Computational Intelligence for Adult Autistic Trait Detection
PhD Researcher exploring the intersection of computational intelligence and psychology to develop innovative methods for adult autistic trait detection.
PhD Researcher exploring the intersection of computational intelligence and psychology to develop innovative methods for adult autistic trait detection.
PhD researcher with a focus on computational intelligence applications in autism identification.
Specialised in software engineering at undergrad and computational intelligence and machine learning at MSc with applications in healthcare.
Developing novel computational methods for detecting and analysing autistic traits in adults.
I am a PhD researcher specialising in the application of computational intelligence techniques for the detection and analysis of autistic traits in adults. My research combines expertise in machine learning, data analysis, and psychology to develop innovative diagnostic tools and methodologies.
With a background in computer science, I aim to bring interdisciplinary techniques to the world of psychology. I am passionate about creating accessible, accurate, and ethical computational methods that can assist in the identification and understanding of autism spectrum conditions in adult populations.
My work aims to address the challenges in adult autism diagnosis through the development of data-driven approaches that can complement traditional clinical assessments.
My research focuses on developing computational intelligence methods for the detection and analysis of autistic traits in adult populations.
Creating and evaluating computational models that can identify patterns associated with autistic traits from various data sources, including behavioural, linguistic, and physiological data.
Applying advanced statistical techniques to analyse and interpret data related to autistic traits, developing metrics that can quantify the presence and severity of these traits.
Selected peer-reviewed publications related to my research on computational intelligence for adult autistic trait detection.
Journal TBD, Year TBC
Authors
D. J. Buxton, J. J. Bird, M. Muhamad, T. Hughes-Roberts and D. J. Brown
Abstract
This systematic review examines computational approaches to identify traits of autism spectrum condition (ASC) among adults, focussing on methodologies involving machine learning and deep learning. A comprehensive search was conducted using the Scopus database, covering studies published between 2014 and January 2024. Of the 454 initial articles retrieved, 30 were included based on their relevance, where the quality of their machine and deep learning methods were assessed using metrics including precision and recall. This review categorises the included studies by the various modalities that are used by the machine and deep learning methods they adopted, such as neuroimaging, facial (eye gaze), questionnaires, and speech and language analysis. Neuroimaging emerged as the most prevalent modality incorporated into models, accounting for 46.7% of the studies. This review highlights the strengths and limitations of various methods adopted by the included studies, emphasising the need for high precision to also be considered as a metric, to minimise the occurrence of false positives and negatives as the outputs of the models. The findings suggest that whilst significant progress has been made, there is still a need for more robust and generalisable models. The review concludes with a call for further research integrating multi-modal data and advanced techniques to enhance the accuracy and reliability of ASC detection.
Interested in my research or potential collaborations? Feel free to reach out.
You can also reach me directly through:
I am open to research collaborations, speaking engagements, and consultations related to computational intelligence and autism research.