How ‘Digital mask’ protects patients’ privacy [Details]

Scientists have created a ‘digital mask’ that will allow facial images to be stored in medical records while preventing potentially sensitive personal biometric information from being extracted and shared.

In research published today in Nature Medicine, a team led by scientists from the University of Cambridge and Sun Yat-sen University in Guangzhou, China, used three-dimensional (3D) reconstruction and deep learning algorithms to erase identifiable features from facial images while retaining disease-relevant features needed for diagnosis.

Facial images can be useful for identifying signs of disease. For example, features such as deep forehead wrinkles and wrinkles around the eyes are significantly associated with coronary heart disease, while abnormal changes in eye movement can indicate poor visual function and visual cognitive developmental problems. However, facial images also inevitably record other biometric information about the patient, including their race, sex, age and mood.

Digital masks

Graphic showing digital masking process/Photo:Professor Haotian Lin’s research group

With the increasing digitalisation of medical records comes the risk of data breaches. While most patient data can be anonymised, facial data is more difficult to anonymise while retaining essential information. Common methods, including blurring and cropping identifiable areas, may lose important disease-relevant information, yet even so cannot fully evade face recognition systems.

Due to privacy concerns, people often hesitate to share their medical data for public medical research or electronic health records, hindering the development of digital medical care.

Professor Haotian Lin from Sun Yat-sen University said: “During the COVID-19 pandemic, we had to turn to consultations over the phone or by video link rather than in person. Remote healthcare for eye diseases requires patients to share a large amount of digital facial information. Patients want to know that their potentially sensitive information is secure and that their privacy is protected.”

Professor Lin and colleagues developed a ‘digital mask’, which inputs an original video of a patient’s face and outputs a video based on the use of a deep learning algorithm and 3D reconstruction, while discarding as much of the patient’s personal biometric information as possible – and from which it was not possible to identify the individual.

Deep learning extracts features from different facial parts, while 3D reconstruction automatically digitises the shapes and movement of 3D faces, eyelids, and eyeballs based on the extracted facial features. Converting the digital mask videos back to the original videos is extremely difficult because most of the necessary information is no longer retained in the mask.

Next, the researchers tested how useful the masks were in clinical practice and found that diagnosis using the digital masks was consistent with that carried out using the original videos. This suggests that the reconstruction was precise enough for use in clinical practice.

Compared to the traditional method used to ‘de-identify’ patients – cropping the image – the risk of being identified was significantly lower in the digitally-masked patients. The researchers tested this by showing 12 ophthalmologists digitally-masked or cropped images and asking them to identify the original from five other images. They correctly identified the original from the digitally-masked image in just over a quarter (27%) of cases; for the cropped figure, they were able to do so in the overwhelming majority of cases (91%). This is likely to be an over-estimation, however: in real situations, one would likely have to identify the original image from a much larger set.

The team surveyed randomly selected patients attending clinics to test their attitudes towards digital masks. Over 80% of patients believed the digital mask would alleviate their privacy concerns and they expressed an increased willingness to share their personal information if such a measure was implemented.

Doctor/IANS

Finally, the team confirmed that the digital masks can also evade artificial intelligence-powered facial recognition algorithms.

Professor Patrick Yu-Wai-Man from the University of Cambridge said: “Digital masking offers a pragmatic approach to safeguarding patient privacy while still allowing the information to be useful to clinicians. At the moment, the only options available are crude, but our digital mask is a much more sophisticated tool for anonymising facial images.

“This could make telemedicine – phone and video consultations – much more feasible, making healthcare delivery more efficient. If telemedicine is to be widely adopted, then we need to overcome the barriers and concerns related to privacy protection. Our digital mask is an important step in this direction.”

Computer scientists address gap in messaging privacy

Researchers have developed a solution to a longstanding problem in the field of end-to-end encryption, a technique that ensures that only sender and recipient can read a message.

With current end-to-end encryption, if an attacker compromises a recipient’s device, they can then put themselves in a position to intercept, read and alter all future communications without sender or recipient ever knowing.

The new protocol, published in IEEE Transactions on Information Forensics and Security, forces attackers to leave evidence of any such activity and alerts users to take action.

Dr. Jiangshan Yu at the University of Luxembourg, Professor Mark Ryan at the University of Birmingham and Professor Cas Cremers at the University of Oxford, were motivated by the discovery of mass software vulnerabilities, such as the Heartbleed bug, that make the majority of devices vulnerable to compromise.

Dr Yu explained, “There are excellent end-to-end encryption services out there, but by definition they rely on your device itself remaining secure; once a device has been compromised there’s little we can do. That’s the problem we wanted to solve.”

Following Edward Snowden’s revelations about government mass surveillance, end-to-end encryption is now widely available through services such as Facebook’s WhatsApp. The approach uses pairs of cryptographic ‘keys’ for the sender to encrypt and the recipient to decrypt messages; anyone wanting to read your messages has to first hack into your phone to steal your latest keys. The attacker then performs a ‘Man-in-the-middle’ (MITM) attack, for example by taking control of your WIFI router to intercept your messages, and uses the stolen keys to impersonate you.

Current encryption protocols such as Signal used by WhatsApp make the most of the fact that a MITM attacker can only intercept messages sent via the compromised network. For example, as soon as you send a message via 3G rather than the compromised WiFi the attacker will no longer be able to act as an intermediary. They will lose track of the keys and be locked out of the conversation.

The solution, called DECIM (Detecting Endpoint Compromise in Messaging), addresses the question of what to do when the attacker is in a position to intercept all of your messages on a long-term basis. Both your Internet Service Provider and messaging service operator are in such positions – all your messages pass through their servers – so that if they obtained your keys, they would never be locked out of a conversation, and you would never know.

With DECIM, the recipient’s device automatically certifies new key pairs, storing the certificates in a tamper-resistant public ledger.

The team undertook a formal security analysis using a symbolic protocol verification tool, the ‘Tamarin prover’, which runs millions of possible attack situations, verifying DECIM’s capabilities. This is a rare step for a messaging protocol, and the same analysis for other protocols revealed several security flaws.

“There’s no silver bullet in the field of end-to-end encryption”, said Dr. Yu, “but we hope that our contribution can add an extra layer of security and help to level the playing field between users and attackers.”

Professor Mark Ryan, from the School of Computer Science at the University of Birmingham, said, “Our Security and Privacy group tries to solve problems that are important to society. Given the prevalence of cyber-attacks on phones and laptops, we are proud of this work on detecting when encryption keys have become compromised. Next, we intend to apply for this work on detecting encryption key compromise to applications, for example in blockchain or in Internet-based voting.”

Older users like to snoop on Facebook, but worried others might snoop on them

Older adults are drawn to Facebook so they can check out pictures and updates from family and friends, but may resist using the site because they are worried about who will see their own content, according to a team of researchers.

In a study of older people’s perception of Facebook, participants listed keeping in touch, monitoring other’s updates and sharing photos as main reasons for using Facebook. However, other seniors listed privacy, as well as the triviality of some posts, as reasons they stay away from the site.

“The biggest concern is privacy and it’s not about revealing too much, it’s that they assume that too many random people out there can get their hands on their information,” said S. Shyam Sundar, distinguished professor of communications and co-director of the Media Effects Research Laboratory, Penn State. “Control is really what privacy is all about. It’s about the degree to which you feel that you have control over how your information is shared or circulated.”

The researchers, who report their findings in a forthcoming issue of Telematics and Informatics, available online now, said that Facebook developers should focus on privacy settings to tap into the senior market.

“Clear privacy control tools are needed to promote older adults’ Facebook use,” said Eun Hwa Jung, assistant professor of communications and new media, National University of Singapore. “In particular, we think that privacy settings and alerts need to be highly visible, especially when they [older adults] are sharing information.”

While older adults are leery about who is viewing their posts, they enjoy using the site to look at pictures and read posts from friends and family, according to the researchers.

“I am more of a Facebook voyeur, I just look to see what my friends are putting out there,” one participant told the researchers. “I haven’t put anything on there in years. I don’t need to say, ‘I’m having a great lunch!’ and things like that, I don’t understand that kind of communication.”

Sundar said that, in fact, many participants mentioned the triviality of the conversation that kept them from using Facebook.

“They believe that people reporting on the mundane and unremarkable things that they did — brushing their teeth, or what they had for lunch — is not worth talking about,” said Sundar. “That’s an issue, especially for this generation.”

Older users could be a significant resource to help drive the growth of Facebook and other social media sites, Sundar said.

“The 55-plus folks were slow initially in adopting social media, but now they are one of the largest growing sectors for social media adoption,” he said.

The researchers suggest that Facebook is helping to serve as a communications bridge between the generations and that young people are prompting their older family members to join the site.

“In particular, unlike younger people, most older adults were encouraged by younger family members to join Facebook so that they could communicate,” said Jung. “This implies that older adults’ interaction via social networking sites can contribute to effective intergenerational communication.”

The researchers recruited 46 participants who were between 65 and 95 years old to take part in in-depth interviews. The group included 17 male participants and 29 female participants, all of whom had a college degree. The participants also said they used a computer in their daily lives.

A total of 20 Facebook users and 26 non-users participated in the study. If participants had a Facebook account, researchers asked them about their experience and their motivations for joining. Participants who did not use Facebook were asked why they did not join.

Because all of the participants in this study lived in a retirement home, the researchers said that future research should look at the perception and use of Facebook by seniors who live alone.