FINAL REVIEW_AD of a college project.pptx

Some people live productively for years in the early stage of Alzheimer’s disease. Others experience rapid changes and need assistance with everyday tasks relatively soon. Early intervention is critical to provide help and support as soon as possible.

There is no cure or effective long-term treatment for Alzheimer’s disease. Some medical treatments may help manage symptoms for a while, but they do not slow how the disease changes over time. In general, treatments are more effective when they are started early. That’s one good reason to see a doctor as soon as you notice symptoms that are not typical signs of aging. Some conditions that cause dementia can be reversed, even though Alzheimer’s disease cannot.

1. COGNITIVE CARE: EARLY INTERVENTION FOR ALZHEIMER’S DISEASE Presented by Thamaraikannan B – 822720104045 Robin V – 822720104031 Vengadesh R – 822720104047 Aswin B K - 822720104008 Guided by Dr. K. Santhosh Kumar M.E., Ph.D., Assistant Professor, Department of Computer Science and Engineering, GCE Thanjavur

2. Alzheimer's disease is a degenerative brain illness, incurable and progressive Alzheimer’s disease(AD)  is the most common cause of dementia among older adults It usually affects people older than 65 Identifying AD in early stages is so difficult as symptoms overlap with normal aging process By using deep learning models like Xception to analyze medical imaging data, it may be able to identify early signs of Alzheimer's disease before symptoms become severe. OBJECTIVE

3. EXISTING WORK Aversen et al. [2] to analyze structural MRI data. They have used both SVM binary classifier and multi-class classifier to detect AD MRI images using Alzheimers Disease Neuroimaging Initiative (ADNI) database HosseiniAs et al. [3] adapted a 3D CNN model for AD diagnostics Liu et al. [ 4 ] developed a deep learning model using a subset of ADNI dataset and classified AD and MCI patients Gupta et al. [5] have developed a sparse autoencoder model for AD, Mild Cognitive Impairment (MCI) and healthy control (HC ) using ADNI dataset Morra et al. [6] compared several model’s performances for AD detection including hierarchical AdaBoost , SVM with manual feature

4. PROPOSED WORK In this project we’re using Xception model in CNN as a feature extractor CNN(CONVOLUTIONAL NEURAL NETWORK): A CNN is a kind of network architecture for deep learning algorithms and is specifically used for image recognition and tasks that involve the processing of pixel data It automatically detects the important features without any human supervision Its built-in convolutional layers reduces the high dimensionality of without losing its information Minimize computation in comparison with a regular neural network

5. XCEPTION MODEL(Extreme Inception): Xception is a CNN model which is 71 layers deep. It can classify images into 1000 object categories Pre-trained Xception model achieved accuracy almost 10% more than the VGG16, Inception-V3 and Resnet This deep learning algorithm assists radiologists in making more accurate diagnoses and treatment decisions Prediction in initial stage helps the patients to get proper treatment and guidelines With the help of this website AD can be easily identified in early stages hence patients can be treated before its too late

6. Pre-processor RAM Hard disk Keyboard Monitor HARDWARE REQUIREMENT : Intel icore7 : 8GB : 60GB : Standard Keyboard : 15 inch colour Monitor Above hardware requirements are essential for seamless project execution

7. JUPYTER NOTEBOOK It’s coding platform where we can process the image, train and test the model HTML, CSS, JAVASCRIPT For creating the webpages where user can upload their MRI scan FLASK It is web framework used for integrating model and webpages SOFTWARE REQUIREMENT

8. Data collection Data pre-processing Model Building Application Building Result Prediction MODULES

9. DATA COLLECTION Alzheimer’s disease classified into four stages i ) Non demented ii) very mild dementia iii) mild dementia iv) moderate dementia. The datasets collected and separated into test and train set Example of different brain MRI images presenting different AD stage. (a) Non demented ; (b) very mild dementia ; (c) mild dementia; (d) moderate dementia.

10. DATA PRE-PROCESSING In this module we will be improving the image data that suppresses unwilling distortions or enhances some image features important for further processing, although performing some geometric transformations of images like rotation, scaling, translation, etc . MODEL BUILDING Pre-trained CNN model as a feature extracter In this phase for one of the models, we will use it as a simple feature extractor by freezing all the five convolution blocks to make sure their weights don’t get updated after each epoch as we train our own model. Here, we have considered images of dimension (180,180,3).

11. Create sequential layers For this purpose we have imported Sequential from tensorflow.keras.models . As the name suggests it is used to arrange the Keras layers in  a sequential manner. It operates the mean on the height and width dimensionalities for all the channels . Configure learning process The compilation is the final step in creating a model. Once the compilation is done, we can move on to the training phase. The loss function is used to find errors or deviations in the learning process.

12. Train the model Train the model up to 16 epochs Accuracy after training the model Training accuracy - 84.1% Validation accuracy - 82.4% Save the model Save the model in h5 extension Test the model Test the model with any input from test set

13. APPLICATION BUILDING WEBPAGES Home page

14. User can upload their MRI scan images here Image upload page

15. User can view their predicted result in this page Result page RESULT

16. REFERENCES [1] Ali , E.M., Seddik , A.F., Haggag , M.H.: Automatic detection and classification of alzheimer’s disease from mri using tannn . International Journal of Computer Applications 148(9) (2016) [2] Arvesen , E.: Automatic Classification of Alzheimers Disease from Structural MRI. Master’s thesis (2015 ) [3] Hosseini-Asl , E., Keynton , R., El- Baz , A.: Alzheimer’s disease diagnostics by adaptation of 3d convolutional network. In: Image Processing (ICIP), 2016 IEEE International Conference on. pp. 126–130. IEEE (2016 )

17. [4] Liu, S., Liu, S., Cai , W., Che , H., Pujol , S., Kikinis , R., Feng , D., Fulham, M.J.: Multimodal neuroimaging feature learning for multiclass diagnosis of alzheimer’s disease. IEEE Transactions on Biomedical Engineering 62(4), 1132–1140 (2015) [5] Gupta, A., Ayhan , M., Maida, A.: Natural image bases to represent neuroimaging data. In: ICML (3). pp. 987–994 (2013) [6] Morra , J.H., Tu , Z., Apostolova , L.G., Green, A.E., Toga, A.W., Thompson, P.M.: Comparison of adaboost and support vector machines for detecting alzheimers disease through automated hippocampal segmentation. IEEE transactions on medical imaging 29(1), 30 (2010) REFERENCES

18. Thank you