Where PhDs and companies meet
Menu
Login

Already registered?

New user?

Model-based reasoning methods for the analysis of cardio-respiratory interactions

ABG-105218 Thesis topic
2022-05-02 Public funding alone (i.e. government, region, European, international organization research grant)
LTSI-Université de Rennes 1
Rennes - Bretagne - France
Model-based reasoning methods for the analysis of cardio-respiratory interactions
  • Engineering sciences
  • Digital
  • Health, human and veterinary medicine

Topic description

Sleep apnea syndrome (SAS) is a multifactorial condition characterized by repeated episodes of apnea and hypopnea during the patient’s sleep. Between 6% to 17% of the adult population suffers from SAS, but the syndrome is highly under-diagnosed. Therefore, there is a need for new methods and tools to better understand the pathophysiology of this condition and new reliable automatic diagnostic strategies for improving the detection and management of patients suffering from SAS. The comprehension and interpretation of the acute response to apnea require the analysis of heterogenous data (physiological signals, categorical and quantitative clinical data, ), acquired on SAS patients. Although supervised and unsupervised methods are usually used in this context, they do not integrate any knowledge about the complex behavior of physiological systems and their interactions. In this context, model-based reasoning method could be proposed in order to improve interpretability of results.


The objective of this thesis is to improve the understanding of the acute cardio-respiratory responses to apnea and hypopnea events by proposing model-based reasoning methods for the analysis of cardiorespiratory interactions. One of the main challenges will be to provide patient-specific and eventspecific interpretations of apnea episodes, taking into account the morphology and the dynamics of the observed clinical data. Explainable artificial intelligence tools should be proposed and developed in order to improve the interpretability of data analysis.

The global approach lies on methodological tools combining computational models, model-based
reasoning, machine-learning and signal processing. The project is built around 3 tasks:

  1. Development of modeling methods for the integration of hybrid models and parametric analysis (sensitivity analysis and identification methods),
  2. Proposition of a novel integrated model of cardio-respiratory interactions, based on previous works of our team [1,2]. A first parametric analysis will be proposed in the context of SAS in order to proposed a first rank of importance between model parameters concerning the physiological response to apnea.
  3. Evaluation on a clinical database (HYPNOS study - ANR PASITHEA project) [3,4] composed of data acquired during polysomnography, which consists of a complete multi-channel recording and monitoring of cardio-respiratory, neurological and sleep characterization signals during a whole night. After a pre-processing of physiological signals, patient-specific parameters will be identified and a phenotyping of the patients will be proposed.

[1] Guerrero G., V. Le Rolle, Loiodice C., Amblard A, Pepin J-L, Hernandez AI. Modeling patient-specific desaturation patterns in sleep apnea. IEEE Trans Biomed Eng. 2021 Oct 19 ;PP. doi :10.1109/TBME.2021.3121170.

 

[2] Guerrero G., Le Rolle V., Hernandez A.I.. Parametric analysis of an integrated model of cardio-respiratory interactions in adults in the context of obstructive sleep apnea. Annals of Biomedical Engineering. doi : 10.1007/s10439-021-02828-6.

 

[3] Hernandez, A. I., Guerrero, G., Feuerstein, D., Graindorge, L., Perez, D., Amblard, A., Mabo, P., Pépin, J. L., Senhadji, L.. PASITHEA: An Integrated Monitoring and Therapeutic System for Sleep Apnea Syndromes Based on Adaptive Kinesthetic Stimulation. Irbm, 2016. 37(2), 81–89. https://doi.org/10.1109/TBME.2015.2498878

 

[4] Hernandez, A. I., Pérez, D., Feuerstein, D., Loiodice, C., Graindorge, L., Guerrero, G., Limousin, N., Gagnadoux, F., Dauvilliers, Y., Tamisier, R., Prigent, A., Mabo, P., Amblard, A., Senhadji, L., & Pépin, J. L.. Kinesthetic stimulation for obstructive sleep apnea syndrome: An “on-off” proof of concept trial. Scientific Reports, 2018. 8(1), 1–7. https://doi.org/10.1038/s41598-018-21430-w

 

Starting date

2022-10-01

Funding category

Public funding alone (i.e. government, region, European, international organization research grant)

Funding further details

Presentation of host institution and host laboratory

LTSI-Université de Rennes 1

The LTSI (Laboratoire Traitement du Signal et de l'Image) - INSERM 1099 lies at the interface of Health, Information Technologies, Communication Sciences and Integrative Physiology, where modeling plays an essential role.

PhD title

Mathèmatique et Sciences et Technologies de l’Information et de la Communication

Country where you obtained your PhD

France

Institution awarding doctoral degree

Université de Rennes 1

Graduate school

MathStic

Candidate's profile

We are looking for a highly motivated candidate with expertise in signal processing and computational modeling. Programming skills will also be appreciated (e.g., C++, Python). Knowledge about physiological systems would be an advantage but is not necessary.

2022-06-30
Partager via
Apply
Close

Vous avez déjà un compte ?

Nouvel utilisateur ?