The Parkinson disease was firstly recognized by a London‟s doctor name James Parkinson in 1817. Later, in 1960‟s it was found that Parkinson is associated with level of dopamine neurons. Dopamine is neurotransmitter plays a very important role in human brain and body. Dopamine releases a chemical that carries signal between nerve cells and as a result do the planning and programming of movements. With the reduced level of dopamine the movements get interrupted. About 70% of people with Parkinson shows tremor that is most prominent in hands and fingers. Stiffness in the muscles, slowness of movements and lack of coordination while doing daily routine activities are also important signs of parkinson. An exact reason for the death of dopamine is not known. Genetic factor is one of the cause of this disease.15% of the patients have their family history. Internal and external toxins reduce the dopamine production. Free radicals are also responsible for the death of dopamine. As age increases the chances of occurring this disease also increases.
Diagnosis of Parkinson disease is very difficult
I. INTRODUCTION
The Parkinson disease was firstly recognized by a London‟s doctor name James Parkinson in 1817. Later, in 1960‟s it was found that Parkinson is associated with level of dopamine neurons. Dopamine is neurotransmitter plays a very important role in human brain and body. Dopamine releases a chemical that carries signal between nerve cells and as a result do the planning and programming of movements. With the reduced level of dopamine the movements get interrupted. About 70% of people with Parkinson shows tremor that is most prominent in hands and fingers. Stiffness in the muscles, slowness of movements and lack of coordination while doing daily routine activities are also important signs of parkinson. An exact reason for the death of dopamine is not known. Genetic factor is one of the cause of this disease.15% of the patients have their family history. Internal and external toxins reduce the dopamine production. Free radicals are also responsible for the death of dopamine. As age increases the chances of occurring this disease also increases.
Diagnosis of Parkinson disease is very difficult
physical diagnosis can also be done but patients are required to be observed for a long time and this diagnosis give results when almost 80% of dopamine gets ended.
Voice of the person shows changes at an earlier stage, so diagnosis of Parkinson using voice analysis can be done at an earlier stage. Reduced in voice level by approx 10 db, whispering, breathiness, tremors, shifting to higher tones are some voice characteristics visible in PD voice. This method is very reliable and of very ultra low cost. Method is completely computerized and no medical professionals are required. As the PD patients have difficulty in clinical visits, in this voice analysis method no clinical visits are required. This method can be done telephonically, so the telediagnosis of the disease can be done by voice analysis with very less costs and efforts.
Voice analysis for diagnosis of disease is not only limited to Parkinson but it can be used for many other diseases. Voice nodules, Reinke‟ edema, asthma can also be diagnosed using this method. Various classifiers are used in such type of diagnosis. With the help of classifiers accuracy and reliability of diagnosis increases.
Multi-Layer Perceptron neural network and Support Vector Machine with linear and puk kernel function were used for classifying PD effected persons and healthy persons. A hybrid technique taking the advantage of both methods is proposed and achieved an accuracy of 90% [1]. In Artificial neural network 70% data as training data and 30% as testing data was used and 93.2% accuracy was achieved. The dataset consist of 23 features and MLP, RBF neural networks are used and their performance comparision is done [2]. By using maximum-relevance-minimum-redundancy criteria features are selected on the basis of mutual information measures between the features and support vector machines are used for building a predictive model [3]. Features are selected on the basis of information gain and twenty two attributes are reduced to sixteen and 83.3% accuracy is achieved with back propagation Multi layer perceptron network
The Parkinson disease was firstly recognized by a London‟s doctor name James Parkinson in 1817. Later, in 1960‟s it was found that Parkinson is associated with level of dopamine neurons. Dopamine is neurotransmitter plays a very important role in human brain and body. Dopamine releases a chemical that carries signal between nerve cells and as a result do the planning and programming of movements. With the reduced level of dopamine the movements get interrupted. About 70% of people with Parkinson shows tremor that is most prominent in hands and fingers. Stiffness in the muscles, slowness of movements and lack of coordination while doing daily routine activities are also important signs of parkinson. An exact reason for the death of dopamine is not known. Genetic factor is one of the cause of this disease.15% of the patients have their family history. Internal and external toxins reduce the dopamine production. Free radicals are also responsible for the death of dopamine. As age increases the chances of occurring this disease also increases. Diagnosis of Parkinson disease is very difficult I. INTRODUCTION The Parkinson disease was firstly recognized by a London‟s doctor name James Parkinson in 1817. Later, in 1960‟s it was found that Parkinson is associated with level of dopamine neurons. Dopamine is neurotransmitter plays a very important role in human brain and body. Dopamine releases a chemical that carries signal between nerve cells and as a result do the planning and programming of movements. With the reduced level of dopamine the movements get interrupted. About 70% of people with Parkinson shows tremor that is most prominent in hands and fingers. Stiffness in the muscles, slowness of movements and lack of coordination while doing daily routine activities are also important signs of parkinson. An exact reason for the death of dopamine is not known. Genetic factor is one of the cause of this disease.15% of the patients have their family history. Internal and external toxins reduce the dopamine production. Free radicals are also responsible for the death of dopamine. As age increases the chances of occurring this disease also increases. Diagnosis of Parkinson disease is very difficult physical diagnosis can also be done but patients are required to be observed for a long time and this diagnosis give results when almost 80% of dopamine gets ended. Voice of the person shows changes at an earlier stage, so diagnosis of Parkinson using voice analysis can be done at an earlier stage. Reduced in voice level by approx 10 db, whispering, breathiness, tremors, shifting to higher tones are some voice characteristics visible in PD voice. This method is very reliable and of very ultra low cost. Method is completely computerized and no medical professionals are required. As the PD patients have difficulty in clinical visits, in this voice analysis method no clinical visits are required. This method can be done telephonically, so the telediagnosis of the disease can be done by voice analysis with very less costs and efforts. Voice analysis for diagnosis of disease is not only limited to Parkinson but it can be used for many other diseases. Voice nodules, Reinke‟ edema, asthma can also be diagnosed using this method. Various classifiers are used in such type of diagnosis. With the help of classifiers accuracy and reliability of diagnosis increases.
Multi-Layer Perceptron neural network and Support Vector Machine with linear and puk kernel function were used for classifying PD effected persons and healthy persons. A hybrid technique taking the advantage of both methods is proposed and achieved an accuracy of 90% [1]. In Artificial neural network 70% data as training data and 30% as testing data was used and 93.2% accuracy was achieved. The dataset consist of 23 features and MLP, RBF neural networks are used and their performance comparision is done [2]. By using maximum-relevance-minimum-redundancy criteria features are selected on the basis of mutual information measures between the features and support vector machines are used for building a predictive model [3]. Features are selected on the basis of information gain and twenty two attributes are reduced to sixteen and 83.3% accuracy is achieved with back propagation Multi layer perceptron network
การแปล กรุณารอสักครู่..