform is classified into normal and abnormal waveform. These abnormal waveform is sent to mobile application where it receives the data and plots the signal curves in real-time. The mobile application acts as the display device and has the capability to upload data to a cloud platform for further analysis and an intimation is sent to the cardiologist. The authorized Cardiologist can get access to the cloud platform to get the dataset and results via any peripheral devices which are equipped with specific software and diagnose the results.
Published by: Esther Grace M, Dr. S. Sobitha Ahila, P. Hari Kumar
Author: Esther Grace M
Paper ID: V4I6-1166
Paper Status: published
Published: November 23, 2018
The classification scheme for the heart disease prediction
With the enormous enhancement of diseases in medical and the other communities of healthcare, it is extremely important to have an analysis of the heart diseases at the early stages. Since, nowadays it is very important to detect the diseases and lessen the death of patients at early stages. Every person has different values of cholesterol, blood pressure and many more that are linked with heart disease prediction. But it has scientifically proven that the normal person blood pressure is counted to be 120/90 along with this the pulse rate and the cholesterol value is 72. In this paper, the various “machine learning algorithms” are explained that include Support Vector Machine, Decision tree, neural network and many more are explained so that complete description can be provided. Along with this, the entire description of the heart disease has been provided that depicts about the need for the topic to be selected. There are some of the issues present in the Data Mining algorithm that are also described in the paper. The ultimate aim is to improve efficiency in different parameters by describing the classification approach for detecting heart disease. The parameters on which the prediction can be done are the age, serum cholesterol, gender, blood pressure, pulse rate. The accuracy and the efficiency in the prediction can be increased only if the number of attributes is more. For the classification of heart disease, the most efficient algorithm is the Support Vector Machine algorithm since it will not only reduce the prediction time but will also improve the efficiency of the algorithm.
Published by: Saloni Kapoor, Ashwinder Tanwar
Author: Saloni Kapoor
Paper ID: V4I6-1260
Paper Status: published
Published: November 23, 2018
Diabetes analysis using machine learning methods
In this paper, various kinds of algorithms are explained that include Support Vector Machine. The aim is to improve efficiency in different parameters by describing the classification approach for detecting diabetes. In this, it will predict diabetes with SVM.SVM will classify the data into positive and negative data points. In this, we predict the diabetes of Type 1and Type 2.Type 1is a type of diabetes that has no cure. Type 2 diabetes is common diabetes. It develops from child. Diabetes is the fastest growing problem with more health and economic results. The increasing rate is predicted to increase to 430 million. Different types of data mining techniques are used. With SVM it will predict better accuracy. When we will predict the result with SVM, it will give accuracy. With prediction of different parameters, we can predict the target value. With diabetes, there can be eye blindness, stress and many more can happen. With the help of data mining, we can aware about diabetes. In this paper, mention all the data mining techniques, types of classifiers. At the end, In this paper describes the diabetes types and what we have done and accuracy of the data. Type 2 diabetes is not easy to predict all the effects.
Published by: Harwinder Kaur, Gurleen Kaur
Author: Harwinder Kaur
Paper ID: V4I6-1261
Paper Status: published
Published: November 23, 2018
Ovarian cancer sign, symptoms and detection techniques
The current medical field is improved in many ways by accessing the applications of the technology. Digital image processing is one of them that being fascinating for researchers as well as for doctors such as ultrasound images and others. Due to various reasons, the diseases are growing rapidly, Cancer is one of them. Basically, cancer is a disease in which the blood cells grow uncontrollably and abnormal that causes diseases. Ovarian cancer is the most occurred form of the disease in females and every year the majority of females are survived from it. The cancer is produced in the ovaries and spread in the other parts of the body. The detection and diagnosis are crucial in the early stages because of the diagnosis become harder at the last stages. In the research, a deep representation of ovarian cancer is described as its generating process, signs and symptoms and the major causes of ovarian cancer. There are also descriptions of various diagnosis techniques that helped to discover the cancer cells and treatment of the patient.
Published by: Uroosa Shafi, Sugandha Sharma
Author: Uroosa Shafi
Paper ID: V4I6-1254
Paper Status: published
Published: November 23, 2018
Development of safe flying protocol aided by Artificial Intelligence
The paper is about how machine learning can impact the air traffic control systems and future proof it.
Published by: Karan Ganesh, Jessysonia S. P., Ajith Raj R., Rohith I. J.
Author: Karan Ganesh
Paper ID: V4I6-1189
Paper Status: published
Published: November 23, 2018
