<pre><FONT COLOR = BLACK>
<B>Alzheimer's Disease diagnosis using MultiModal biomarker's
 </B>
   Lead Investigator:    Yubraj Gupta
   Institution      :    Chosun university
   E-Mail           :    gshilpa416@gmail.com
   Proposal ID      :    1125
</FONT>
<BR>
   Proposal Description:
   <textarea name="description" rows="12" cols="85" wrap="virtual" readonly>Our project is to develop a CAD system for early diagnosis of AD patients. For, our project we are using ADNI dataset which offers CSF, APOE, SMRI and FDG-PET modality biomarkers, we ran our model and it shows the decent result on it.        Now, we are planning to run our model on your dataset if you give access to us. so, that to compare our result on both datasets.    So, please can you provide access to your dataset to fulfil the project requirement.    This is a machine learning project. Where, we are planning to use one or two different types of image (sMRI or PET or both) for early prediction of Alzheimer's disease. As, we already mention our research title ''Alzheimer's disease diagnosis using different types of Biomarkers". For this experiment, we need four different types of data (CSF value, APOE value, sMRI and PET scans of each subjects). As, recent research shows that combining these four different data can provide good accuracy for early detection of AD, and there are some papers which has used ADNI dataset to implement it. So, our aim is to combine different specific types of biomarker which helps to predict the Alzheimer's disease. Recently, we try our idea with ADNI dataset and get good result. In first step, we download the dataset from the ADNI website. In ADNI cloud, we didn't find enough data which contain four biomarkers of each subject (CSF, APOE, sMRI and PET scans) but there are somes like for (AD(30-40), MCI(40-50), LMCI(30-40) and for HC(30-40)). After downloading the dataset from ADNI website, we extract cortical and sub-cortical features for each image by applying some manual extraction technique. And at last, we combine all features and send it to the classifer to predict the output.      As we already implemented our experiment on ADNI dataset. So, now we are trying to find new dataset which consist of sMRI and PET scans of same subjects with their APOE and CSF values for comapring with ADNI dataset result.   We will use NACC dataset </textarea>