Machine learning-driven prediction of decisional uncertainty among medical students post-Kahramanmar

admin

https://healthpr.org/journal/HPR/articles/online_first/7917

admin

correction of title: post-kahramanmaras earthquake.

admin

related link:
https://healthpr.org/journal/HPR/articles/online_first/7917
abstract:
The article is titled "Machine learning-driven prediction of decisional uncertainty among medical students post-Kahramanmaras earthquake." It's a research paper investigating the psychological effects of a major natural disaster on people not directly involved.

Here is a summary of the key points:

Research Gap & Objective: It's challenging to predict mental health outcomes for "indirect victims" of a disaster (such as family and friends of those directly affected). This study explored a new method, using machine learning to see if it could predict "decisional uncertainty" (answering "not sure" to personal questions) in medical students following the Kahramanmaras earthquake.

Methods: The study used a proxy measure for decisional uncertainty, calculated from participants' "not sure" responses to questions about their hobbies, nutrition, job satisfaction, and academic success. Researchers then used machine learning models to analyze this data.

Key Findings:

    A weak positive correlation was found between the decisional uncertainty proxy and depression scores.

    Academic and nutritional status were stronger predictors of decisional uncertainty than depression or anxiety scores.

    The machine learning models showed some promise, achieving better-than-random accuracy in classifying decisional uncertainty.

Conclusion & Limitations:

    The findings provide a new, preliminary pathway for measuring indirect disaster effects but are not yet definitive.

    The authors emphasize the need for future studies to validate these results, use a clinically approved tool for decisional uncertainty, and note that the current study did not have external funding.