PARKINSON ML using machine learning and SVM

NA

1. Neurological Disorders Detection

2. Parkinson Disease What is Parkinson Parkinson’s disease is a complex neurodegenerative disease that affects a large portion of the worldwide population. With current prevalence rates, ranging from 10 to 800 people per 100,000, PD is one of the most common neurodegenerative disorders. What is the Problem? There is currently no objective method for diagnosing PD. It can take months to get a reliable PD diagnosis, and symptoms need to be carefully monitored. Even then the probability of an inaccurate diagnosis is approximately 25%. A large, global burden of disease study identified PD as one of the top 5 leading causes of death from neurological disorders in the US. It is estimated that there were approximately 6.1 million people with PD ( PwP ).

3. Symptoms Parkinson's disease signs and symptoms can be different for everyone. Early signs may be mild and go unnoticed. Parkinson's signs and symptoms may include: Tremor.  A tremor, or rhythmic shaking, usually begins in a limb, often your hand or fingers. Slowed movement (bradykinesia).  Over time, Parkinson's disease may slow your movement, making simple tasks difficult and time-consuming. Rigid muscles.  Muscle stiffness may occur in any part of your body. The stiff muscles can be painful and limit your range of motion. Impaired posture and balance.  You may fall or have balance problems as a result of Parkinson's disease. Loss of automatic movements.  You may have a decreased ability to perform blinking, smiling or swinging your arms when you walk. Speech changes.  You may speak softly, quickly, slur or hesitate before talking. Writing changes.  It may become hard to write, and your writing may appear small.

4. Parkinson’s Disease Brain vs. Normal Brain

5. Tools We Used G oogle colab Python (Version 3.7) D ataset ( The dataset was created by Max Little of the University of Oxford, in collaboration with the National Centre for Voice and Speech, Denver, Colorado, who recorded the speech signals.) SkLearn Pandas Numpy

6. Dataset Description Data Set Information: This dataset is composed of a range of biomedical voice measurements from 31 people, 23 with Parkinson's disease (PD). Each column in the table is a particular voice measure, and each row corresponds one of 195 voice recording from these individuals ("name" column). The main aim of the data is to discriminate healthy people from those with PD, according to "status" column which is set to 0 for healthy and 1 for PD. The data is in ASCII CSV format. The rows of the CSV file contain an instance corresponding to one voice recording. There are around six recordings per patient, the name of the patient is identified in the first column.

7. Attribute Information: Matrix column entries (attributes): name  - ASCII subject name and recording number MDVP:Fo (Hz)  - Average vocal fundamental frequency MDVP:Fhi (Hz)  - Maximum vocal fundamental frequency MDVP:Flo (Hz)  - Minimum vocal fundamental frequency MDVP:Jitter (%)  ,  MDVP:Jitter (Abs)  ,  MDVP:RAP  ,  MDVP:PPQ  ,  Jitter:DDP  - Several measures of variation in fundamental frequency MDVP:Shimmer  ,  MDVP:Shimmer (dB)  ,  Shimmer:APQ3  ,  Shimmer:APQ5  ,  MDVP:APQ  ,  Shimmer:DDA  - Several measures of variation in amplitude NHR  ,  HNR  - Two measures of ratio of noise to tonal components in the voice status  - Health status of the subject (one) - Parkinson's, (zero) - healthy RPDE  ,  D2  - Two nonlinear dynamical complexity measures DFA  - Signal fractal scaling exponent spread1  ,  spread2  ,  PPE  - Three nonlinear measures of fundamental frequency variation

8. SUPPORT VECTOR MACHINE Support Vector Machine or SVM is one of the most popular Supervised Learning algorithms, which is used for Classification as well as Regression problems. However, primarily, it is used for Classification problems in Machine Learning. The goal of the SVM algorithm is to create the best line or decision boundary that can segregate n-dimensional space into classes so that we can easily put the new data point in the correct category in the future. This best decision boundary is called a hyperplane..

9. DATA PRE-PROCESSING

10. SPLITTING THE DATA INTO TEST DATA & TRAINING DATA

11. MODEL TRAINING – SUPPORT VECTOR MACHINE

12. Classifier Used and result MODEL EVALUATION ACCURACY SCORE

13. PREDICTIVE SYSTEM – DISEASE DETECTION

14. Future Vision Diagnosis can be possible in the remote area and It can make this process cost efficient. Decision support tools are gaining significant research interest due to their potential to improve health-care provision. Among many possible approaches, those that provide noninvasive monitoring and diagnosis of diseases are of increased interest to clinicians and biomedical engineers. We aim to provide this diagnosis to people in remote areas where healthcare is not just lacking but extremely inadequate.

15. Conclusion Support Vector Machine SVM Classifier gave the best result . ACCURACY: approx. 88% Accuracy score of Training Data – 88.46 Accuracy Score of Test Data – 87.

16. References and Bibliography https://www.mayoclinic.org/diseases-conditions/parkinsons-disease/symptoms-causes/syc-20376055 Kaggle dataset (university of oxford ) http://parkinson.org/understanding-parkinsons https://www.ninds.nih.gov/Disorders/All-Disorders/Parkinsons-Disease-Information-Page

17. Rest of all the results, source code and the visualization can be found at our GitHub repo.