Apple’s iPhones are said to help track down depression and cognitive problems

Apple is said to be working with the University of California and the pharmaceutical company Biogen on diagnostic functions for the iPhone. Its sensor technology should allow conclusions to be drawn about mental disorders in the future.

Can conclusions be drawn about health problems based on typing behavior or facial expressions or other behavioral data available via the iPhone sensors? The alleged project partners Apple, the University of California (UCLA) and the drug manufacturer Biogen are said to be Wall Street Journal (WSJ) want to find out. The WSJ got its information from people in the project environment and, according to its own statements, was able to view relevant documents.

Specifically, Apple is to work under the project name “Seabreeze” with UCLA on the diagnosis of depression and with Biogen under the project name “Pi” on the diagnosis of mild cognitive impairments. In the summer, Biogen received FDA approval for a new drug for the treatment of mild cognitive disorders. Solid diagnostics with the iPhone could help to find patients at an early stage of the disease at which drug interventions are still showing a good effect. None of the project partners has confirmed the cooperation so far.

The sensor data collected by Apple’s iPhones, which include mobility, physical activity, sleep patterns, typing behavior and more, play a key role in this. The researchers want to examine these for patterns that are related to the target states. Algorithms would then be developed from this in order to reliably recognize them.

Earlier academic studies had already suggested that there is a chance of such solutions. It was found that people with certain mental illnesses use their digital devices differently than others. To what extent this knowledge can be translated into reliable algorithms is still completely unclear.

A success of Apple research would be of benefit, especially for the early diagnosis of mental illnesses and brain disorders. Because access to specialists who are able to reliably perform such diagnostics is not completely barrier-free, even in Western countries. Apple and its partners hope with their work to create a generally available alternative for the first specialist contacts.

According to information from the WSJ, the project will receive enthusiastic support at Apple. Apple manager Jeff Williams, who is responsible for the health department of the iPhone manufacturer, is said to have spoken enthusiastically in front of his employees about the potential to combat the increasing rates of depression and anxiety as well as other brain disorders.

Since the data to be collected will be very sensitive, Apple should want to rely on algorithms right from the start that can process the data on the users’ devices – i.e. without the involvement of the cloud. It remains to be seen to what extent Apple will have to worry about the loss of trust after the at least communicative disaster surrounding the child pornography scan on customer iPhones.

Apple has worked with researchers in the past to develop health functions. For example, the detection of atrial fibrillation using the Watch’s ECG function was developed in cooperation with Stanford University.

For depression research, the University of California relies on data from the video camera, keyboard and audio sensors on the iPhone, as well as data on movement, vital parameters and sleep. Among other things, the typing speed, the frequency of typing errors and the content of what is written should be measured. In each individual data element, the researchers are looking for clues about the emotions, concentration, energy level, state of mind and much more of the device users. At the same time, physical data are also collected, such as the amount of the stress hormone cortisol in the hair follicles of the participants.

In a pilot phase that started last autumn, the researchers are collecting this data from 150 people using Apple Watch and iPhone. The study is to be broadened shortly. Then the data from 3,000 study participants will be collected and evaluated.