Stanford researchers develop an algorithm that outperforms radiologists on diagnosing pneumonia; Researchers at University of Southern California develop a first ever brain implant that can effectively improve the human memory
Stockholm-based Natural Cycles raises $30M by developing the first CE-certified contraceptive app in Europe
A group of Stanford researchers led by Andrew Ng, the former Chief Scientist of Baidu, develop an algorithm that outperforms radiologists on diagnosing pneumonia; The algorithm, CheXNet, is a convolutional neural network trained on the largest publicly available chest X-ray dataset; containing over 100,000 frontalview X-ray images with 14 diseases.
In this study, 420 of the images for possible indications of pneumonia are annotated independently by four Stanford radiologists; And in just over a month, the algorithm could beat the best published results in all 14 identification tasks; The research result is published on arXiv. (Stanford News)
Researchers at University of Southern California develop a brain implant that can improve short-term memory by roughly 15 percent and working memory by 25 percent; This is the first device ever to effectively improve the human memory according to a New Scientist report.
The device is tested on 20 volunteers having brain electrodes implanted previously for the treatment of epilepsy; it works by determining the pattern associated with optimal memory performance from their brain activity, and then stimulate the brain with small electric currents following that pattern in later tests to improve the memory; The device could have great potentials to treat those affected by dementia and Alzheimer’s says the researcher. (Futurism, BigThink)
Stockholm-based Natural Cycles, the first CE-certified contraceptive app in Europe, raises $30 million led by EQT Ventures; Raises $38 million to date; The startup is borrowing ideas from high energy physics to develop algorithms that can be used to predict the risk of pregnancy.
The app is sold on a yearly subscription basis along with a basal thermometer; Currently has over 500,000 users across around 160 countries; New funding will be used for international expansion and algorithm development for detecting early signs of polycystic ovary syndrome, endometriosis, and infertility. (MobiHealthNews)
