Artificial Intelligence of Things (AIoT)-Based Strengthening Ergometer Mitigates Cancer Cachexia Progression and Boosts Patient Physical Fitness

File
Li-Chieh Kuo, Chih-Chieh Yen, Ruei-Hung Tsai, I-Ting Liu, Bo-Gen Lai, Yi-Ching Yang, Yu-Sheng Lin, Yi-Jing Chen, Yi-Lin Wu, Chih-Chun Lin
Purpose:

In this study, we developed an AIoT-based Feedback Assistive Strengthening Ergometer, AIFASE, to deliver tailored physical training with safety monitoring for patients with CC.

Methods:

AIFASE provides personalized exercise courses based on a decision tree algorithm, while the heart rate prediction model ensures safe monitoring, and IoT integration enables real-time medical alerts to healthcare providers. In this study, eleven older cancer patients with pre-cachexia/cachexia participated (age 74±5.31 years). Personalized physical training was provided by AIFASE, consisting of sessions twice a week, each lasting 12 to 25 minutes, over a 12-week intervention period. The Mini CAchexia SCOre (MCASCO), and Senior Fitness Test (SFT) were evaluated before and after the intervention. Wilcoxon method was used to analyze the changes in values before and after the intervention.  

Results:

After 12 weeks of AIFASE intervention, MCASCO scores significantly decreased from 23.68±13.72 to 10.51±6.62 (p0.05). Significant improvements were observed across multiple physical fitness parameters after the intervention. Hand grip strength increased notably in both hands, with the left hand improving from 19.21±6.88 kg to 20.94±7.43 kg and the right hand from 20.34±8.80 kg to 22.91±8.83 kg, indicating enhanced overall muscle strength (p0.05, and p0.01, respectively). Aerobic endurance, as measured by 2 minute step test, improved from 74.18±31.04 to 90.09±20.34 steps, reflecting better cardiovascular performance (p0.05). Walking speed increased from 1.03±0.32 m/s to 1.10±0.33 m/s, demonstrating enhanced mobility and functional capacity (p0.05). Additionally, dynamic balance, measured by the Timed Up and Go (TUG) test, improved significantly, increasing from 13.47±16.72 seconds to 10.41±10.94 seconds, suggesting better stability and reduced fall risk (p0.01).

Conclusion(s):

AIFASE provides personalized and safely monitored exercise programs, which can slow the progression of cancer cachexia and improve patient physical fitness. The application of AIFASE provides a scalable solution for long-term management of cancer cachexia.

Implications:

The AIoT-based strengthening ergometer system provides safe and personalized exercise recommendations, enhancing the role of physical therapists in physical and exercise training for cancer cachexia, and potentially improving patient outcomes.

Funding acknowledgements:
National Health Research Institutes ( NHRI-13A1-CG-CO-04-2225-1), National Science and Technology Council (NSTC113-2221-E214-006), and MoE in Taiwan (Higher Education Sprout Project: MDIC).
Keywords:
Artificial Intelligence of Things
Ergometer
Cancer Cachexia
Primary topic:
Innovative technology: information management, big data and artificial intelligence
Second topic:
Health promotion and wellbeing/healthy ageing/physical activity
Third topic:
Older people
Did this work require ethics approval?:
Yes
Name the institution and ethics committee that approved your work:
This study was approved by the Institutional Review Board of National Cheng Kung University Hospital (NCKUH)
Provide the ethics approval number:
A-BR-112-025-T
Has any of this material been/due to be published or presented at another national or international conference prior to the World Physiotherapy Congress 2025?:
No

Back to the listing