Proposed MA Thesis Topic: AI & Communication Disorders

AI-Powered Speech, Language, and Communication Assessment: Unlocking Insights into Developmental and Acquired Communication Challenges

Master's Supervision • Department of Special Needs Education (ISP), University of Oslo

Overview

We invite motivated Master's students in Speech-Language Pathology (Logopedi), Special Needs Education, and Cognitive/Computational Sciences to join our research environment investigating novel Artificial Intelligence (AI), Machine Learning (ML), and Natural Language Processing (NLP) methods for developmental and acquired neurocognitive communication disorders.

Students will gain hands-on experience with state-of-the-art computational language assessment tools—including our cloud platform Open Brain AI (openbrainai.com)—to investigate diagnostic, prognostic, and therapy-monitoring biomarkers. Specific research questions can be tailored to each student's clinical or methodological interests.

Goals

Research Objectives

  • Explore AI techniques (machine learning, deep learning, and natural language processing) to automatically analyze spoken and written language productions of individuals with developmental or acquired neurogenic communication disorders.
  • Compare the performance and reliability of automated AI-based assessment methods with traditional manual clinical assessments in terms of accuracy, efficiency, objectivity, and validity.
  • Investigate the clinical utility of AI-based language biomarkers for differential diagnosis, prognosis, treatment efficacy evaluation, and individualized therapy planning.
Core Questions

Research Questions

  • How can multimodal AI and NLP techniques be applied to speech, language, and discourse assessment in neurogenic and neurodevelopmental conditions?
  • How do automated digital biomarkers compare with standardized manual clinical scoring across phonology, morphosyntax, lexical semantics, and prosody?
  • What are the practical benefits and ethical/clinical considerations of integrating AI decision-support tools into speech-language pathology practice?

Proposed Research Plan

  1. Literature Review: Survey state-of-the-art computational and AI methods for speech and language assessment alongside the strengths and limitations of traditional manual batteries.
  2. Computational Pipeline & Open Brain AI: Utilize or adapt automated assessment modules within Open Brain AI to extract acoustic, phonological, morphosyntactic, and lexical-semantic metrics from spoken or written samples.
  3. Clinical or Normative Corpus Analysis: Analyze spoken or written discourse productions (e.g., picture descriptions, narrative retells, procedural discourse) from clinical cohorts and healthy controls.
  4. Validation & Statistical Modeling: Benchmark automated AI metrics against expert clinician ratings and standardized clinical test scores using modern statistical and machine learning models.
  5. Clinical Translation: Synthesize implications for early screening, subtyping, longitudinal monitoring, and individualized speech-language therapy target selection.

Interested in Discussing an MA Thesis Project?

Please reach out via email with a brief description of your background and research interests:

Prof. Charalambos Themistocleous: charalth@uio.no / charalampos.themistokleous@isp.uio.no