China
Africa
Explore chapters and articles related to this topic
Smart Homes: Practical Guidelines
Published in Nazmul Siddique, Syed Faraz Hasan, Salahuddin Muhammad Salim Zabir, Opportunistic Networking, 2017
Kevin Bouchard, Bruno Bouchard, Abdenour Bouzouane
The guidelines base the selection of prompts accordingly on the problem implied in the current context. AD is well known for the memory impairment it provokes. The consequence is that the person may not remember something from general knowledge (e.g., the correct steps to make a cake) or from events. Aphasia and agnosia are, however, much less known. Aphasia is the difficulty for a person to produce or understand spoken language or written language even if he or she hears and sees correctly through intact organs. Agnosia is the general difficulty to recognize stimuli while still being able to perceive them. AD also often leads to ideational apraxia, which is characterized by a lack of knowledge about the sequence of actions (e.g., substitution of objects, omissions, mistakes in the sequence of actions) needed to accomplish a certain task. Finally, persons afflicted by AD also often suffer from deficits in executive functions and sensory problems that can be the simple consequence of aging.
Cervical Cancer Classification from Pap Smear Images Using Modified Fuzzy C Means, PCA, and KNN
Published in IETE Journal of Research, 2022
N. Lavanya Devi, P. Thirumurugan
Many research works are going on for the efficient and rapid classification of cervical cancer. The paper [5] compared the efficiency of machine learning algorithms like KNN, SVM, and decision tree for the detection of multiple sclerosis and concluded that KNN gives better results by computing accuracy, precision, and specificity. A novel method based on stationary wavelet entropy and KNN was suggested for the detection and gave superior results [6] suggested a novel approach based on Kernel Support Machine for the classification of MR brain images into normal and abnormal images. Seven brain diseases like Pick’s disease, Alzheimer’s disease, meningioma, Huntington’s disease, Alzheimer’s disease plus visual agnosia, glioma, and sarcoma are taken into consideration as abnormal cases. Discrete Wavelet Transform is used for extracting the feature; PCM is applied for reducing the dimension of features.
Classification of SGS-SRAD Denoised MRI Using GWO Optimized SVM
Published in IETE Journal of Research, 2022
Sonal Goyal, Navdeep Yadav, Asha Rani, Vijander Singh
The effectiveness of designed hybrid method, SGS-SRAD + SIFT + PCA + GWO-SVM, is implemented on a real magnetic resonance imaging dataset obtained from “the whole brain atlas” provided by Harvard Medical School [22]. The databases comprise high contrast brain MR-T2 images in axial plane with 256 × 256 image size. In this work, experiments are performed using five different databases. The first dataset (DS-66) consists of 66 brain MR images including 18 healthy and 48 unhealthy images. The second dataset (DS-160) consists of 160 brain MR images including 20 healthy and 140 unhealthy images. The unhealthy images in above two databases consist of seven types of diseases, i.e. glioma, meningioma, Alzheimer's disease, Alzheimer's disease plus visual agnosia, Pick's disease, sarcoma and Huntington's disease. The third dataset (DS-255) consists of 255 brain MR images including 35 healthy and 220 unhealthy images. This database contains four additional types of disease in addition to the previous seven diseases, i.e. chronic subdural hematoma, cerebral toxoplasmosis, herpes encephalitis and multiple sclerosis. It is observed that the dataset contains more healthy images than unhealthy images leading to unbalanced dataset. Therefore analysis is also carried out for two subsets of these datasets having equal number of healthy and unhealthy images. First subset contains images with 11 diseases (DS-11) whereas second comprises unhealthy images with Alzheimer's disease only (DS-1). Figure 2 shows some examples of healthy and unhealthy brain MR-T2 images used in the classification analysis.