A Method to Detect Obstructive Sleep Apnea Using Neural Network Classification of Time-Frequency Plots of the Heart Rate Variability

• Published in: 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society Vol. 2007; pp. 6101 - 6104
• Main Authors: Al-Abed, M., Manry, M., Burk, J.R., Lucas, E.A., Behbehani, K.
• Format: Conference Proceeding Journal Article
• Language: English
• Published: United States IEEE 01.01.2007
• Subjects: Algorithms; Artificial Neural Networks; Co-occurrence Matrix; Data mining; Diagnosis, Computer-Assisted - methods; Discrete Fourier transforms; ECG; Electrocardiography; Electrocardiography - methods; Expert Systems; Feature extraction; Female; Heart Rate; Heart rate detection; Heart rate variability; Humans; Image generation; Male; Middle Aged; Multi-Layer Perceptron; Neural networks; Neural Networks (Computer); Pattern Recognition, Automated - methods; Reproducibility of Results; Sensitivity and Specificity; Sleep apnea; Sleep Apnea, Obstructive - diagnosis; Sleep Apnea, Obstructive - physiopathology; Sleep Disordered Breathing; Textural Features; Time frequency analysis; Time-Frequency Plots
• ISBN: 9781424407873; 1424407877
• ISSN: 1094-687X; 1557-170X; 1558-4615
• DOI: 10.1109/IEMBS.2007.4353741

This paper presents a new method of analyzing time-frequency plots of heart rate variability to detect sleep disordered breathing from nocturnal ECG. Data is collected from 12 normal subjects (7 males, 5 females; age 46 plusmn 9.38 years, AHI 3.75 plusmn 3.11) and 14 apneic subjects (8 males, 6 females; age 50.28 plusmn 9.60 years; AHI 31.21 plusmn 23.89). The proposed algorithm uses textural features extracted from normalized gray-level co-occurrence matrices (NGLCM) of images generated by short-time discrete Fourier transform (STDFT) of the HRV. Using feature selection, seventeen features extracted from 10 different NGLCMs representing four characteristically different gray-level images are used as inputs to a three-layer Multi-Layer Perceptron (MLP) classifier. After a 1000 randomized Monte-Carlo simulations, the mean training classification sensitivity, specificity and accuracy are 99.00%, 93.42%, and 96.42%, respectively. The mean testing classification sensitivity, specificity and accuracy are 94.42%, 85.40%, and 90.16%, respectively.