Vidhusha Srinivasan*, Udayakumar N and Kavitha Anandan Pages 1 - 15 ( 15 )
Background: The spectrum of autism includes High Functioning Autism(HFA)and Low Functioning Autism(LFA). Brain mapping studies have revealed that, the autism individuals have overlaps in brain behavioural characteristics. Generally, the high functioning individuals are known to exhibit higher intelligence and better language processing abilities. However, the specific mechanisms associated with the functional capabilities are still under research.
Objective: This paper attempts to address the overlapping phenomenon present in autism through functional connectivity patterns along with, brain connectivity parameters and distinguish the classes using deep belief networks.
Methods: The task-based functional Magnetic Resonance Images (fMRI) of both groups were acquired from ABIDE database, for 58 low functioning against 43 high functioning individuals while they were involving in a defined language processing task. The language processing regions of the brain, along with the Default Mode Network (DMN) have been focussed. The functional connectivity maps have been plotted through graph theory. Brain connectivity parameters such as Granger Causality (GC) and Phase Slope Index (PSI) have been calculated for the groups. These parameters have been fed to the Deep Belief Networks (DBN) to classify the input as either LFA or HFA.
Results: Results showed an increased functional connectivity in high functioning subjects. It was found that, the additional interaction of the Primary Auditory Cortex with the other regions of interest complimented their enhanced connectivity. Results were validated using DBN measuring the classification accuracy of 85.85% for high functioning and 81.71% for low functioning group.
Conclusion: Since it is known that autism involves enhanced, but imbalanced components of intelligence, the reason behind the supremacy of high functioning group in language processing and region responsible for enhanced connectivity has been identified. Therefore, this work that signifies the effect of Primary Auditory Cortex in characterizing the dominance of language processing in high functioning young adults seems to be highly significant in discriminating the different groups in autism spectrum.
Autism, fMRI, Functional connectivity, Primary auditory cortex, Language Processing, Deep belief networks
Department of Information Technology, Centre for Healthcare Technologies, SSN College of Engineering, Rajiv Gandhi Salai (OMR), Chennai, Department of Pediatrics,Sri Ramachandra Institute of Higher Education and Research,Sri Ramachandra Medical University, Chennai, Department of Biomedical Engineering, Centre for Healthcare Technologies, SSN College of Engineering, Rajiv Gandhi Salai (OMR)