Social platform Litmus Health can predict if you are sick sooner

On June 28th, the article "Connected" published an article on the early clinical trial data science platform Litmus Health co-founder Sam Volchenboum said that one day, social media networks may be able to diagnose users. Diseases may be able to alert them before they know they are ill.

The world is becoming a huge clinical trial. Humans generate data streams from different sources every minute. These constant sources of information come from social media, mobile GPS, WiFi locations, search history, pharmacy membership cards, wearables, and more, all of which provide insight into people's health.

Today, Facebook or Google, the world's two largest data platforms and the largest human behavior forecasting engine, is entirely likely to tell people that they may have cancer when they have no doubts. People who complain about night sweats and weight loss on social media may not know that these are symptoms of lymphoma, and they may not know that the joints are stiff and prone to sunburn in the morning or indicate that they have lupus. Robots that crawl social networking posts are likely to notice these clues.

Sharing these insights and predictions can save people's lives and improve their health, but the data platform currently has good reason not to do so. The next question is, is the risk of doing so more than the return?

Thinking experiment

Social media platforms can be praised by the media by helping to predict and even prevent users from committing suicide, but for now, it is unrealistic for those platforms to predict their health before they see a doctor. But that is not a fantasy.

Suppose, Facebook puts out a huge amount of information to identify data sets, such as user location, travel, likes, frequency of postings, opinions, and search habits. Based on this data, researchers can build models that predict physical and emotional states.

For example, a data set containing social media posts from tens of thousands of people may record an entire life course before someone is diagnosed with cancer, depression, or inflammatory bowel disease. With machine learning technology, researchers can use that data to study the language, style, and content of posts posted by patients before and after the disease is diagnosed. Researchers can also design additional models that, after being injected with new user data sets, predict who will likely develop similar disease symptoms.

Such systems do not need to look for obvious symptoms like fever or weight loss. Data that appears to be insignificant and irrelevant—such as buying a drug that avoids nausea or watching a documentary about insomnia—may eventually lead to a predictive set of rules that predict users or have specific medical conditions. The point is that our digital traces leave a lot of clues about our physical and mental health, whether it is a dominant clue or a hidden clue. How we use that data effectively is another matter.

As a clinician, I support the integration of data and the use of vast amounts of information for the benefit of society. One of the reasons I co-founded data science company Litmus Health is to help researchers better collect, organize, and analyze data from clinical trials, and in turn use that data to improve the health of society as a whole. However, major issues involving regulation, ethics, technology, and society need to be handled with caution.

From a regulatory perspective, all companies are responsible for looking after user data as stated in their Terms of Service. Unfortunately, from a Facebook study in 2014, a study from Carnegie Mellon University, etc., the company's terms of service and privacy policy are usually difficult to read, no one will read, users are very It will be easy to sign.

By developing simple, easy-to-understand data policies without applying personal data to inappropriate uses, companies can demonstrate to their users the moral responsibility of “doing no evil”. The ethical framework for big data must consider the identity, privacy, data ownership, and reputation issues of users. For many companies today, providing user data externally without the user's permission to develop predictive models is against their established value system. However, the process of obtaining user licenses may be as cumbersome as the terms of service agreement for users to sign.

 Ambulance Stretcher

Medical Emergency Trolley is Medical Equipment, surgical instruments, medicines, and transporting patients. It can greatly reduce the operational burden of caregivers. In terms of product classification, medical carts are luxurious, medium and ordinary. According to the material of the product, medical carts are made of ABS, stainless steel and plastic sprayed. According to the types of products, medical carts include rescue carts, emergency carts, treatment carts, medical record carts, equipment carts, instrument carts, medicine delivery carts, anesthesia carts, dirt carts, infusion carts, medicine pickup carts, and drop-off carts. , pick up the car and Emergency Ambulance Equipment.

Ambulance Stretcher,Folding Aluminum Ambulance Stretcher,Folding Ambulance Stretcher,Medical Aluminum Ambulance Stretcher

Haloxylon Ammodendron Medical Equipment Co., Ltd. , https://www.ssmedicaldevic.com