International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 1669

ISSN 2229-5518

Autism Child Survey and Discussion

Ramya,Dr.Savithri

Abstract— Learning disability (Dyslexia) is a neurological disorder. It affects the brains ability to receive process, store, respond to and communicate information. It is a language based disability in which a person has trouble understanding written words. It may also be referred to as reading disability or reading disorder. Some states and schools do not allow the term dyslexia to be used to describe a student’s reading difficulties. Our dyslexia information statistics tells us that 70 – 80% of people with poor reading skills across the world. In India 12 million children were affected from dyslexia. This paper presents a brief overview of dyslexiaand how it affects the population by adopting Artificial Neural Network techniques which has been appliedsuccessfully to solve problems in numerous fields. This methodology helps to identify and diagnosis problem using Artificial Neural Networks.

Index Terms— Neurological disorder, reading disability, dyslexia, Artificial Neural Networks, Mental disorder, Disability, Loss of Memory.

——————————  ——————————

1. Introduction
Dyslexia was identified by Oswald Berkhan but the term dyslexia was coined in 1887 by Rudolf Berlin, an ophthalmologist. He used the term to refer to a case of a young boy who had a severe impairment in learning to read and write despite showing typical intelligence and physical abilities in all other respects. In 1896, W. Pringle Morgan, a British physician published a description of a reading-specific learning disorder in a report titled "Congenital Word Blindness"[1]. The distinction between phonological and surface types of Dyslexia is only descriptive, and devoid of any etiological assumption as to the underlying brain mechanisms[2]. However, studies have alluded to potential differences due to variation in performance[3]. Systematic approach detecting dyslexia using artificial neural network is applied in order to segregate the stage of disability in person for better understanding[4].
2.Methodology
In the presence study image segmentation technique based on edge detection algorithm are examined to extract the boundary of left
temporal lobe in the brain. The feature are extracted and used as inputs to the artificial neural network for the classification. Wavelet algorithm is applied to know the classification during contraction and relaxation. This will provide a faster solution and effectively for classification of normal and abnormal dyslexia where it reduce the burden of the conventional way of manual observation through images.
In conclusion these methodology provide a reliable to detect dyslexia and convey used effectively and secondary observer in clinical decision making to detect the damage left temporal lobe in the brain with its inner and outer wall.
The purpose of this study is to investigate the inter observer variability of manual and also of computer software measurement of inner and outer wall. To investigate whether inter observer errors measurement of inner and outer wall in left temporal lobe in the brain can be detected. To find out the variability using image processing and classification with neural networks provides proper diagnosis for the patient with different stages. To conclude that an automatic procedure can reply the manual procedure and leads to an improved performance.

International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 1670

ISSN 2229-5518

III. Results and Discussions
64 slice CT Scan image is taken as an input in oder to diagnose whether the image of the person has dyslexia and if so the stages of the disability is classified based on the study applied using image processing and classification with neural networks for proper diagnosis.

Fig 1. CT Scan input image

Fig 2. Dyslexia symptom
Age and Disability of the children are taken and the same were used for classification to know about the stage of children affects with autusm and based on the training and testing the result has been evaluated using confusion matrix. The data collected is based on the child age group of the children who treated as special child. The original dataset of the children are gathered and evaluated.

classification of normal and dyslexia

data 1

Figure represent normal brain and dyslexic brain

Classification with Neural Networks
The segmented image is given as an input in Radial Basis Function classifiers, where three tier layers helps to classify the input, output and hidden layer information and displays the output with single data.
The Fig 1. CT Scan Input images of the autusm children is taken for processing and the features to be extracted based on Name,

International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 1671

ISSN 2229-5518

Acknowledgement
We wish to thank the center for special child at Mogappair, Chennai for providing the datas and information to proceed this research in a successful manner.
In future, the research can be implemented to find the way such that without surgery any type of clinical type of technology can be applied to the special child to increase their attentation and reduce mental retardation.
References
[1] Breiman, L. M. (1996). Some properties of splitting criteria. Machine Learning,
24(1), 41–47. Coskunoglu, O., Hansotia, B.,
&Muzaffar, S. (1996). A New Logit model
for decision making and its applications. The Journal of the Operations Research Society, 36(1), 35–41.
[2] Breiman, L., Friedman, J. H., Olshen, R.
A., & Stone, C. J. (1984). Classification and regression trees. Monterey, CA: Wadsworth
& Brooks/Cole Advanced Books & Software.
[3] Cheng, B., &Titterington, D. M. (1994). Titterington Neural networks: A review
from a statistical perspective. Statistical
Science, 9(1), 2–54.
[4] Coskunoglu, O., Hansotia, B.,
&Muzaffar, S. (1985). A New Logit
modelfor decision making and its applications. The Journal of the Operations Research Society, 36(1), 35–41.
4. Conclusion
Our proposed system helps to identify the
different stages of dyslexia and based on the stages the treatment may be given to the special child. This research helpful in medical field where such research is not much implemented so far. This system is also referred as second practitioner to the clinical triats.