Sheikh Shanawaz Mostafa1, Darío Baptista2, Antonio G Ravelo-García3, Gabriel Juliá-Serdá4, Fernando Morgado-Dias5. 1. ITI/Larsys/Madeira Interactive Technologies Institute, Portugal; Universidade de Lisboa, Instituto Superior Técnico, Portugal. Electronic address: sheikh.mostafa@tecnico.ulisboa.pt. 2. ITI/Larsys/Madeira Interactive Technologies Institute, Portugal; Universidade de Lisboa, Instituto Superior Técnico, Portugal. Electronic address: dario.baptista@tecnico.ulisboa.pt. 3. Universidad de Las Palmas de Gran Canaria, Institute for Technological Development and Innovation in Communications, Spain; ITI/Larsys/Madeira Interactive Technologies Institute, Portugal. Electronic address: antonio.ravelo@ulpgc.es. 4. Pulmonary Medicine Department, Hospital Universitario de Gran Canaria Dr. Negrın, Las Palmas de Gran Canaria 35010, Spain. Electronic address: jjulser@gobiernodecanarias.org. 5. ITI/Larsys/Madeira Interactive Technologies Institute, Portugal; Universidade da Madeira, Portugal. Electronic address: morgado@uma.pt.
Abstract
BACKGROUND AND OBJECTIVE: Sleep apnea is a common sleep disorder, usually diagnosed using an expensive, highly specialized, and inconvenient test called polysomnography. A single SpO2 sensor based on an automated classification system can be developed to simplify the apnea detection. The main objective of this work is to develop a classifier based on a convolution neural network with the capability of detecting apnea events from one dimensional SpO2 signal. However, to find an optimum convolution neural network structure is a daunting task is usually done by a trial-and-error method. To solve this problem, a method is proposed to save time and simplify the process of searching for an optimum convolution neural network structure. METHODS: Greedy based optimization is proposed to search for an optimized convolution neural network structure. Three different variants of greedy based optimization are proposed: the topology transfer, the weighted-topology transfer with rough estimation, and the weighted-topology transfer with fine tuning. The subject independent and the cross-database test are performed for the analysis. RESULTS: Considering the balance between the execution time and the performance, the weighted-topology transfer with rough estimation is the best. An accuracy of 88.49% for the HuGCDN2008 database and 95.14% for the Apnea-ECG database are obtained for apnea events detection per minute. Regarding the apnea patient detection, also referred to as global classification, an accuracy of 95.71% is achieved for the HuGCDN2008 database, and 100% is achieved for the AED database without removing any subjects from both databases. CONCLUSIONS: The proposed one-dimensional convolution neural network performs better in a similar situation than those presented in the literature. The greedy based methods, mainly the weighted-topology transfer with rough estimation, is an alternative method to extensive trial and error method.
BACKGROUND AND OBJECTIVE: Sleep apnea is a common sleep disorder, usually diagnosed using an expensive, highly specialized, and inconvenient test called polysomnography. A single SpO2 sensor based on an automated classification system can be developed to simplify the apnea detection. The main objective of this work is to develop a classifier based on a convolution neural network with the capability of detecting apnea events from one dimensional SpO2 signal. However, to find an optimum convolution neural network structure is a daunting task is usually done by a trial-and-error method. To solve this problem, a method is proposed to save time and simplify the process of searching for an optimum convolution neural network structure. METHODS: Greedy based optimization is proposed to search for an optimized convolution neural network structure. Three different variants of greedy based optimization are proposed: the topology transfer, the weighted-topology transfer with rough estimation, and the weighted-topology transfer with fine tuning. The subject independent and the cross-database test are performed for the analysis. RESULTS: Considering the balance between the execution time and the performance, the weighted-topology transfer with rough estimation is the best. An accuracy of 88.49% for the HuGCDN2008 database and 95.14% for the Apnea-ECG database are obtained for apnea events detection per minute. Regarding the apneapatient detection, also referred to as global classification, an accuracy of 95.71% is achieved for the HuGCDN2008 database, and 100% is achieved for the AED database without removing any subjects from both databases. CONCLUSIONS: The proposed one-dimensional convolution neural network performs better in a similar situation than those presented in the literature. The greedy based methods, mainly the weighted-topology transfer with rough estimation, is an alternative method to extensive trial and error method.
Authors: Tanmoy Paul; Omiya Hassan; Khuder Alaboud; Humayera Islam; Md Kamruz Zaman Rana; Syed K Islam; Abu S M Mosa Journal: AMIA Annu Symp Proc Date: 2022-05-23
Authors: Daniel Álvarez; Gonzalo C Gutiérrez-Tobal; Fernando Vaquerizo-Villar; Fernando Moreno; Félix Del Campo; Roberto Hornero Journal: Adv Exp Med Biol Date: 2022 Impact factor: 3.650
Authors: Fernando Vaquerizo-Villar; Daniel Álvarez; Gonzalo C Gutiérrez-Tobal; C A Arroyo-Domingo; F Del Campo; Roberto Hornero Journal: Adv Exp Med Biol Date: 2022 Impact factor: 3.650