Enhancing breast cancer awareness through artificial intelligence: Bridging the gap for low-literacy populations.

Abstract

e13519 Background: Breast cancer poses a significant threat to women's health, as evidenced by the alarming statistics that it is the second most common form of cancer among females. It is responsible for a staggering 30% of all female cancers, according to data from the American Cancer Society. Furthermore, the incidence of breast cancer is projected to rise by 1% annually by 2025. This can be attributed to various factors, including low literacy levels and lack of health care awareness. In this project, we explore how Chat Generated Pre-Trained Transform (ChatGPT), an emerging artificial intelligence, can be leveraged to simplify the abundant data about breast cancer available on the National Breast Cancer Foundation website, making it more accessible to low-literacy populations. Methods: We aimed to examine the readability of the 17 breast cancer-related questions and answers provided by the National Breast Cancer Foundation on their website. Additionally, we investigated the effectiveness of using ChatGPT in simplifying the content to ensure it is easily understood by individuals with a sixth-grade reading level. To compare the original text and the readability of ChatGPT-generated text, we used the Flesch-Kincaid scoring criteria to understand the education and reading levels needed to interpret this text. We then employed STATA BE 18.5 software using a paired t-test to statistically analyze the Flesch-Kincaid grade level and readability ease score of the original text and ChatGPT-generated text. Results: The results of the paired t-test analysis revealed a significant difference between the mean values of original and ChatGPT-modified text, with a standard error of 0.54 and a standard deviation of 2.23. The 95% confidence interval for the mean difference ranged from 1.38 to 3.67, with the mean difference being 2.523 and a corresponding p-value of 0.0003, which is statistically significant (p < 0.05). Conclusions: Breast cancer continues to be a major public health concern, particularly among women, and there is a pressing need to develop innovative approaches to enhance awareness and encourage early detection. Emerging research suggests that artificial intelligence technologies, such as chatbots, may provide a promising solution to address the barriers faced by individuals with low literacy rates in accessing and comprehending crucial health information.