The vocal professionals are prone to voice disorders as a result of excessive exposure to high intensity voices that can impact their job performance and the quality of their life. Standardized treatments often suffer from lack of multi-dimensional optimization and lack of personalization. The aim of this study is to overcome such shortcomings by offering a multi-objective optimization algorithm-based strategy of building a vocal rehabilitation model and personalized treatment pathways with the view to achieving synergistic optimization on various therapeutic dimensions. Based on the combination of acoustic properties, physiologic measurements, and subjective perception ratings, a multi-objective optimization model is constructed. An improved Non-dominated Sorting Genetic Algorithm II (NSGA-II) is created to effectively search through the set of Pareto-optimal solutions to the treatment parameters. The experimental evidence shows that the proposed solution is much more effective than the traditional single-objective optimization and manual empirical solutions in terms of reducing the duration of the treatment and increasing patient comfort as well as ensuring high accuracy in acoustic restoration. These findings offer a new data-driven framework in personalized medical care in voice rehabilitation and extend the use of multi-objective optimization algorithms in biomedical engineering.
