Background & Aims

An estimated 19% of Canadians are living with a chronic pain condition1, and 2/3 of chronic pain patients express dissatisfaction with available treatment options2. The benefits of art therapy to reduce self-reported pain levels, anxiety, and depression — which contribute to the maintenance of chronic pain — have been extensively documented. However, while only 2% of chronic pain patients have access to pain clinics3, even fewer have access to the support of art therapy professionals. To address this critical need, we are developing novel artificial intelligence (AI) supported computer and extended reality (XR) programs that assist patients in generating and exploring visual representations of their chronic pain experience, both for their own direct benefit. We suspect that in the future, this could be used as a tool to aid in communication with other members of the chronic pain patient community.

Methods

To meet the need for effective non-pharmacological interventions for chronic pain conditions and to facilitate new opportunities for patient therapeutic self-expression, we will create a set of machine learning-based tools specifically designed for this population and systematically test its therapeutic potential. We will: 1) Create the first database of visual art relevant to the experience of living with a chronic pain condition. This database will be used to seed a first-of-its-kind application of a machine learning algorithm which will generate novel images for patient self-expression purposes; 2) Evaluate the efficacy of this self-expression tool, delivered through a computer interface, as a therapeutic for chronic pain conditions; 3) Generate immersive extended reality (XR) interfaces. We will also experiment with several strategies to support transformations, such as combining and morphing elements from several selected images, without requiring artistic skills.

Results

Through a collaboration with the National Gallery of Art, we have assembled an extensive database of digitized artworks from their collection to train machine learning algorithms and systematically test their utility as part of computer-based and XR-based interventions for individuals living with chronic pain conditions. We will work with an art historian and an art therapist to curate several versions of machine learning algorithms and investigate which of these approaches provides the most useful material for patient-led journaling of the chronic pain experience.

Conclusions

Artificial intelligence-based approaches in the field of chronic pain have thus far taken diagnostic and prognostic approaches. The application of machine learning techniques to therapeutic applications have thus far been underexplored. This research seeks to address a gap in the literature and open a new avenue for non-pharmacological interventions which seek to enhance current approaches that act on key psychosocial factors underlying the chronic pain experience and provide a new patient-led journaling tool which can be used outside of clinical spaces.

References

Schopflocher D., T., P., Jovey, R. (2011). The prevalence of chronic pain in Canada. Pain Research and Management, 16, 445-450.

Geurts, J. W., Willems, P. C., Lockwood, C., van Kleef, M., Kleijnen, J., & Dirksen, C. (2017). Patient expectations for management of chronic non-cancer pain: A systematic review. Health expectations : an international journal of public participation in health care and health policy, 20(6), 1201–1217. https://doi.org/10.1111/hex.12527

Breivik, H., Collett, B., Ventafridda, V., Cohen, R., & Gallacher, D. (2006). Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. European journal of pain (London, England), 10(4), 287–333. https://doi.org/10.1016/j.ejpain.2005.06.009

Presenting Author

Hannah Derue

Poster Authors

Hannah Derue

BSc(Hons)

McGill University

Lead Author

E-mail

James Newman

McGill University

Lead Author

Olivia Song

McGill University

Lead Author

Samantha Astles

McGill University

Lead Author

Topics

  • Novel Experimental/Analytic Approaches/Tools