[home] [Personal Program] [Help]

---

Session: Respiration & Pregnancy
Session starts: Thursday 26 January, 10:30
Presentation starts: 11:00
Room: Room 531

---

Liming Hao ()
Tom Bakkes ()
Anouk van Diepen ()
Ashley de Bie Dekker ()
Pierre Woerlee ()
Massimo Mischi ()
Simona Turco ()

Abstract:
Patient-ventilator asynchrony (PVA) during mechanical ventilation can lead to pulmonary damage, complications and even increase mortality. A machine learning method was proposed in [1] for detection and diagnosis of PVAs. Proper training of these machine learning detection models requires a large amount of labelled data. However, the availability of clinical data with annotations is limited and simulated data is not ventilator specific. In our research, a new framework based on a generative adversarial network was developed to improve simulated data by learning the ventilator specific artefacts from clinical unlabelled data, with the goal of providing a more powerful training dataset for algorithms as in [1]. The input of the network is the simulated data obtained by the patient-ventilator model developed in [2]. The GAN model consists of a generator and a discriminator. The generator reproduces the characteristics of the clinical data while preserving the annotations of the simulated ones. The role of the discriminator is to distinguish clinical data from generated data, thus pushing the generator to produce more realistic data. Qualitative validation was performed visually by the clinical expert. Our qualitative analysis suggests that GAN model produces more realistic ventilatory waveforms, while maintaining the annotations unchanged. For a quantitative validation of our method, we compared the PVA detection and classification performance of the model in [1] by using as a training dataset the simulated data before and after passing it through GAN model. The TPR and PPV of the PVA detection using generated data were 91.4% and 99.3%, while the ones using simulated data were 94.8% and 98.5%, respectively. In terms of the classification performance, the model trained by generated data performs better than simulated data, with an increase in the F1-score for a majority of the asynchrony types. The ability of the GAN model to produce more realistic simulated data was demonstrated qualitatively and quantitatively. This framework can be integrated with the PVAs detection models to further improve the accuracy of PVA detection and classification, possibly decreasing the risk of pulmonary damage during mechanical ventilation.