Here’s the latest in health and biotech analytics:
Googleâs DeepMind and NHS restart data-sharing deal with greater transparency: Google has announced an agreement with the UKâs National Health Service to develop a mobile app, Streams, which will provide data-driven insights to doctors. There have been concerns about data sharing with past agreements, and Google has promised to implement âan unprecedented level of data security.â
IBM and Pfizer to Accelerate Immuno-oncology Research with Watson for Drug Discovery: Pfizer is looking for drugs that will modify the immune system to fight cancer. Watson will help them find new drug targets by analyzing millions of scientific journal articles, as well as other primary data sources.Â
Scientists created a mobile game to help detect the early onset of dementia: In âSea Hero Quest,â a collaboration between game designers and scientists studying dementia, players do tasks intended to measure their spatial navigation skills. The project has collected 9,500 years of data, which has helped the researchers measure how age, geography, and gender affect these skills.
Mount Sinai Researchers Demonstrate Ability of Machine-Learning Algorithms in Echocardiographic Interpretation and Diagnosis of HCM:Â Pathological hypertrophic cardiomyopathy (HCM, a thickening of the walls of the heart) is difficult to distinguish from non-pathological cases. The research team trained machine learning algorithms on echocardiograms (heart ultrasounds), which improved on non-automated techniques for diagnosing the condition.
Will Big Data Finally Demystify the Brain and Alzheimerâs?:Â At the USC Neuroimaging and Informatics Institute, researchers use brain imaging data from all NIH-funded studies on brain disease.. To date, they have found physical changes associated with schizophrenia, depression, and Alzheimerâs disease.
A genome every 12 minutes:Â The Harvard/MIT Broad Institute recently hosted the Program in Quantitative Genomics Conference. Speakers highlighted the growth in genomes sequenced at the institute (3 per year in 2006 to 75,000 per year today), insights that can be gained from large amounts of sequences, and the pitfalls of relying too much on genetic data.
One Obstacle to Curing Cancer: Patient Data Isnât Shared:Â For a variety of reasons, data critical for the development of precision therapies are largely fragmented into inaccessible siloes. There are some government and private efforts to make large, publically available datasets; however there are also opportunities to patients themselves to spur the generation and sharing of data. The Harvard Business School Precision Medicine Accelerator wants to spread knowledge of these opportunities.
Why Scientists Are Sequencing Newborns’ Genomes: Robert Green is leading the BabySeq program, which he hopes will lead to better personalized treatments.
Roanoke Memorial mines patient records to spot trouble early: The hospital uses input from lab tests as well as nurseâs observations to calculate a “Rothman score.” This score can be used to detect when a patientâs condition is starting to deteriorate.