Automated Physiological Assessment of Chronic Pain in Daily Life

点击翻译按钮获取中文摘要

The United States is in the midst of dual epidemics of chronic pain and opioid abuse, with approx. 20% of the population in persistent pain, and over 40,000 lives lost each year to opioid misuse. Chronic back pain (CBP) is the most common pain disorder and one of the major reasons for prescribing opioids. Strategies to help reduce CBP pain without opioids are therefore urgent. A promising opioid alternative are psychological interventions that reduce pain intensity, interference and negative emotions, and do not just target the physical pain intensity as many of the traditional pharmacological approaches do. However, these interventions are not often temporally aligned with pain episodes. We propose to establish diagnostic physiological markers of ongoing clinical pain by capturing ongoing clinical pain and the associated physiological fluctuations and psychological processes. We will develop fully automated real-time detection of ongoing pain in N=80 CBP patients from physiological signs collected in everyday life. We will record multiple physiological signals (electroencephalogram (EEG), facial electromyography (EMG), electrooculography (EOG), electrodermal activity (EDA), and heart rate (HR)) from two wearable device, one worn around the ears (Earable) and one worn around the wrist (Empatica). The sensing system will be integrated with an experience sampling method (ESM) smartphone app to collect ratings of pain and psychological processes associated with pain episodes. Our goal in Aim 1 is to establish computational physiology-based models that can predict clinical pain in real-life. To achieve this, we will apply machine-learning techniques to physiological data preceding pain self-reports to build predictive models of ongoing pain, with the ultimate goal for these computational models to be able to trigger psychological interventions when needed most, which we aim to develop in our future research. Our goal in Aim 2 is to field-test these computational models in a new group of N=20 CBP patients. The proposed work will afford, for the first time, autonomous monitoring of clinical pain in real-life. If the real-life pain experience of patients can be captured in physiological patterns preceding pain, then automated tracking of physiology has considerable potential to improve the efficacy of psychological treatments, by providing signals to trigger just-in-time interventions. Overall, the proposed project will contribute fundamental scientific knowledge about psycho-physiological signs of real-life pain and lay the groundwork for translational efforts to improve outcomes of pain self-management and reduce opioid use.