Vestibüler sistem kaynaklı rahatsızlıkların kuvvet algılayıcıları tabanlı analizi
Force sensor based analysis of vestibular system disorders
- Tez No: 601281
- Danışmanlar: DOÇ. DR. SERHAT İKİZOĞLU
- Tez Türü: Yüksek Lisans
- Konular: Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2019
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Kontrol ve Otomasyon Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Kontrol ve Otomasyon Mühendisliği Bilim Dalı
- Sayfa Sayısı: 59
Özet
Bu çalışma, denge sorunu yaşayan kişilerin rahatsızlıklarının belirlenmesine yönelik geniş kapsamlı bir projenin önemli bir parçasıdır. Tüm proje temel olarak iki aşamadan oluşmaktadır. İlk aşamada kişinin“hasta”ya da“sağlıklı”gruplardan hangisine ait olduğu belirlenmekte; kişi hasta grubuna dahil olduğu takdirde, ikinci aşamada hastalığı tespit edilmeye çalışılmaktadır. Çalışmamızda, öncelikle birinci aşamada kullanılacak olan verilerin elde edilmesi faaliyetleri yürütülmüştür. Bu amaçla, üzerine kuvvet algılayıcılarının yerleştirildiği tabanlar üretilmiştir. Tabanlar, deneklere ait ayakkabılar içerisine konularak yürüyüş boyunca gerçek zamanlı veri toplanmış, sağlanan verilerden hareketle yürüyüşü karakterize edecek çeşitli öznitelikler elde edilmiştir. Bu öznitelikler değerlendirilerek hasta-sağlıklı şeklinde bir ikili sınıflama yapılmıştır. Bu işlemdeki başarı %96 olarak belirlenmiştir. Her ne kadar gelinen nokta önemli bir başarımı işaret ediyor ise de, çalışma, güvenilirliğin artırılması amacıyla denek sayısı hacminin yükseltilmesi, ayrıca yeni öznitelikler elde edilmesi doğrultusunda sürdürülmektedir. Sağlık dışında spor vb. farklı alanlarda da uygulanabilecek olması, çalışmanın değerini önemli ölçüde arttırıcı niteliktedir.
Özet (Çeviri)
The vestibular system is a system that allows us to keep our balance. This system senses the position and motion of our head in three-dimensional space through its sensors in our inner ear, automatically generates the necessary signals and thus enables us to maintain our balance. Vestibular system-related disorders can significantly affect the daily life of an individual and reduce the quality of life. This study is an important part of the comprehensive project that aims to identify the discomfort of people who experience balance problems. The project consists of two phases. The first phase is to determine that an individual belongs to which of the“patient”or“healthy”groups. The second phase is to determine the type of disease of the individual identified as being included in the patient group. The activities of gathering the data to be used in the first stage were firstly carried out in this study. For this purpose, it was aimed to walk the subject along the specified path and to measure the weight changes under the soles of the feet during this walk. Accordingly, a system was designed which could be placed into the shoes of the subjects and would allow real-time data collection during their walks. The characteristics and requirements of such systems were investigated by making a literature review before designing the system. Products were made by taking into consideration such data. The components of this system consist of the insoles with force sensors placed into the subject's own shoes and of an electronic unit that transfers data from these insoles wirelessly to the computer. The durable and soft plastazote material used in the production of orthopedic products were used in the production of insoles. In addition, insoles were manufactured for five different sizes of the shoe so that different subjects could use them (36-38-40-42-44). Literature research has been made for the selection of the sensor to be used, sensor types have been investigated and, consequently, it has been decided to use the PL-2728 coded resistive sensor. The sensor has high repeatability of measurement and light, thin, flexible and durable structure, ease of application, cost-effective and low power consumption. All these factors have become effective in its selection. The literature review was also made for the sensor placement to be made inside the insoles, the sensor placement in the previous studies was examined and the locations were determined for accurate measurement of body weight distribution. Arduino Mega set and HC-06 Bluetooth module connected to this set were used to transfer the data obtained from the soles to the computer in the wireless environment. In addition, the data from the sensors were transferred to a voltage divider circuit, and thus the values corresponding to the weight of the subject were obtained. Data from four sensors in each shoe of the subject were transferred to analog inputs of Arduino Mega set placed on the subject. Next, these data were transferred to a computer via Bluetooth module in real-time. Since the output of the sensor and of the voltage divider circuit was not linear, it was needed to obtain the characteristic curve of the sensor. Necessary apparatus was established, weights were placed on the sensor, output values were recorded and the characteristic curve was drawn. Then, the equation giving the result closest to the characteristic curve was obtained by the curve fitting method. The tension values were calculated through this equation and the weight values were obtained. The data required for walking analysis were obtained through the walking of diseased and healthy subjects on the path in the Audiology Department, Cerrahpaşa Medical School, İstanbul University; all following ethics committee approval. Data analysis studies were performed in time and frequency domains in order to determine the attributes to be used in the analysis. The aim of this study was to examine individually each of the steps taken during the walking analysis and to evaluate any gait abnormalities that might occur. Time-dependent studies were conducted in two phases. In the first phase, at what rate the force sensors are active during a step was evaluated. For this purpose, the time the sensor is active during a step was proportioned to the duration of the step. Here, the step time was defined as the time passing from the heel touching to the ground to discontinuation of contact of the big toe with the ground. In the second phase, the distribution of the weight applied by the subject to the sensors in one foot was examined during a step and the result of the relation used here was called the sensor loading rate. The sensor loading ratio was calculated by dividing the area value under the weight-time curve of the respective sensor in one step by the sum of the area values of all sensors in the same foot. In the frequency domain analysis, harmonic distortion variables of the force sensor data were evaluated as a measure of staggering. In our analysis, the force data of the sensors in each step were examined by Fourier Transform and the harmonic distortion variables were calculated using the data. In the experimental results section, sensor activation rates, sensor loading rates and sensor harmonic distortion variables of a patient and a healthy subject were presented. When the data were examined, it was seen that different sensor data were effective for different attributes. It has been found necessary and sufficient to consider the data of certain sensors in order to obtain the attributes to be used in the classification process. In light of this information, 18 attributes based on the correlations between the selected sensors were decided to use in the data analysis. By evaluating these attributes, a dual classification was made as patient-healthy. The machine learning algorithms were assessed by transferring the attributes obtained from the subjects to Matlab. As a result of 12 machine learning algorithms, the highest performance was achieved with Support Vector Machines (SVM) by 96% accuracy. Support Vector Machines are an effective and simple method used to classify data. For classification, two groups are formed in the plane using the data obtained from the patient and healthy subjects, then parallel lines are drawn closest to the two groups in the same plane and a common boundary line is created by being approached these lines. In this respect, it should be noted that each attribute used to separate the two groups will define a new plane. The evaluation at this point shows that increasing the number of attributes will increase the performance rate. Although the present situation is an important achievement, the study should be carried out in a way that will increase the number of subjects and obtain new features in order to increase reliability. In addition, the creation of disease libraries using the data on subjects diagnosed with the disease for the disease diagnosis phase is one of the objectives in comprehensive research including this study. The fact that the data obtained through the studies carried out within the scope of this project can be applied to different fields such as sports and games, as well as health, would increase the value of the study significantly.
Benzer Tezler
- Septorinoplasti olgularında vestibüler sistem fonksiyonlarının değerlendirilmesi
Evaluation of vestibular system in septorhinoplasty cases
F. GÜLİZ ALP
Yüksek Lisans
Türkçe
2012
Kulak Burun ve BoğazBaşkent ÜniversitesiKulak Burun Boğaz Ana Bilim Dalı
PROF. DR. LEVENT N. ÖZLÜOĞLU
- Normal işiten bireylerde 500HZ'te level spesifik chirp ve tone- burst uyaranların servikal vestibüler uyarılmış miyojenik potansiyel (CVEMP) yanıtlarına etkisinin değerlendirilmesi
Evaluation of the effect of level specific chirp and tone-burst stimuli at 500HZ on cervical vestibular evoked miyogenic patential (CVEMP)responses in individuals with normal hearing
FEHİME ÇARKIT
Yüksek Lisans
Türkçe
2021
Kulak Burun ve BoğazKapadokya ÜniversitesiOdyoloji Ana Bilim Dalı
DR. ÖĞR. ÜYESİ MEHMET CELALETTİN CİHAN
- COVID-19'un işitme ve vestibüler sistem üzerine etkilerinin incelenmesi
Investigation of the effects of COVID-19 on the hearing and vestibular system
BETÜL ÖZDEMİR
Yüksek Lisans
Türkçe
2021
Kulak Burun ve Boğazİstanbul Gelişim ÜniversitesiOdyoloji Ana Bilim Dalı
DR. ÖĞR. ÜYESİ SELVA ZEREN
- Normal işitmeye sahip tinnituslu bireylerde vestibüler uyarılmış miyojenik potansiyellerin (VEMP) değerlendirilmesi
Evaluation of vestibular evoked myogenic potentials (VEMP) on individuals who have tinnitus with normal hearing
AYHAN ADMIŞ
Yüksek Lisans
Türkçe
2015
Sağlık EğitimiTurgut Özal ÜniversitesiOdyoloji ve Konuşma Bozuklukları Ana Bilim Dalı
PROF. DR. MEHMET GÜNDÜZ
- Diyabetik nöropatili hastalarda vestibüler sistemin değerlendirilmesi
Vestibular system evaluation in patients with diabetic neuropathy
MERVE KANDAZOĞLU
Yüksek Lisans
Türkçe
2019
Kulak Burun ve BoğazOndokuz Mayıs ÜniversitesiKulak Burun Boğaz Ana Bilim Dalı
PROF. DR. FİGEN BAŞAR