Multiple Sclerosis, Pain
Conditions
Keywords
Pain assessment, physiological signals, wearable sensors, artificial intelligence, machine learning
Brief summary
This exploratory interventional study aims at exploring the feasibility of using physiological signals recorded through wearable devices, together with artificial intelligence techniques, to assess pain automatically and objectively. Automatic methods to assess presence/absence of pain, discern nociceptive from neuropathic pain, and estimate the intensity of pain will be trained an tested on a population of multiple sclerosis patients undergoing neurorehabilitation.
Detailed description
In patients with Multiple Sclerosis (MS), pain is one of the most common symptoms. The pain described by MS patients is often diffuse, chronic, and debilitating, generally associated with psychological distress and decreased daily functioning. The presence of pain adversely affects the neurorehabilitation process itself. Patients with pain may refuse to participate in therapy sessions or request to terminate early. However, the link between the frequency and/or intensity of pain and the rehabilitation process is largely unexplored. This is also exacerbated by the different sources of pain experienced by MS patients who require neurorehabilitative interventions. In clinical practice, pain assessment is conducted mainly using self-administered questionnaires or scales. These tools however can be influenced by many factors, including emotional or cognitive aspects and cannot give an objective measure of the pain experience. To date, there are no objective and simple-to-use clinical methods that allow objective quantification of the painful experience and a diagnostic differentiation between the two main types of pain, which are nociceptive pain (arising from nociceptive stimuli), and neuropathic pain (caused by a lesion or a pathology of the somatosensory nervous system). In this sense, wearable technologies which can continuously monitor physiological parameters related to pain can be used for the quantification of physiological measures related to pain experience. AIMS: This study aims at exploring the feasibility of developing methods based on wearable sensors and artificial intelligence algorithms to assess pain objectively and automatically in patients undergoing neurorehabilitation. The specific objectives of this study are the following: * To assess the feasibility of developing a differential diagnosis method to evaluate the absence or presence of pain * To assess the feasibility of developing a regression model to evaluate the intensity of pain * To assess the feasibility of developing a differential diagnosis method to discern the type of pain (i.e., nociceptive vs. neuropathic pain)
Interventions
DIAGNOSTIC_TESTmonitoring
The intervention consists of 48-h monitoring by using two types of monitoring: an objective monitoring, through a class IIa wearable medical device recording four physiological signals, and a subjective monitoring through a questionnaire developed with Microsoft Forms that can be compiled with a smartphone. The monitoring will be conducted during a motor neurorehabilitation treatment, 24 hours before and 24 hours after the treatment at the participant's home. Besides this monitoring, stratification questionnaires will be administered to each participant to be stratified in one of the three categories (absence of pain, nociceptive pain, or neuropathic pain) based on the following timeline: t0: baseline t1: pre-treatment t2: post-treatment t3: follow-up
Sponsors
Azienda Usl di Bologna
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
No
Inclusion criteria
* Age between 18 and 75 years * Diagnosis of Multiple Sclerosis for at least three months post-onset * Prescription of a physiotherapy-based motor rehabilitation program * Signature of the informed consent to participate in the study
Exclusion criteria
* Heart rhythm modifying disease and/or factors such as arrhythmogenic heart disease (e.g., atrial fibrillation), presence of pacemakers and/or use of drugs capable of affecting heart rhythm, such as beta blockers (C07) or other antiarrhythmic drugs (C01) * Cognitive impairments that preclude the possibility of providing valid informed consent, such as a disorder of consciousness or confusional state, the latter defined by temporal and/or spatial disorientation detected during ordinary conversation. In case of doubt, a simple confusional state assessment test (4AT) will be administered before enrollment * Language comprehension skills lower than 75% in an ordinary conversation due to aphasic disorder of severe deafness despite the use of a hearing aid. In case of doubt, a simple language comprehension test (token test) will be administered before enrollment * Linguistic expression less than 75%. In case of doubt, a simple verbal fluency test (verbal fluency by phonemic category) will be administered before enrollment * Severe psychiatric comorbidity that may interfere with adherence to the study protocol (e.g., severe personality disorders, severe psychomotor agitation) * History or current use of narcotic drugs (including marijuana) * Modification in the two weeks prior to enrollment or foreseeable modification during enrollment of any chronic pain management program, both pharmacological (cortisone for systemic use, H02; antirheumatics, M01; analgesics, N02; antiepileptics, N03; antidepressants tricyclics, N06AA; atypical antidepressants such as duloxetine or venlafaxine, N06AX) and non-pharmacological (e.g., acupuncture or other manual therapies, physical therapies, such as tecar therapy)
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of registrations | The monitoring will be conducted during the intervention time frame [48 hours] | Number of concurrent physiological signal registrations and pain assessments through CRF monitoring questionnaire and CRF monitoring questionnaire-intervention. Diagnostic performance of the classifier (i.e., sensitivity, specificity, predictive value) against the gold standard (outcomes from CRF monitoring questionnaire and CRF monitoring questionnaire-intervention). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| regression model - intensity of pain | 24 hours before and 24 hours after the treatment at the participant's home | Number of concurrent physiological signal registrations and pain assessments through CRF monitoring questionnaire and CRF monitoring questionnaire-intervention. Coefficient of determination of the regression model against the gold standard (outcomes from CRF monitoring questionnaires and CRF monitoring questionnaires-intervention). |
Countries
Italy
Contacts
Primary ContactFabio La Porta
Outcome results
None listed