Last year, more than 1 million Americans attempted suicide, and 47,000 succeeded. While some people display warning signs, many others do not, which makes suicide difficult to predict and leaves family members shocked — and anguished that they couldn’t do something.

Medical providers and tech companies, including the Department of Veterans Affairs and Facebook, are increasingly applying artificial intelligence to the problem of suicide prediction. Machine learning software, which excels at pattern recognition, can mine health records and online posts for words and behaviors linked to suicide and alert physicians or others to impending attempts. The potential upside of this effort is huge, because even small increases in predictive accuracy could save thousands of lives each year.

Do you think you own your own medical data? Your hospital and doctor records, lab and radiology tests, genetic information, even the actual tissue removed during a biopsy or other surgical procedure? Well, you don’t.

It’s a good bet that the fine print of the consent form you signed before your latest test or operation said that all the data or tissue samples belong to the doctor or institution performing it. They can study it, sell it or do whatever they want with it, without notifying or compensating you, although the data must be depersonalized in their best effort to make sure you are anonymous.

But as ubiquitous as the phenomenon is, and as plentiful the studies that demonstrate it, the placebo effect has yet to become part of the doctor’s standard armamentarium — and not only because it has a reputation as “fake medicine” doled out by the unscrupulous to the credulous. It also has, so far, resisted a full understanding, its mechanisms shrouded in mystery. Without a clear knowledge of how it works, doctors can’t know when to deploy it, or how.

Here’s the thing: Even if the Apple Watch excels at detecting undiagnosed AF (a big if), using it to screen large numbers of asymptomatic people isn’t necessarily a good idea.

The executive team at Moderna raised a cheer today after publishing their first early snapshot of human efficacy data that demonstrate their messenger RNA tech works — at least on the first try. The biotech tested their H10N8 flu vaccine on a small group of 31 subjects, looking at their response in two different measures. All demonstrated a sufficient immune response to fight off the virus in the first measure, and all but 3 in the second, for a total of 23 who received the vaccine.  None of the 8 subjects who received a placebo responded.

Now, 22 teams of computer scientists have unveiled a set of algorithms able to predict the odor of different molecules based on their chemical structure. It remains to be seen how broadly useful such programs will be, but one hope is that such algorithms may help fragrancemakers and food producers design new odorants with precisely tailored scents.

The departments and agencies listed in this document announce revisions to modernize, strengthen, and make more effective the Federal Policy for the Protection of Human Subjects that was originally promulgated as a Common Rule in 1991. This final rule is intended to better protect human subjects involved in research, while facilitating valuable research and reducing burden, delay, and ambiguity for investigators.

By the time 8-year-old Ava Christianson got to the National Institutes of Health this summer, she had lost several grueling rounds to leukemia and was bracing for the next one.

The European Union is beefing up protections for volunteers in phase I clinical trials in the wake of a disastrous clinical study in Rennes, France, that resulted in the death of one volunteer and the hospitalization of five others. On 21 July, the European Medicines Agency (EMA) in London announced in a “concept paper” that it wants to improve strategies to identify and reduce risks in “first-in-human” (FIH) studies on healthy volunteers. EMA is asking for input from stakeholders.

President Barack Obama’s ambitious 1-million-person personalized medicine study began to take shape this week with the announcement of four medical centers that will recruit volunteers starting this fall. A fifth center aims to sign up 350,000 participants by blasting the general public with ads coming soon to your web browser or mobile phone.