A radial basis function neural network (RBFNN) approach for structural classification of thyroid diseases.

Area of Science:

• Endocrinology
• Medical Informatics
• Artificial Intelligence

Background:

• Thyroid diseases impact multiple organ systems, necessitating accurate diagnosis for effective treatment and prognosis.
• Accurate diagnosis of thyroid conditions is crucial for guiding therapeutic decisions and assessing patient outcomes.
• The complexity of thyroid disease classification highlights the need for advanced diagnostic tools.

Purpose of the Study:

• To investigate the efficacy of Multilayer Perceptron Neural Network (MLPNN) and Radial Basis Function Neural Network (RBFNN) for the structural classification of thyroid diseases.
• To compare the performance of MLPNN and RBFNN models in classifying thyroid disease structures using a patient dataset.
• To assess the utility of artificial neural networks in supporting the diagnosis of thyroid disorders.

Main Methods:

• A dataset comprising 487 patients with thyroid disease was utilized.
• Two types of neural networks, MLPNN and RBFNN, were built, trained, and tested.
• Structural classification was initially performed by expert physicians to establish a benchmark for the neural network models.

Main Results:

• Both MLPNN and RBFNN models achieved highly satisfactory predictions on the learning datasets.
• The Radial Basis Function Neural Network (RBFNN) model exhibited superior performance compared to the Multilayer Perceptron Neural Network (MLPNN) model on the evaluation dataset.
• The study confirmed the potential of artificial neural networks in accurately classifying thyroid disease structures.

Conclusions:

• Artificial neural networks, particularly RBFNN, show significant promise for the structural classification of thyroid diseases.
• The findings support the integration of AI-driven tools in clinical practice for improved thyroid disease diagnosis.
• This research underscores the strong utility of neural network models in enhancing the accuracy and efficiency of thyroid disease assessment.