NAVIA: Neurodiversity Assessment and Voice-enabled Intervention AI

Project: Applied Research

Project Details

Abstract

The NAVIA project aims to pioneer a holistic framework for advancing early screening and intervention for Autism Spectrum Disorder (ASD) by integrating multimodal AI technologies, with a strong emphasis on ethical and inclusive practices. Its core objective is to develop AI models capable of detecting early indicators of autism and generating comprehensive pre-diagnostic reports to support clinical assessments. A key component is the creation of a novel, richly annotated dataset capturing speech and behavioral cues associated with ASD. These signals will be analyzed to identify salient markers that will drive the development of accurate, culturally sensitive AI models. Building on the screening insights, NAVIA will develop an AI-powered platform to support the clinicians and caregivers in generating personalized speech and language therapy exercises, tailored to each child’s unique communication profile and challenges. The project recognizes autistic speech and behavioural patterns as culturally meaningful forms of communication and commits to designing tools that ethically respect these expressions. A core pillar of the project is the integration of ethical AI principles, ensuring that all stages, from data collection to model deployment, are guided by transparency, fairness, cultural sensitivity, and accountability, promoting inclusive design principles that respect the communicative legitimacy of autistic expressions.

Submitting Institute Name

Hamad Bin Khalifa University (HBKU)
Sponsor's Award NumberHBKU-INT-VPR-FRG-03-09
Proposal IDHBKU-OVPR-FRG-03-3
StatusActive
Effective start/end date10/07/259/07/27

Primary Theme

  • Artificial Intelligence

Primary Subtheme

  • AI - Healthcare

Secondary Theme

  • Artificial Intelligence

Secondary Subtheme

  • AI - Education

Keywords

  • AI Application
  • Autism: Causes, Diagnosis, and Intervention
  • Multimodal

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.