A pinboard by
George Ng

Ph.D in Biotechnology who has recently joined an eCommerce startup in Hong Kong


Do our memories, which fade as we age define who we are? Do we not exist if our memories fail us?

How Exactly Does Memory Work? The function for short term memory storage came to light from the case of Henry Molaison in the 1950s, during epilepsy surgry both his hippocampus were damaged and was unable to make new memories but retained his old ones before the operation (read more). A recent study has found that 2 duplicate memories are created simultaneously and are stored in both the hippocampus - for short term and the cortex - for long-term retrival. (read more)

I Am Because I Loved Alzheimer's disease is an irreversible brain disorder that slowly and progressively wipes away memory and cognitive skills. It is caused by faulty proteins amyloid and tau that build up in the brain. In the movie "Eternal Sunshine of the Spotless Mind" Jim Carrey and Kate Winslet play lovers who after a terrible fight (read more) hires Lacuna, a firm that specialises in wiping away all references of one's memory for a particular episode in life. The movie had a happy ending, meeting unwittingly 2 years after the procedure only to fall in love again.

Helping Family Members With Dementia Caring for family members with dementia is heartbreaking. Joey Daley whose mother Molly, breaks down in front of the camera when she struggles to remember who he is - he documents his life with his mother on a daily basis (read more), his pain is evident. David Baddiel uses humour to manage his pain by creating a reality tv show 'My Family:Not the Sitcom', an exploration of his father - diagnosed with picks disease, a form of dementia (read more).

Staying Sharp Researchers at Massachusetts General Hospital, scanned brain areas of a unique group of adults in their 60s and 70s with minds as sharp as people n their 20s, in that are related to memory such as the anterior insula and orbitofrontal cortex and the hippocampus using Magnetic Resonance Imaging (MRI). They called super agers whose scanned areas were thicker than their younger counterpartrs. Factors that prevent dementia include having a good nights sleep, not smoking and overall low cholesterol (read more)


Hierarchical feature representation and multimodal fusion with deep learning for AD/MCI diagnosis.

Abstract: For the last decade, it has been shown that neuroimaging can be a potential tool for the diagnosis of Alzheimer's Disease (AD) and its prodromal stage, Mild Cognitive Impairment (MCI), and also fusion of different modalities can further provide the complementary information to enhance diagnostic accuracy. Here, we focus on the problems of both feature representation and fusion of multimodal information from Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). To our best knowledge, the previous methods in the literature mostly used hand-crafted features such as cortical thickness, gray matter densities from MRI, or voxel intensities from PET, and then combined these multimodal features by simply concatenating into a long vector or transforming into a higher-dimensional kernel space. In this paper, we propose a novel method for a high-level latent and shared feature representation from neuroimaging modalities via deep learning. Specifically, we use Deep Boltzmann Machine (DBM)(2), a deep network with a restricted Boltzmann machine as a building block, to find a latent hierarchical feature representation from a 3D patch, and then devise a systematic method for a joint feature representation from the paired patches of MRI and PET with a multimodal DBM. To validate the effectiveness of the proposed method, we performed experiments on ADNI dataset and compared with the state-of-the-art methods. In three binary classification problems of AD vs. healthy Normal Control (NC), MCI vs. NC, and MCI converter vs. MCI non-converter, we obtained the maximal accuracies of 95.35%, 85.67%, and 74.58%, respectively, outperforming the competing methods. By visual inspection of the trained model, we observed that the proposed method could hierarchically discover the complex latent patterns inherent in both MRI and PET.

Pub.: 22 Jul '14, Pinned: 17 Apr '17