Medopad go to China
The UK-based startup Medopad recently closed $143 million of funding in China for its remote patient monitoring apps and wearables, and advanced patient analytics.
Among the companies investing in Medopad are Chinese tech giant Tencent, Ping An, GSK China and Lenovo.
Medopad joined British Prime Minister Theresa May and 50 other companies on a business delegation to China this February.
Dan Vahdat, Medopad CEO, said China’s large population provides a huge data set to train with that can lead to smarter predictive capabilities.
The company will access one million patients in a study with Capital Medical University in Beijing, where Medopad will track the walking gaits of patients with Parkinson’s.
Disrupting health insurance
Amazon have partnered with Berkshire Hathaway and JPMorgan Chase to create a not-for-profit health-care company for their own employees.
In a typically disruptive move, Amazon will use its own technology to provide cheaper care than conventional health insurers for one million employees in the first instance.
There are few details so far, but most speculate Amazon will train their algorithms on huge quantities of patient data and build patient monitoring tools.
Amazon’s new partnerships will give the tech giant access to more data, leading to continuous improvements in its system and the potential for a monopoly.
Two of America’s largest insurance firms, UnitedHealth Group and CVS Health, saw shares fall by 4% at the news of Amazon’s plans.
Singled out for recognition
YITU Tech, a Chinese startup, impressed at the annual 2017 Radiology Society of North America (RSNA) meeting in Chicago by revealing tools for assisted diagnosis of chest CT, bone age, pediatric clinic, ultrasonic, and a clinical research platform.
While it initially focused on facial recognition, YITU has branched out into more complex image recognition challenges, like cancer scans.
For their AI products, covering both images and texts simultaneously, YITU received tens of millions of dollars from Yunfeng Capital, the personal fund of Alibaba founder Jack Ma.
Lin Chenxi, the YITU co-founder who left Alibaba to establish the company in 2012, said he hopes the technology can ensure equal access to medical treatment across China.
Google predicts hospital visits
The tech giant is developing AI that could predict the outcome of hospital visits at the moment of admission.
The algorithms Google has developed, based on neural networks, could show the probability that a patient will be discharged, readmitted, or die. They say their mortality predictor is 24-48 hours more accurate than current methods.
Google trained its neural networks on anonymized medical information from 216,221 adult patients, combining more than 46 billion units of individual data sets, including clinical notes.
Newborn AI in Africa
Last year, a Nigerian startup exhibited at the first United Nations AI for Good Summit and was selected as one of 141 international teams to compete in the IBM AI XPrize worth $5 million.
Ubenwa, which means baby’s cry, has developed a machine learning system to detect child birth asphyxia by analysing the amplitude and frequency patterns of an infant’s wail.
The founders say the AI solution, currently deployed as an embedded model on an Android app, achieved over 95% prediction accuracy in trials with nearly 1,400 pre-recorded baby cries.
Aggregator for clinical trials
Patients will now find it easier to find and sign up for clinical trials thanks to a company democratising the process.
Antidote was founded by Pablo Gravier, former VP at Kayak, to simplify clinical trials as companies like Kayak simplified booking travel.
The company’s system was trained to identify studies using language and it recognises multiple different ways of describing the same disease, like adult onset diabetes, T2DM and type 2 diabetes.
Through 231 partnerships, including with the American Kidney Fund, Muscular Dystrophy Association, and Lung Cancer Alliance, Antidote says it reaches more than 15 million people per month.