Menu
Index

Sample run 2

 
In this sample run, a sample application was performed to analyze hypopnea. Although it is based on real data, the purpose of this example is simply to explain how PSGMiner could be used for testing a hypothesis, and it is not meant to suggest a clinical procedure for experiments of sleep apnea. In this sample study it is aimed to investigate differences which occur in the EEG features extracted from O1–A2 channel during 1 seconds before hypopnea occurrence and during 1 seconds following the onset of the apnea. In other words, this was done to see whether the before apnea period and onset of apnea could be classified using signals from O1 – A2 channel.The statistical verification of the study results for the sample run is shown in the user interface of Machine Learning in last figure. According to the results for the sample run, the accuracy rate is 76.52%. A careful study of the results indicates that there is no important distinction between before hypopnea period and onset of hypopnea. However, this is by no means a demanding study, and it was only carried out to illustrate how our software can be used in clinical events.
PSG data
PSG data that are used at the test run 2 were taken from two real patients, whose identification information is kept as secret. (Patient 1 age: 48, weight: 75 kg, height: 161 cm, neck circumference: 36 cm, sex: female, patient 2 age: 55 weight: 96kg, height: 173 cm, neck circumference: 43 cm, sex male). These patients were actual patients from the archive of a sleep laboratory. Recordings were used in retrospective manner. All patients were diagnosed with obstructive sleep apnea and underwent a one-night PSG for diagnosis.These PSG records are recorded with 44-channel polygraph (Compumedics 44E series, Avustralia) and these archive records are obtained through the use of complete PSG techniques in the Sleep Laboratory of the Faculty of Medicine, Trakya University.
 
Patient 1 Hypopnea count: 83
Patient 2 Hypopnea count: 48
 
Used channels and features for classification.
No         
Channel
Feature
1-50
O1-A2
All available EEG features in PSGMiner
51
 
SleepStage
52
Airflow2
class {0,1}
class 0:Not Hypopnea
class 1:Hypopnea
 
Classification Results (PSGMiner)
 
PSGMiner
 
Neural Network (Rprop)
 
Neuron Type : Sigmoid
Hidenlayer Count : 1
Hidenlayer 0 Neuron Count : 32
Learning Rate : 0.3
Maximum Iteration : 500
Epsilon : 1E-8
Weights Initial : 0.1
Weights Max : 50
 
---------------
 
Instances :230
Attributes :58
   
--Summary-- 
   
Accuracy :%76.522
Root mean squared error :0.439
Kappa Value :0.534
Average==>
Sensitivity :0.765
Specificity :0.774
Precision :0.803
False Negative Rate :0.235
F1 score :0.759
ROC Area :0.77
Confusion Matrix 0 (Classified as)  1 (Classified as)
            0             71             47
            1             7              105
 
 
____________________________________________
 
 
Click on [Create New Database] button
 
 
Enter database name
 
 
Enter Epoch Length
 
 
Click on [OK] button
 
 
Enter database values and then click on [Connect] button
 
 
Select features and click on [Event Based] radio button
 
 
Choose parameters and click on [Hypopnea] radio button
 
 
Click on [Create Analysis Table] Button
 
 
Click on [Feature Extraction] Button
 
 
Click on [Start] Button
 
 
After Finish click on [OK] button
 
 
Click on [Close] button
 
 
Click on [Remove Artifact] Button
 
 
Click on [AWAKE] item and click on [Remove Selected Epochs] Button
 
 
Click on [OK] button
 
 
Click on [Machine Learning] Button
 
 
Click on [Hypopnea] item and Click on [Output] Button
 
 
Click on [inputs you want] items and Click on [Input] Button and than Click on [Create Machine Learning Table] Button
 
 
Type [Table name] and click on [OK] button
 
 
Click on [Neural Network] Button
 
 
Setup Parameters and Click on [Start] Button
 
 
Classification Results
 
 
 
 
 
 
 
 
The online help was made with Dr.Explain