This post is also available in: Español (Spanish)

Personalized pain treatment: the role of technology for taking decisions through big data and AI

Colloquially, the term artificial intelligence (AI) is applied when a machine imitates the “cognitive” functions that humans associate with other human minds, such as “perceive,” “reason,” “learn,” and “solve problems.” Machine learning (ML) is a reduced form of AI where algorithms make predictions to interpret data or “learn” without static instructions.

Modern physiotherapy has progressively and increasingly incorporated technology, but despite all the advances human cognition has limitations in the processing of complex data, in addition to the time consumption that this implies for the professional. Today we have access to a large amount of data from our patients through multiple sensors, APPs or smartwatches, which measure different variables more or less reliably, but offer non-integrated information and therefore difficult to process for the physiotherapist and with a high cost of time that makes it unfeasible on a day-to-day basis.

As a result of this need to provide physiotherapists with information ready to consume that they can apply to the treatment of their patients and based on the scientific evidence provided by the dynamic analysis of this big data, the Hefora project arises, in which the group of iPhysio research of the San Jorge University together with the Howlab research group of the University of Zaragoza and the companies Geoslab and Fisio Consultors. The platform, which is initially in a beta version, allows structured patient data to be collected, as well as prescribing exercises and keeping a personalized follow-up of patients. The structuring of all these data and their changes over time allow subsequent analysis in an anonymous way to identify related factors, causality and / or patterns.

Supervised ML has great potential to evolve physiotherapy towards a 3.0 physiotherapy. An important challenge could be the prediction of the risk of developing chronic pain based on modifiable (lifestyle) and non-modifiable (genetic) variables, to see to what extent a phenotype is more or less responsible for the risk of suffering pain or other problems of health, as well as to what extent the interrelation of different factors allows to modify the expression of a genetic conditioner in real life. Another example could be the measurement of adherence and its improvement, or to know if the exercise has been well done by the patient without being supervised by the physiotherapist.

Regarding predictive models in physiotherapy, there are great challenges such as analyzing to what extent reliable models can be developed that show how certain educational programs or the empowerment of the patient regarding their lifestyle can achieve changes in the determining variables of chronic pain and help Improve the health of individuals.

Pablo Herrero

  • Dr. Pablo Herrero is Head of Research at iPhysio Research Group and Lecturer at San Jorge University.

  • He holds a doctorate in Physical Therapy from Zaragoza University (Spain) with European Mention after spending a research stage at Keele University (UK).

  • Dr. Herrero is specialist in invasive techniques, teaching about dry needling for the treatment of myofascial pain at pre-graduate (San Jorge University) and post-graduate courses in different countries.

  • He is the author of DNHS® technique and method, a dry needling technique to treat spasticity, which has been expanded to different countries all through the world. He is also inventor of the patented 3TOOL®, an innovative tool to help physios with treatments and also for self-treatment.

  • He is also President of the Association for Research in Motor Handicap (AIDIMO) and Editor-in-Chief of the Journal of Invasive Techniques in Physical Therapy.

  • He has experience in European projects related to technologies for rehabilitation, leading the project “3TOOLing your Health” which ended in a platform to prescribe therapeutic exercises to patients remotely and to monitor their evolution (see www.hefora.com) and he is currently leading the European Project Prevent4Work, awarded with 898476€ for the years 2018-2021 with European SMEs, enterprises and Universities.

Pablo Herrero

  • Dr. Pablo Herrero is Head of Research at iPhysio Research Group and Lecturer at San Jorge University.

  • He holds a doctorate in Physical Therapy from Zaragoza University (Spain) with European Mention after spending a research stage at Keele University (UK).

  • Dr. Herrero is specialist in invasive techniques, teaching about dry needling for the treatment of myofascial pain at pre-graduate (San Jorge University) and post-graduate courses in different countries.

  • He is the author of DNHS® technique and method, a dry needling technique to treat spasticity, which has been expanded to different countries all through the world. He is also inventor of the patented 3TOOL®, an innovative tool to help physios with treatments and also for self-treatment.

  • He is also President of the Association for Research in Motor Handicap (AIDIMO) and Editor-in-Chief of the Journal of Invasive Techniques in Physical Therapy.

  • He has experience in European projects related to technologies for rehabilitation, leading the project “3TOOLing your Health” which ended in a platform to prescribe therapeutic exercises to patients remotely and to monitor their evolution (see www.hefora.com) and he is currently leading the European Project Prevent4Work, awarded with 898476€ for the years 2018-2021 with European SMEs, enterprises and Universities.

Personalized pain treatment: the role of technology for taking decisions through big data and AI

Colloquially, the term artificial intelligence (AI) is applied when a machine imitates the “cognitive” functions that humans associate with other human minds, such as “perceive,” “reason,” “learn,” and “solve problems.” Machine learning (ML) is a reduced form of AI where algorithms make predictions to interpret data or “learn” without static instructions.

Modern physiotherapy has progressively and increasingly incorporated technology, but despite all the advances human cognition has limitations in the processing of complex data, in addition to the time consumption that this implies for the professional. Today we have access to a large amount of data from our patients through multiple sensors, APPs or smartwatches, which measure different variables more or less reliably, but offer non-integrated information and therefore difficult to process for the physiotherapist and with a high cost of time that makes it unfeasible on a day-to-day basis.

As a result of this need to provide physiotherapists with information ready to consume that they can apply to the treatment of their patients and based on the scientific evidence provided by the dynamic analysis of this big data, the Hefora project arises, in which the group of iPhysio research of the San Jorge University together with the Howlab research group of the University of Zaragoza and the companies Geoslab and Fisio Consultors. The platform, which is initially in a beta version, allows structured patient data to be collected, as well as prescribing exercises and keeping a personalized follow-up of patients. The structuring of all these data and their changes over time allow subsequent analysis in an anonymous way to identify related factors, causality and / or patterns.

Supervised ML has great potential to evolve physiotherapy towards a 3.0 physiotherapy. An important challenge could be the prediction of the risk of developing chronic pain based on modifiable (lifestyle) and non-modifiable (genetic) variables, to see to what extent a phenotype is more or less responsible for the risk of suffering pain or other problems of health, as well as to what extent the interrelation of different factors allows to modify the expression of a genetic conditioner in real life. Another example could be the measurement of adherence and its improvement, or to know if the exercise has been well done by the patient without being supervised by the physiotherapist.

Regarding predictive models in physiotherapy, there are great challenges such as analyzing to what extent reliable models can be developed that show how certain educational programs or the empowerment of the patient regarding their lifestyle can achieve changes in the determining variables of chronic pain and help Improve the health of individuals.