An Introduction to Machine Learning for Speech-Language Pathologists: Concepts, Terminology, and Emerging...

Area of Science:

• Speech-language pathology
• Computational linguistics
• Artificial intelligence

Background:

• Machine learning (ML) offers advanced analytical capabilities.
• The integration of ML in speech-language pathology is an emerging area.
• Understanding ML is crucial for future research and clinical practice.

Purpose of the Study:

• To introduce clinicians and researchers to machine learning (ML) approaches in speech-language pathology.
• To provide an overview of key ML concepts, terminology, and models.
• To discuss the challenges and limitations of implementing ML in the field.

Main Methods:

• Introduction to fundamental ML concepts (e.g., classification, regression).
• Explanation of various ML models (e.g., random forest, support vector machines, neural networks).
• Review of six exemplar studies applying ML in aphasiology.

Main Results:

• Demonstration of ML applications in speech-language pathology through case studies.
• Discussion of critical considerations for ML model development and evaluation.
• Identification of limitations including reliability, validity, bias, and explainability.

Conclusions:

• ML holds significant potential for advancing speech-language pathology.
• Addressing challenges in reliability, validity, and explainability is key for clinical adoption.
• Responsible implementation of ML is essential for meaningful clinical impact.