There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | NAGARJUNA REDDY G |
| Institution: | NBKR INSTITUTE OF SCIENCE & TECHNOLOGY, VIDYANAGAR |
| Department: | ECE |
| Country: | |
| Proposed Analysis: | Alzheimer’s disease is a most common cause of dementia. It is a neurodegenerative disease that damage brain’s intellectual functions such as memory, calculations and orientation, etc. by preserving the motor functions. People’s brain with Alzheimer's also have a shortage of some important chemicals in their brain which transmit signals around the brain. Due to the shortage of these chemicals, the signals are not transmitted throughout the brain as effectively. Even though it is generally a disease of later life i.e. after the age of 60, rarely it affect the young persons with age 30 also because of the people habits such as cigarette smoking and alcohol consumption. Genetic reasons and people with a history of head injury may increase the possibly of getting this disease. In the initial stages of Alzheimer's disease short-term memory is affected, and patient face many problems in planning his daily life activities, managing money, learning and recalling new information. Personality changes, anxiety, or depression are also the serious problems in day to day life. Ultimately, older or distant memory is lost slowly, becomes difficult to recover memories of events from earlier life. Next, other indications such as difficulty in putting thoughts into words, difficulty in carrying out simple acts and difficulty in recognizing well-known things may develop. Even in a familiar area, these persons lose their sense of direction and get lost while driving or walking. As Alzheimer's disease progresses to its middle and final stages, the patient may face delusions and hallucinations and become antagonistic or may wander away from home. Around 46.8 million people are living in the world with dementia according to the survey World Alzheimer Report 2015 prepared by King's College London. This report expected this count will be doubled for every 20 years so that it will become 74.7 million by 2030 and 131.5 million people by 2050. Researchers also found that there are more than 9.9 million new cases of dementia each year worldwide, implying one new case every 3.2 seconds. At the country level, ten countries are home to over a million people with dementia in 2015: China (9.5 million), US (4.2 million), India (4.1 million), Japan (3.1 million), Brazil (1.6 million), Germany (1.6 million), Russia (1.3 million), Italy (1.2 million), Indonesia (1.2 million) and France (1.2 million). These new findings are taken into account both the people age, new and better evidence on the number of people living with dementia, and costs incurred for care. These findings show that the cost of dementia has increased by 35 percent in 2015 since the 2010 World Alzheimer Report estimate of USD 604 billion and it is expected that this will become a trillion dollar disease in just three years' time. These views are creating emergency to pay the attention towards future issues of this dementia. Since the prevention is better than cure, this dementia must be diagnosed in earlier stages so that people will undergo through proper remedial care. But only 1-in-4 people with Alzheimer’s disease are being diagnosed due to the complexity in early detection of disease. The current diagnosis is made purely by clinical, neuropsychological and neuroimaging assessments. General structural neuroimaging assessment is based on nonspecific features such as atrophy which is a late feature in the progression of image. Therefore, developing new approaches for early and specific recognition of this dementia at prodromal stages is of crucial importance. So structural MRI, functional MRI and Positron Emission Tomography (PET) images are playing crucial role in the early detection of Alzheimer’s disease. The proposed system has steps to classify the Multimodal images for early detection of Alzheimer’s disease using improved deep learning framework with modified k-sparse auto encoder classification for early detection of Alzheimer’s disease. In first phase image data sets are retrieved from ADNI (Alzheimer’s Disease Neuroimaging Initiative) database and pre-processed mainly for removing noise, normalizing intensity values, improving contrast and extracting brain from skull. Pre-processing helps in separating desired regions more accurately and also improves classification accuracy. After pre-processing, image data sets are divided into homogeneous regions with similar properties called segmentation and features are extracted in feature extraction step. Most commonly seen Pattern Recognition problem in the domain of medical image analysis is finding discriminating features from a set of medical images and classifying them into set of classes. Feature selection, extraction and classification stages in this procedure play a very important role for detection of brain lesions. Features are extracted mainly based on intensity values, texture properties, intensity gradients and edge detection. Important features are selected in the feature selection phase by applying algorithms and based on these selected features. Once the features are selected, the kernel (mask) matrix of optimal size is computed and those kernels are combined to get the combination of kernels. These combined kernels are applied to the inputs of Deep learning neural networks for classification. By choosing the wider data sets instead of small data sets and optimal implementation of this procedure may produce more accuracy. To achieve optimal solution, the selection of optimal kernel in size, combining and implementation of 3D Convolutional Neural Networks play crucial role. Once the Image data sets are classified into Alzheimer’s Disease (AD) and Normal Control (NC) and Mild Cognitive Impairment (MCI), The performance of the proposed approach is analyzed by measuring the accuracy and standard deviation for every data set. |
| Additional Investigators |
