‘We will help non-verbal people to express their emotions’

10 Feb 2021

Dr Aviva Cohen. Image: UCD

Dr Aviva Cohen shares the personal story behind her research and how she plans to help non-verbal people to communicate their needs independently.

Dr Aviva Cohen is the principle investigator on SeamlessCare, a project developing technologies to enhance the care of people with a wide range of disabilities.

Her work is funded by Enterprise Ireland and hosted in the School of Mechanical and Materials Engineering at University College Dublin (UCD). It began with postdoctoral research backed by an Assistid Marie Curie fellowship, which has funded researchers developing technology for use by people with intellectual disabilities or who are on the autism spectrum.

SeamlessCare is set to spin out of UCD later this year, but it’s not Cohen’s first research-led business. Neuro Hero, founded in 2011, arose out of a personal crisis.

In 2006, Cohen’s partner, Steve, suffered a severe stroke. Cohen explored clinically tested but lesser-known therapies for stroke to find ways to help Steve regain his cognition, speech and movement.

While he regained some of his mobility and cognition, speech continued to be an issue for Steve. So Cohen led an expert team to develop a suite of speech and language therapy apps. Today, Neuro Hero spans 19 apps available in six languages and has won numerous awards.

‘I understood that, if I wanted things to improve, l would have to work on the solutions myself’

Cohen returned to academia when she was invited to become an Assistid Marie Curie research fellow in 2016, and the organisation introduced her to healthcare professionals and families of people with additional needs.

This experience led her to devise a new app – one that would capture the insights that primary carers acquire over years, so that this valuable knowledge can be passed on. Further development, testing and funding applications led Cohen and her colleague Ian Kennedy to enhance the app with new ideas to support the care of people with additional needs.

What inspired you to become a researcher?

I loved poetry in school, especially Patrick Kavanagh and the metaphysical poets. After reviewing my Leaving Cert essays, one of my English teachers suggested that I study philosophy.

In 1984 I went to Trinity College Dublin to begin a four-year degree in pure philosophy. I loved every moment and desperately wanted to lecture because I could see how a philosophical approach could lead to creativity and foster the type of awareness that is necessary for political freedom and equality.

When Steve became ill, I realised how few resources were available to support people with additional needs and their families. I understood that, if I wanted things to improve, l would have to work on the solutions myself. I was lucky, I could use the skills I learned as a researcher and a communications consultant. I also drew heavily on the tenacity I learned from Steve, who had been my karate sensei for almost 20 years. I started to identify gaps and to build solutions wherever I could.

It still surprises me that I ended up working in technology – this had never been my forte! However, the more I researched, the more I became inspired by the potential for technological innovation. From the very start, I had looked for information and technologies to help Steve and our family. When I failed to find what we needed, I knew that I had to learn about technology so that I could begin to address this deficit.

Can you tell us about the research you’re currently working on?

SeamlessCare is now being developed as a cloud-based solution designed to ensure the best possible care for everyone, no matter how complex their needs. Our team has been joined by the talented Dr Cagri Cubukcu – he is building our APIs on the HL7 FHIR framework so that they are interoperable and comply with international healthcare and World Health Organization standards. [FHIR is a standard for healthcare data exchange published by Health Level Seven International (HL7), a not-for profit standards-developing organisation.]

We are developing four APIs that support interactions between families and formal carers, enhance the flow of information within caring facilities and promote communication with people who don’t have the capacity to use functional words.

The first API is accompanied by a user-friendly mobile app that was inspired by my postdoctoral research. It is a video-based care plan that enables carers to record and archive short videos that show the daily living requirements of each care recipient. The aim is to relay complex information quickly and accurately.

The second API and mobile app combination offers a unique opportunity to use machine learning to analyse the sounds made by non-verbal people so that other people can understand up to 10 emotional states. Using data analytics, it becomes possible to record vocalisations and generate text messages that express emotions. Our aim is to enable non-verbal people to convey their feelings independently, often for the first time in their lives.

The third API is a practical tool that will match the best qualified care provider to the specific needs of each person. This will enhance safety and quality of care. We are also building a heat-mapping system that will help to identify gaps in care, spikes in incidence and other patterns in order to enhance patient care, safety and quality of life.

In your opinion, why is your research important?

Imagine if only one or two people in your world could interpret what you are trying to say. Imagine what it’s like to work in a caring environment where you desperately want to know what a person is trying to communicate but it is simply impossible.

People with complex needs, such as intellectual disability, brain injury or communication difficulty, often become frustrated because they cannot communicate their needs and preferences. This reduces their quality of life and sometimes leads to challenging behaviours that can result in harm to themselves or to those providing care. SeamlessCare is working to change this.

At first, we will help non-verbal people to express their emotions. In the future we hope to expand upon our research so that, one day, we can translate a wide range of vocalisations.

Testing has shown that our video care plan has numerous benefits, including shorter handover times between staff members and improved interactions between family members and staff. In addition, testers reported a reduction in frustration for non-verbal people which led to fewer falls and a reduction in challenging behaviours.

Our matching algorithm goes beyond aligning clinical needs with the skills written on a care provider’s CV. We want to match people who will connect on a personal level so that people enjoy their lives rather than simply receiving care. Our heat-mapping system has the potential to quickly identify patterns that could lead to medical, behavioural or staffing issues. Initial research suggests that it could improve safety and reduce risk to both staff and patients.

What’s the future for your research?

SeamlessCare APIs will be available to healthcare providers including hospitals, rehabilitation centres, care facilities and nursing homes. Each institution can select a single API or license all four as a bundle. They can be easily integrated into existing systems and can be white-labelled if desired. The user-friendly mobile apps will be provided to care providers and to families.

As we provide a cloud-based service, there is no limit to the number of users we can support, or to the length of their contract. This means that it is possible to maintain an evolving archive of data that grows and changes throughout the lifetime of each care recipient.

Last year, we completed the 2020 UCD VentureLaunch Accelerator Programme. This was a wonderful opportunity to learn advanced business skills and to enhance our personal development. The VentureLaunch Accelerator is run by NovaUCD, who have been incredibly supportive of our work. With NovaUCD’s help, we will spin our company out of UCD before the summer of 2021.

We are now entering the second phase of testing in institutions and family homes. This is an exciting stage in the project. We will be ready to roll out the first of our APIs to the public by this summer and all four will be available by the end of 2021.

Updated, 11.15am, 10 February 2021: This article was updated to clarify that Cohen’s postdoctoral research has finished and she is no longer working in the UCD Smartlab.

Want stories like this and more direct to your inbox? Sign up for Tech Trends, Silicon Republic’s weekly digest of need-to-know tech news.

Are you a researcher with an interesting project to share? Let us know by emailing editorial@siliconrepublic.com with the subject line ‘Science Uncovered’.