Related Experiment Video
Updated: May 16, 2025
09:47
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
918
Beyond the Posts: Analyzing Breast Implant Illness Discourse With Natural Language Processing and Deep Learning.
Aesthetic Surgery Journal
|April 2, 2025
Summary
Patient concerns about Breast Implant Illness (BII) on social media are largely negative and fearful, correlating with increased breast implant removal rates. This highlights social media
Area of Science:
- Social media analysis
- Natural language processing (NLP)
- Patient-reported outcomes
Background:
- Breast Implant Illness (BII) is a collection of symptoms attributed to breast implants, with growing patient interest despite unproven causality.
- Social media platforms significantly influence healthcare decisions, making patient perceptions of BII on these platforms critical.
- Understanding online discourse is essential for healthcare providers and researchers addressing patient concerns.
Purpose of the Study:
- To analyze patient perceptions and emotional responses related to Breast Implant Illness (BII) discussed on X (formerly Twitter).
- To utilize RoBERTa, a natural language processing model, to analyze a large dataset of social media posts.
- To compare emotional trends and sentiment over time and correlate them with breast implant explantation data.
Main Methods:
- Analysis of 6,099 posts mentioning BII from 2014-2023 using NLP models for sentiment and emotion detection (fear, sadness, anger, disgust, neutral, surprise, joy).
- Classification of posts by highest-scoring emotion and comparison of results between 2014-2018 and 2019-2023 periods.
- Pearson correlation analysis was performed between social media data and published breast implant explantation and augmentation statistics.
Main Results:
- 75.4% of analyzed posts exhibited negative sentiment, with a notable peak in March 2019.
- Neutral and fear were the dominant emotions (35.9% and 35.6% respectively), with fear scores increasing significantly from 2014-2018 to 2019-2023.
- Strong positive correlations (r>0.70) were found between explantation rates and the volume of negative, neutral, and fear-based posts.
Conclusions:
- Online discourse on Breast Implant Illness (BII) peaked in 2019, characterized by predominantly negative sentiment and fear.
- The significant correlation between fear-based social media content and increased breast implant explantation rates suggests a strong influence on patient decisions.
- Social media plays a crucial role in shaping patient perceptions and choices regarding breast implant removal.
