Automatic Detection of Asynchrony Levels of Mechanically Ventilated Patients

• Published in: 2023 19th IEEE International Colloquium on Signal Processing and Its Applications, CSPA 2023 - Conference Proceedings
• Main Author: Damanhuri N.S.; Bakar I.K.N.A.; Sauki N.S.M.; Othman N.A.; Chiew Y.S.; Meng B.C.C.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2023
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85153706688&doi=10.1109%2fCSPA57446.2023.10087410&partnerID=40&md5=9b2b137828aeb7da45513128ead598e9

A patient with breathing failure or Acute Respiratory Distress Syndrome (ARDS) needs a mechanical ventilator (MV) as a breathing support. However, some patients may produce spontaneous breathing (SB) which results in producing asynchrony events (AEs) even though they are fully sedated. The AEs is referring to the mismatch between the patient's mechanical ventilator and the patient's respiratory system and may worsen the patient's condition. Hence, this will lead to the inaccurate measurement of lung conditions in MV patients. Furthermore, the negative elastance is another metric that can occur during mechanical ventilation when the patient's inspiratory demand is less than zero. Thus, this study aims to calculate the asynchrony events (AEs) with negative elastance in mechanically ventilated patients and develop the graphical user interface (GUI) for automated detection of asynchrony levels that potentially helps clinicians in monitoring the AEs in MV patients. Data on the patients is obtained from the International Islamic University Malaysia (IIUM) Hospital with a total number of 9 patients. The asynchrony events (AEs) with negative elastance are measured based on ± 50% of the median (AUC) over each breathing cycle of a given patient and the negative elastance is obtained based on AUC Edrs that is less than 0. Then, the asynchrony index (AI) is calculated to indicate the asynchrony level of each patient. The result shows the AI that includes the negative elastance produced higher values with 13.02 [1.40 - 17.74] than the AI without negative elastance with 10.32 [1.22 - 17.28] for all 9 MV patients. Thus, the estimation of the asynchrony level that includes the negative elastance is more accurate and can help the clinician for better MV management. This automated system of the GUI enables the detection of the AEs with negative elastance by displaying all the information of AI, MV mode, AEs, and negative elastance automatically which potentially help the clinician in optimizing the management setting of MV. © 2023 IEEE.