Prediction of recurrent spontaneous abortion using evolutionary machine learning with joint self-adaptive sime mould...

Area of Science:

• Reproductive endocrinology
• Medical informatics
• Computational biology

Background:

• Recurrent spontaneous abortion (RSA) presents significant challenges in reproductive health, often exacerbated by thyroid dysfunction.
• The underlying causes and effective treatments for RSA remain incompletely understood, necessitating advanced analytical approaches.
• Thyroid disorders are frequently implicated in RSA, highlighting the need for integrated diagnostic strategies.

Purpose of the Study:

• To develop an effective analytical framework for recurrent spontaneous abortion (RSA), particularly when associated with thyroid disorders.
• To integrate clinical data, vitamin D levels, and thyroid function measurements for improved RSA analysis.
• To introduce a novel computational framework, JASMA-SVM, for enhanced RSA diagnosis.

Main Methods:

• The study employed the joint self-adaptive sime mould algorithm (JASMA) combined with support vector machine (SVM) theory, termed JASMA-SVM.
• JASMA was utilized for its adaptive parameter optimization and feature selection capabilities, enhancing SVM performance.
• The JASMA-SVM model was trained and validated using clinical data including vitamin D levels, thyroid hormones, and autoantibodies from women with RSA.

Main Results:

• The JASMA-SVM framework demonstrated high diagnostic performance for RSA.
• Achieved accuracy of 92.998%, Matthews Correlation Coefficient (MCC) of 0.92425, sensitivity of 93.286%, and specificity of 93.064%.
• The JASMA algorithm's global search and optimization capabilities were validated on CEC 2014 benchmarks.

Conclusions:

• The JASMA-SVM model shows promise as a valuable tool for the diagnosis of recurrent spontaneous abortion (RSA).
• Integrating thyroid function and vitamin D measurements within this computational framework improves diagnostic accuracy.
• This approach offers a potential pathway for more effective management of RSA in clinical practice.