The article addresses the ubiquitous challenge of uncertainty in decision-making, with a particular focus on medical decision-making, through the innovative application of logistic regression enhanced by interval-valued fuzzy set theory. Traditional logistic regression relies on a linear combination of variables and a uniform set of regression coefficients, which can inaccurately represent the variability and uncertainty inherent in real-world data. Our proposed methodology differs in that it incorporates weights with interval values into the logistic regression model, allowing for a more nuanced and flexible representation of the data. This approach allows the model to adjust the weights independently in terms of values, offering a fit to interval data and improving the precision of predictions. By developing a specialized algorithm to calculate weighted coefficients adjusted to specific inputs or attributes, we demonstrate the practical effectiveness of our method in dealing with uncertainty. Experimental results highlight the potential of interval-valued fuzzy sets in improving machine learning techniques and enhancing the accuracy of decision-making models in complex, uncertain environments.