#ChronicPain: Automated Building of a Chronic Pain Cohort from Twitter Using Machine Learning

Abstract

Background: Due to the high burden of chronic pain, and the detrimental public health consequences of its treatment with opioids, there is a high-priority need to identify effective alternative therapies. Social media is a potentially valuable resource for knowledge about self-reported therapies by chronic pain sufferers. Methods: We attempted to (a) verify the presence of large-scale chronic pain-related chatter on Twitter, (b) develop natural language processing and machine learning methods for automatically detecting self-disclosures, (c) collect longitudinal data posted by them, and (d) semiautomatically analyze the types of chronic pain-related information reported by them. We collected data using chronic pain-related hashtags and keywords and manually annotated 4,998 posts to indicate if they were self-reports of chronic pain experiences. We trained and evaluated several state-of-the-art supervised text classification models and deployed the best-performing classifier. We collected all publicly available posts from detected cohort members and conducted manual and natural language processing-driven descriptive analyses. Results: Interannotator agreement for the binary annotation was 0.82 (Cohen’s kappa). The RoBERTa model performed best (F 1 score: 0.84; 95% confidence interval: 0.80 to 0.89), and we used this model to classify all collected unlabeled posts. We discovered 22,795 self-reported chronic pain sufferers and collected over 3 million of their past posts. Further analyses revealed information about, but not limited to, alternative treatments, patient sentiments about treatments, side effects, and self-management strategies. Conclusion: Our social media based approach will result in an automatically growing large cohort over time, and the data can be leveraged to identify effective opioid-alternative therapies for diverse chronic pain types.

Overview of the 8thSocial Media Mining for Health Applications (#SMM4H) Shared Tasks at the AMIA 2023 Annual Symposium

“…[3] For Task 2, the dataset included 5364 English-language tweets that mentioned a total of 32 therapies, such as medication, behavioral, and physical. [4] These tweets were posted by a cohort of users who self-reported chronic pain on Twitter, [11] so it is likely that the sentiments associated with the therapies are being expressed by patients who are actually experiencing them. For Task 3, the dataset included 10,150 tweets that were written in Latin American Spanish and reported COVID-19 symptoms of the user or someone known to the user, building on a previous iteration of this task involving the multi-class classification of Spanish-language tweets that mentioned COVID-19 symptoms.…”

A Bibliometric Study of Natural Language Processing Using Dimensions Database: Development, Research Trend, and Future Research Directions

“…Every new trend has its challenges, and since natural language processing tasks are fundamental in nature, this fact can be used to carefully suggest the direction of future research. There are a few areas of inquiry that will need further study in the future: (1) It is difficult to mine the medical text data stored in EHR or social media platforms because of imbalanced data, misspellings or ambiguity, high lexical variability, the difficulty of de-identification, the lack of key annotation, and the difficulty of obtaining all the semantic and syntactic attributes from complex sentences due to the lack of efficient methods that can capture all of them (Li et al, 2019;Pandey, Pandey, Mishra, & Rhmann, 2021;Sarker, Gonzalez-Hernandez, & Perrone, 2019;X. Wang, Hripcsak, Markatou, & Friedman, 2009).…”