My Way to Health ‘trajectories medicine’

Description

In short
The ambition of MyWaytoHealth is to implement artificial intelligence for developing 'trajectories medicine', which will be pioneered by addressing Obstructive Sleep Apnea (OSA) affecting one billion people worldwide.

Scientific objectives and context
The innovation of MyWaytoHealth 'trajectories medicine' chair is encapsulated in the concept of disease trajectome. Trajectome encompasses the conditions, risk factors, environmental and societal determinants that modulate disease trajectory in individual patient’s ecosystem requiring a new organization for value-based care.

Research directions
Artificial intelligence in MyWaytoHealth is fundamental for providing unique insights for organizing the huge diversity of factors contributing to individual patient’s trajectories into a rational framework amenable for clinical decision, therapeutic interventions and reform of the health system.

Activities

The chair will develop artificial intelligence and integrative data analytics merging routinely collected information with innovative data acquisitions:

  • Structured and unstructured data captured by individual Digital Health Passports (industrial partnership with home care providers, EIT health POLLAR, MARS electronic medical records);
     
  • International/national disease registries (120,000 patients in French sleep apnea registry and European Sleep apnea database (Principal investigators); 50,000 patients followed at home by Agiradom, over 25,000 patients with sleep apnea assessed; cardiovascular/sleep cohorts (G Derumeaux, Paris);
     
  • Longitudinal data generated by innovative sensors (Dreem headband (longitudinal, large-scale EEG at-home), CPAP treatment devices (international industrial and academic partnerships: data on millions OSA patients available from RESMED; Mandibular movements: SUNRISE and University of Namur Belgium);
     
  • National administrative healthcare databases covering medications, ambulatory and hospital care (Private insurance MGEN); ALASKA and COVISAS projects;
     
  • Socio-economic, attitude and behavioral determinants (CDP LIFE program (1.7 M€)).

Chair events

ISRAEL-FRANCE webinar in artificial intelligence applications in the COVID 19 pandemic, 23/11/2020
  • Sleep at time of COVID-19: Big data and artificial intelligence for improving knowledge and continuing sleep care

Workshop Swansea – Grenoble, virtual meeting, 22/10/2020
  • Al & medicine

Keio University, Tokyo, 06/02/2020
  • Keynote lecture: 'big data analyses and AI for supporting integrated care in sleep apnea'

The 5th PKU-UPENN Sleep Medicine Forum & Xinyue Health International Sleep Forum, Pékin, 16 & 17/11/2019
  • 'OSA management: knowledge gained from big data analyses an AI' (Keynote speech)

SENSAPNEA EIT health (success story)

Scientific publications


Tamisier R, Treptow E, Joyeux-Faure M, Levy P, Sapene M, Benmerad M, Bailly S, Grillet Y, Stach B, Muir JF, Pegliasco H, Pépin JL ; OPTISAS trial investigators. Impact of a multimodal telemonitoring intervention on CPAP adherence in symptomatic low-cardiovascular risk sleep apnea : a randomized controlled trial. Chest 2020, in press. Obstructive sleep apnea, chronic obstructive pulmonary disease and NAFLD : an individual participant data meta-analysis. Sleep Med 2020, in press. PMID :The Somatotropic Axis in the Sleep

Apnea-Obesity Comorbid Duo. Front Endocrinol (Lausanne) 2020 ;11 :376. Long-term variations of arterial stiffness in patients with obesity and obstructive sleep apnea treated with continuous positive airway pressure. PLoS One 2020 ;15 : e0236667.

Mendelson M, Gentina T, Gentina E, Tamisier R, Pépin JL, Bailly S. Multidimensional Evaluation of Continuous Positive Airway Pressure (CPAP) Treatment for Sleep Apnea in Different Clusters of Couples. J Clin Med. 2020 ;9 : E1658.

Pépin JL, Bruno RM, Yang RY, Vercamer V, Jouhaud P, Escourrou P, Boutouyrie P. Wearable Activity Trackers for Monitoring Adherence to Home Confinement During the COVID-19 Pandemic Worldwide: Data Aggregation and Analysis. J Med Internet Res. 2020 ;22: e19787.

Viglino D, Martin M, Piché ME, Brouillard C, Després JP, Alméras N, Tan WC, Coats V, Bourbeau J, Pépin JL, Maltais F ; CanCOLD Collaborative Research Group and the Canadian Respiratory Research Network. Metabolic profiles among COPD and controls in the CanCOLD population-based cohort. PLoS One 2020 ;15: e0231072.

Revol B, Jullian-Desayes I, Bailly S, Tamisier R, Grillet Y, Sapène M, Joyeux-Faure M, Pépin JL; OSFP national French registry scientific council. Who May Benefit from Diuretics in OSA? A Propensity Score-Match Observational Study. Chest 2020 ;158 :359-364.

Viglino D, Daoust R, Bailly S, Faivre-Pierret C, Charif I, Roustit M, Paquet J, Debaty G, Pépin JL, Maignan M, Chauny JM. Opioid drug use in emergency and adverse outcomes among patients with chronic obstructive pulmonary disease: a multicenter observational study. Sci Rep. 2020 ; 10(1) :5038.

Revol B, Jullian-Desayes I, Guichard K, Micoulaud-Franchi JA, Tamisier R, Philip P, Joyeux-Faure M, Pépin JL. Valproic acid and sleep apnoea: A disproportionality signal from the WHO pharmacovigilance database. Respirology 2020; 25 :336-338.

Pépin JL, Letesson C, Le-Dong NN, Dedave A, Denison S, Cuthbert V, Martinot JB, Gozal D. Assessment of Mandibular Movement Monitoring With Machine Learning Analysis for the Diagnosis of Obstructive Sleep Apnea. JAMA Netw Open 2020 ; 3 : e1919657.

Bailly S, Daabek N, Jullian-Desayes I, Joyeux-Faure M, Sapène M, Grillet Y, Borel JC, Tamisier R, Pépin JL. Partial failure of CPAP treatment for sleep apnoea: Analysis of the French national sleep database. Respirology 2020; 25:104-111.

   

Research topics

Topic 1: Large-scale processing and Interactive multidisciplinary exploration of patients’ trajectories
  • Informative representation of trajectories (visualization, dashboards) allowing identifying clinically useful trajectories in large, high-dimensional data sets (in partnership with Probayes an AI SME subsidiary of “La Poste” and SEMEIA an AI SME working on national claims and predictive medicine).
  • Identify subtle and complex associations between diseases and environmental factors that are unavailable with traditional analytic approaches: Compare trajectories to understand differences and contextualize decisions
  • Hypothesis generation for later validation in clinical trials and real-life health care

Topic 2: Prediction of health progression and aggregation of comorbidities
  • Deep learning techniques to predict health progression and treatments impacts at a collective level (cohorts and real-life data) and at an individual level

Topic 3: Patients centered and integrated care and reform the health system and cost effectiveness evaluations
  • AI for improving sleep apnea diagnosis pathways: SENSAPNEA EIT health (success story), Reimbursement of a new diagnosis pathway in FRANCE (4M for a nation-wide validation study)
  • Validation of new payment models using deep learning of administrative databases
  • Transfer learning to generate multilingual textual narratives for different expertise levels (medical professionals, patients, etc.)