Download PDF
B&G Bewegungstherapie und Gesundheitssport 2022; 38(05): 208-215
DOI: 10.1055/a-1909-5766
Wissenschaft

Entwicklung einer KI-gestützten Bewegungstherapie bei onkologischen Palliativpatienten

Eine PilotstudieDevelopment of AI-supported Exercise Therapy in Oncological Palliative PatientsA pilot study
Nico De Lazzari
1   Westdeutsches Tumorzentrum – Comprehensive Cancer Center, Innere Klinik (Tumorforschung), Universitätsklinikum Essen, 45122 Essen, Deutschland
,
Felix Wichum
2   Fraunhofer IMS, Universität Duisburg-Essen, 47057 Duisburg, Deutschland
,
Miriam Götte
5   Westdeutsches Tumorzentrum – Comprehensive Cancer Center, Klinik für Kinderheilkunde 3, Universitätsklinikum Essen, 45122 Essen, Deutschland
,
Corinna David
4   Fachhochschule Münster, Fachbereich Physikingenieurwesen
,
Karsten Seid
3   Fraunhofer-Institut für Mikroelektronische Schaltungen und Systeme (IMS), 47057 Duisburg und Fachgebiet Elektronische Bauelemente und Schaltungen (EBS), Universität Duisburg-Essen
,
Mitra Tewes
6   Palliativmedizin der Universitätsmedizin Essen, Universitätsklinikum Essen, 45122 Essen, Deutschland
› Author Affiliations
› Further Information
    Permissions and Reprints

Zusammenfassung

HintergrundDie wechselnde Symptomlast ist eine große Hürde in der Sporttherapie von onkologischen Palliativpatienten. Die täglich variierende Symptomstärke erschwert die Einstellung einer optimalen Trainingsbelastung und stellt neben der Motivation eine große Barriere für die Teilnahme an bewegungstherapeutischen Interventionen dar. Ein durch Künstliche Intelligenz (KI) gesteuertes Training könnte helfen, die Trainingseinheiten individuell anzupassen und die Autonomie von Palliativpatienten zu erhalten.

Methoden Fünf Patienten mit fortgeschrittener unheilbarer Krebsdiagnose haben im Rahmen der Routineversorgung eine supervidierte Bewegungstherapie absolviert. Dabei wurde ein Elektrokardiogramm über einen Polar H10 Brustgurt aufgezeichnet und daraus kardiale und respiratorische Vitalparameter extrahiert. Eine Klassifikation in drei Intensitätsstufen über KI erfolgte anhand von neuronalen Netzen.

Ergebnisse Das KI-gesteuerte Training hat eine sehr hohe Klassifikationsgüte (F1-Score: 0,95±0,05) durch die Vereinigung von respiratorischen und kardialen Vitalparametern. Diese Kombination erzielt genauere Klassifikationsergebnisse als die einzelnen Datensätze für kardiale Parameter (0,93±0,06) und respiratorische Parameter (0,72±0,06). Die Berücksichtigung einer Baselinemessung hat eine positive Wirkung auf die Klassifikationsgenauigkeit.

Diskussion Diese Studie stellt die erste Untersuchung zum Einsatz von KI zur Klassifizierung von trainingswissenschaftlichen Inhalten bei onkologischen Palliativpatienten dar. Diese vulnerable Patientengruppe kann von einer objektiven Erfassung des Belastungsniveaus anhand von Parametern des kardiovaskulären Systems profitieren. Mit nur fünf Patienten wird die Aussagekraft dieser explorativen Studie über Kreuzvalidierung hergestellt. Zukünftig sollen weitere Parameter wie ein subjektives Empfinden, Alter, Größe und Geschlecht die Klassifikation weiter verbessern. In einem integrierten System ist eine individuelle Trainingssteuerung in Echtzeit möglich.

Summary

Background The changing symptom burden is a major hurdle in the exercise therapy therapy of oncological palliative patients. The daily varying severity of symptoms makes it difficult to set an optimal training load and, in addition to motivation, represents a major barrier to participation in exercise therapy interventions. Training controlled by artificial intelligence (AI) could help to customize the training sessions and maintain the autonomy of palliative patients.

Methods Five patients with advanced incurable cancer diagnosis underwent supervised exercise therapy as part of routine care. An electrocardiogram was recorded using a Polar H10 chest strap, from which cardiac and respiratory vital parameters were extracted. This led to a classification into three intensity levels via artificial intelligence using neural networks.

Results Through the combination of respiratory and cardiac vital parameters, the AI-controlled training has a very high classification quality (F1-Score: 0.95±0.05). This combination achieves more accurate classification results than the individual data sets for cardiac parameters (0.93±0.06) and respiratory parameters (0.72±0.06). When taken into account, the baseline measurement has a positive effect on classification accuracy.

Discussion This study represents the first investigation into the use of artificial intelligence to classify training intensities for patients in oncological palliative care. This vulnerable patient group can benefit from an objective recording of the stress level based on parameters of the cardiovascular system. It was possible to establish the validity of this exploratory study on cross-validation with only five patients. In the future, other parameters such as subjective perception, age, size and gender should be considered in order to further improve the classification. Thus, in an integrated system, individual training control is possible in real time.

Stichworte

Bewegungstherapie - fortgeschrittene Krebspatienten - Künstliche Intelligenz - Maschinelles Lernen - Trainingssteuerung - Belastungsklassifikation - Herzratenvariabilität - Respirationsrate - Herzfrequenz.

Key words

exercise therapy - advanced cancer patients - artificial intelligence - machine learning - training control - load classification - heart rate variability - respiration rate - heart rate.


Publication History

Received: 01 March 2022

Accepted after revision: 22 June 2022

Article published online:
10 October 2022

© 2022. Thieme. All rights reserved.

Georg Thieme Verlag
Rüdigerstraße 14, 70469 Stuttgart, Germany

 

  • Literatur

  • 1 Henson LA, Maddocks M, Evans C. et al. Palliative care and the management of common distressing symptoms in advanced cancer: pain, breathlessness, nausea and vomiting, and fatigue. Journal of clinical oncology 2020; 38: 905
    CrossrefPubMedGoogle Scholar
  • 2 Seow H, Barbera L, Sutradhar R. et al. Trajectory of performance status and symptom scores for patients with cancer during the last six months of life. Journal of clinical oncology 2011; 29: 1151-1158
    CrossrefPubMedGoogle Scholar
  • 3 Pyszora A, Budzynski J, Wojcik A. et al. Physiotherapy programme reduces fatigue in patients with advanced cancer receiving palliative care: randomized controlled trial. Support Care Cancer 2017; 25: 2899-2908 DOI: 10.1007/s00520-017-3742-4.
    CrossrefPubMedGoogle Scholar
  • 4 Toohey K, Chapman M, Rushby A-M. et al. The effects of physical exercise in the palliative care phase for people with advanced cancer: a systematic review with meta-analysis. Journal of Cancer Survivorship 2022; 1-17
    PubMedGoogle Scholar
  • 5 Heywood R, McCarthy AL, Skinner TL.. Efficacy of exercise interventions in patients with advanced cancer: a systematic review. Archives of physical medicine and rehabilitation 2018; 99: 2595-2620
    CrossrefPubMedGoogle Scholar
  • 6 De Lazzari N, Niels T, Tewes M. et al. A Systematic Review of the Safety, Feasibility and Benefits of Exercise for Patients with Advanced Cancer. Cancers 2021; 13: 4478
    CrossrefPubMedGoogle Scholar
  • 7 Hefferon K, Murphy H, McLeod J. et al. Understanding barriers to exercise implementation 5-year post-breast cancer diagnosis: a large-scale qualitative study. Health education research 2013; 28: 843-856
    CrossrefPubMedGoogle Scholar
  • 8 Frikkel J, Gotte M, Beckmann M. et al. Fatigue, barriers to physical activity and predictors for motivation to exercise in advanced Cancer patients. BMC Palliat Care 2020; 19: 43 DOI: 10.1186/s12904-020-00542-z.
    CrossrefPubMedGoogle Scholar
  • 9 Knowlton SE, O’Donnell EK, Horick N. et al. Moving forward on all fronts: impact, patterns, and barriers to exercise in cancer survivors and patients living with advanced disease. Supportive Care in Cancer 2020; 28: 4979-4988
    CrossrefPubMedGoogle Scholar
  • 10 Patel AV, Friedenreich CM, Moore SC. et al. American College of Sports Medicine roundtable report on physical activity, sedentary behavior, and cancer prevention and control. Medicine and science in sports and exercise 2019; 51: 2391
    CrossrefPubMedGoogle Scholar
  • 11 de Raaf PJ, Sleijfer S, Lamers CH. et al. Inflammation and fatigue dimensions in advanced cancer patients and cancer survivors: an explorative study. Cancer 2012; 118: 6005-6011
    CrossrefPubMedGoogle Scholar
  • 12 Chowdhury AK, Tjondronegoro D, Chandran V. et al. Prediction of relative physical activity intensity using multimodal sensing of physiological data. Sensors 2019; 19: 4509
    CrossrefPubMedGoogle Scholar
  • 13 Aguirre A, Pinto MJ, Cifuentes CA. et al. Machine learning approach for fatigue estimation in sit-to-stand exercise. Sensors 2021; 21: 5006
    CrossrefPubMedGoogle Scholar
  • 14 Qin P, Feng W.. Design of the Exercise Load Data Monitoring System for Exercise Training Based on the Neural Network. Journal of Healthcare Engineering 2021; 2021
    PubMedGoogle Scholar
  • 15 Nevill A, Atkinson G, Hughes M.. Twenty-five years of sport performance research in the Journal of Sports Sciences. Journal of sports sciences 2008; 26: 413-426
    CrossrefPubMedGoogle Scholar
  • 16 Pfeiffer M, Hohmann A.. Applications of neural networks in training science. Human movement science 2012; 31: 344-359
    CrossrefPubMedGoogle Scholar
  • 17 Zhang Y, Zhang Y.. Sports Training System Based on Convolutional Neural Networks and Data Mining. Computational Intelligence and Neuroscience 2021; 2021
    PubMedGoogle Scholar
  • 18 Makivić B, Nikić Djordjević M, Willis MS.. Heart Rate Variability (HRV) as a tool for diagnostic and monitoring performance in sport and physical activities. Journal of Exercise Physiology Online 2013; 16
    PubMedGoogle Scholar
  • 19 Albert JA, Herdick A, Brahms CM. et al. Using Machine Learning to Predict Perceived Exertion During Resistance Training With Wearable Heart Rate and Movement Sensors. In Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) 2021; 801-808
    PubMed
  • 20 Kumar P, Das AK, Halder S.. Statistical heart rate variability analysis under rest and post-exercise. International Journal of Computational Science and Engineering 2022; 25: 186-197
    CrossrefPubMedGoogle Scholar
  • 21 Joyce D, Barrett M.. State of the science: heart rate variability in health and disease. BMJ supportive & palliative care 2019; 9: 274-276
    CrossrefPubMedGoogle Scholar
  • 22 MacDonald AM, Chafranskaia A, Lopez CJ. et al. CaRE@ Home: Pilot study of an online multidimensional cancer rehabilitation and exercise program for cancer survivors. Journal of Clinical Medicine. 2020; 9: 3092
    PubMedGoogle Scholar
  • 23 Cheng KKF, Lim YTE, Koh ZM. et al. Home-based multidimensional survivorship programmes for breast cancer survivors. Cochrane Database of Systematic Reviews. 2017
    PubMedGoogle Scholar
  • 24 Kim JY, Lee MK, Lee DH. et al. Effects of a 12-week home-based exercise program on quality of life, psychological health, and the level of physical activity in colorectal cancer survivors: a randomized controlled trial. Supportive Care in Cancer 2019; 27: 2933-2940
    CrossrefPubMedGoogle Scholar
  • 25 Baumann FT, Hallek M, Meyer J. et al. Evidence and recommendations for oncologic clinical exercise-a personalized treatment concept for cancer patients. Deutsche Medizinische Wochenschrift (1946) 2015; 140: 1457-1461
    PubMedGoogle Scholar
  • 26 Charlton PH, Birrenkott DA, Bonnici T. et al. Breathing rate estimation from the electrocardiogram and photoplethysmogram: A review. IEEE reviews in biomedical engineering 2017; 11: 2-20
    CrossrefPubMedGoogle Scholar
  • 27 Sznajder M, Lukowska M.. Python online and offline ECG QRS detector based on the pan-Tomkins algorithm. Zenodo 2017; 2: 5
    PubMedGoogle Scholar
  • 28 Fujisawa D, Temel JS, Greer JA. et al. Actigraphy as an assessment of performance status in patients with advanced lung cancer. Palliative & Supportive Care 2019; 17: 574-578
    CrossrefPubMedGoogle Scholar
  • 29 Gresham G, Hendifar AE, Spiegel B. et al. Wearable activity monitors to assess performance status and predict clinical outcomes in advanced cancer patients. NPJ digital medicine 2018; 1: 1-8
    CrossrefPubMedGoogle Scholar
  • 30 Teunissen SC, Wesker W, Kruitwagen C.. et al. Symptom prevalence in patients with incurable cancer: a systematic review. Journal of pain and symptom management 2007; 34: 94-104
    CrossrefPubMedGoogle Scholar
  • 31 Cheung WY, Le LW, Zimmermann C.. Symptom clusters in patients with advanced cancers. Supportive care in cancer 2009; 17: 1223-1230
    CrossrefPubMedGoogle Scholar
  • 32 Kravchenko J, Berry M, Arbeev K. et al. Cardiovascular comorbidities and survival of lung cancer patients: Medicare data based analysis. Lung Cancer 2015; 88: 85-93
    CrossrefPubMedGoogle Scholar
  • 33 Herrero Rivera D, Nieto-Guerrero Gómez J, Cacicedo Fernández de Bobadilla J. et al. Cardiovascular disease and survival in non-small cell lung cancer: a multicenter prospective assessment. Clinical and Translational Oncology 2019; 21: 1220-1230
    CrossrefPubMedGoogle Scholar
  • 34 Nicolò A, Massaroni C, Passfield L.. Respiratory frequency during exercise: the neglected physiological measure. Frontiers in physiology 2017; 8: 922
    CrossrefPubMedGoogle Scholar