The top three news stories of the week, as chosen by our resident students. This week’s top three news stories are A.I. doctors, time crystals, and developments in dementia.

Article written by Fuad Mosis.

Artificial Intelligence as a diagnostic tool for cancer

In 2011, IBM’s computer system, Watson, defeated game show champions of the popular game show Jeopardy! Just last year, Korean Go champion Lee Sedol was defeated by Google’s DeepMindsoftware. Artificial intelligence is developing at a terrifying rate, so much so that in 2015 an open letter was signed by the likes of Elon Musk and Stephen Hawking amongst many others to prevent the weaponisation of AI. Needless to say, AI is in invaluable tool in many global disciplines, with medical diagnostics scoring very highly. Deep learning is one area of machine learning where software mimics how we think by recognising patterns in images and sound. This tool has been employed by researchers from Stanford University to identify skin cancer and was able to match dermatologists in its performance of diagnosing skin lesions. This was achieved by training a neural network with a dataset of 129,450 images of 2,032 diseases. The algorithm was then trained to make a visual diagnosis based on what it had ‘learned’. The researchers suggest that this diagnostic tool could be fitted on smartphones and could potentially provide quick and universal screening.

picture1
Image credit: Matt Young

Time crystals are a new form of matter

The idea of a time crystal sounds like something out of the Marvel universe, but it really does exist – as we’ve recently found out! In the field of quantum physics, a ground state is a point in which a material is at its lowest energy state. The movement of materials in this stage is theoretically impossible. One feature is that they have a repeating atomic structure that you can imagine as a fixed lattice, and this exists in space. For now, let’s call this a ‘normal crystal’. With a time crystal, the same repeating patterns exist on the atomic level, the difference being that these also repeat in time as well as space. The key thing here is that this allows it to oscillate in its ground state – in other words, it’s projecting motion without any energy. A method to make these crystals has been recently published and two independent teams have confirmed they were able to successfully do just this. These results have effectively demonstrated the existence of a new form of matter and make a fascinating new avenue of research that will undoubtedly continue to excite us with novel breakthroughs in the coming years. Click here for a fantastic report on the paper.

picture2
Image credit: Chris Monroe.

Alzheimer’s-linked pathologies reversed in mice through genetic modification

Alzheimer’s disease is the most common form of dementia, making up 62% of all cases. In the United States alone, there are 5 million individuals living with the disease, and is also the 6th leading cause of death. Alzheimer’s is a disease which continues to elude us, but ground-breaking research continues every day to try and tackle its devastating effects. Neurofibrillary tangles (NFTs) act as a histopathologic marker for Alzheimer’s disease. These are aggregates of the tau protein, which in its pathological form generates toxic oligomers. In a new study, researchers used antisense oligonucleotides to target and reduce the expression of human tau mRNA, which the mice have been genetically predisposed to express. Not only did they observe a reduction in the development of further tau aggregates, but their data also suggests that this destroyed pre-existing oligomeric tau. Compared to controls, mice which underwent this treatment also had greater survival statistics and a reduction in neuronal death. Before testing in humans can commence, the safety of the treatment needs to be further assessed as we do not currently understand the full function of tau. It may also be possible, however, to target other neuropathology’s in which tau is implicated if such a treatment is safe for human use.

picture3Image credit: DeVos et al., Science Translational Medicine 2017.

Article written by Fuad Mosis.