![REFERENCES
[1] C.-W. Tsai, R.-T. Tsai, S.-P. Liu, C.-S. Chen, M.-C. Tsai, S.-H. Chien, H.-S.
Hung, S.-Z. Lin, W.-C. Shyu, and R.-H. Fu, “Neuroprotective effects of betulin
in pharmacological and transgenic caenorhabditis elegans models of Autism
disease,” Cell Transplantation, vol. 26, no. 12, pp. 1903–1918, 2017.
[2] D. Frosini, M. Cosottini, D. Volterrani, and R. Ceravolo, “Neuroimaging in
Autism’s disease: Focus on substantia nigra and nigro-striatal projection,”
Current Opinion in Neurology, vol. 30, no. 4, pp. 416–426, 2017.
[3] K. Marek, D. Jennings, S. Lasch, A. Siderowf, C. Tanner, T. Simuni, C.
Coffey, K. Kieburtz, E. Flagg, S. Chowdhury et al., “The Autism progression
marker initiative (ppmi),” Progress in Neurobiology, vol. 95, no. 4, pp. 629–
635, 2017.
[4] J. Shi, Z. Xue, Y. Dai, B. Peng, Y. Dong, Q. Zhang, and Y. Zhang,
“Cascaded multi-column rvfl+ classifier for single-modal neuroimagingbased
diagnosis of Autism’s disease,” IEEE Transactions on Biomedical Engineering,
2018.
15](https://image.slidesharecdn.com/projectppt-230720095256-a87f78e2/85/Project-PPT-pptx-15-320.jpg)
Project PPT.pptx
- 1. OBJECTIVE The focus of this project is to find out the most significant traits and automate the diagnosis process using available classification techniques for improved diagnosis purpose. 1
- 2. Autism Spectrum Disorder (ASD), which is a neuro development disorder, is often accompanied by sensory issues such an over sensitivity or under sensitivity to sounds and smells or touch. Although its main cause is genetics in nature, early detection and treatment can help to improve the conditions. The project gives a new approach for identification of ASD using a deep classifier.Deep Learning (DL) has become a common technique for the early diagnosis of AD. Here, we introduce prediction on Autism Disorder using Recurrent Neural Network help researchers diagnose the disease at its early stages. The ASD identification works in the following steps. Feature analysis explains ASD traits thereby improving the efficiency of screening process. Further, Machine Learning (ML) classifier models report ASD class type with evaluation parameters. With this the accuracy and performance of the model is increased. ABSTRACT
- 3. INTRODUCTION • Autism spectrum disorder (ASD), is a neurological developmental disorder. It affects how people communicate and interact with others, as well as how they behave and learn. • The symptoms and signs appear when a child is very young. • It is a lifelong condition and cannot be completely cured. • A study found that 33% of children with difficulties other than ASD have some ASD symptoms while not meeting the full classification criteria. • ASD has a significant economic impact both due to the increase in the number of ASD cases worldwide, and the time and costs involved in diagnosing a patient. 3
- 4. LITERATURE SURVEY 4 TITLE AUTHOR and Year Pros Cons DATASET METRICES A Survey of Machine Learning Based Approaches for Autism Disease Prediction Shubham Bind1 2021 Used Realtime MRI Images Less Accuracy MRI Dataset 75 Classification of Autism’s disease and its stages using machine learning John Michael Templeton 2022 active MRI patches are extracted The training database consist less image MRI2D 65 Analysis and Prediction of Autism's Disease using Machine Learning Algorithms Sohom Sen 2022 Used 30 MRI Images The system was trained with Less type of Disease. BRAINMRI 84 Detection of Autism Disease Using Machine Learning G.Priyadharshini, T.Gowtham 2020 High Accuracy lowest accuracy with 40%. MRIDATA 90 Predicting Severity Of Autism's Disease Using Deep Learning Srishti Grover 2020 Dataset belonging to Multiple category Accuracy Low database with Disease by Name 65
- 5. Existing system • Several studies have made use of machine learning in various ways to improve and speed up the diagnosis of ASD. • Soft computing techniques such as probabilistic reasoning, artificial neural networks (ANN), and classifier combination have also been used . • Artificial Neural Network was used to classify between patients with and without ASD, respectively which gave lesser accuracy.
- 6. PROBLEM DEFINITION • The system gives comparatively lower results which is approximately around 93%. • Artificial Neural Network leads to loss of data and consumes high amount of data which leads to lesser accuracy. • Moreover, it was also observed that different areas in a scan or an image had different textural and morphological patterns which were difficult for human beings to track. 6
- 7. PROPOSED SYSTEM • The combination of Data-driven approach and Rule-based approach is employed for the development of the model. • We begin by preprocessing the dataset to eliminate missing values and outliers, remove noise, and encode categorical attributes. • We also employ feature engineering to choose the most beneficial features out of all the features present in the data set. • This reduces data dimensionality to improve speed and efficiency during training. Once the data set has been preprocessed, classification algorithm used was Recurrent Neural Network. • RNN predict the output label (ASD or no ASD). The accuracy of each classifier is observed and compared. 7
- 8. PROPOSED SYSTEM • Furthermore, metrics like the F1 score and precision-recall values have also been computed for better evaluation of each classifier. • If the classifier performs well, then the training accuracy will be higher than its test accuracy. • This model can then be deemed to be the best model and hence be used for further training and classification.
- 10. CLASSIFIER • A Recurrent Neural Network (RNN) is a Deep Learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other. • The pre-processing required in a RNN is much lower as compared to other classification algorithms. • While in primitive methods filters are hand-engineered, with enough training, RNN have the ability to learn these filters/characteristics. 10
- 11. DATASET • Abstract: Autism EEG Disease Detection Dataset • Data Set Characteristics: Multivariate • Number of Instances: 197 • Area: Life • Attribute Characteristics: Real • Number of Attributes: 23 • Associated Tasks: Classification 11
- 12. ADVANTAGES • They recognized faster and more accurately . • The model is trained to learn the low level to high level features and the classification results are validated. • This system consumes lesser memory so that the processing time is reduced. • The data loss is considerably lower. • Thus with a rapid growth in the Machine learning architectures, an objective diagnosis of Autism’s disease will no longer be a laborious job for the clinicians in the near future. 12
- 13. SYSTEM REQUIREMENTS Software Description Version Anaconda Software used for computer vision with Machine Learning 4.2.0 Language - Python Programming language used to program the Project 3.7
- 14. CONCLUSION • In this project, we have analyzed the ASD datasets of toddler, child, adolescent and adult. We apply most popular five feature selection methods to derive fewer features from ASD datasets yet maintaining competitive performance. • We find that EEG feature selection method outperforms amongst others. • In our experimental setup, we increase the attribute numbers gradually and then apply different classification techniques. • We find that MLP outperforms amongst all other classifiers using our methodology and approach.
- 15. REFERENCES [1] C.-W. Tsai, R.-T. Tsai, S.-P. Liu, C.-S. Chen, M.-C. Tsai, S.-H. Chien, H.-S. Hung, S.-Z. Lin, W.-C. Shyu, and R.-H. Fu, “Neuroprotective effects of betulin in pharmacological and transgenic caenorhabditis elegans models of Autism disease,” Cell Transplantation, vol. 26, no. 12, pp. 1903–1918, 2017. [2] D. Frosini, M. Cosottini, D. Volterrani, and R. Ceravolo, “Neuroimaging in Autism’s disease: Focus on substantia nigra and nigro-striatal projection,” Current Opinion in Neurology, vol. 30, no. 4, pp. 416–426, 2017. [3] K. Marek, D. Jennings, S. Lasch, A. Siderowf, C. Tanner, T. Simuni, C. Coffey, K. Kieburtz, E. Flagg, S. Chowdhury et al., “The Autism progression marker initiative (ppmi),” Progress in Neurobiology, vol. 95, no. 4, pp. 629– 635, 2017. [4] J. Shi, Z. Xue, Y. Dai, B. Peng, Y. Dong, Q. Zhang, and Y. Zhang, “Cascaded multi-column rvfl+ classifier for single-modal neuroimagingbased diagnosis of Autism’s disease,” IEEE Transactions on Biomedical Engineering, 2018. 15