Dinesh Kumar leads revolutionary stroke detection app at RMIT

By Indira Laisram
Professor Dinesh Kumar from RMIT University (right) // Pic supplied

Experts typically keep their breakthroughs to themselves, but this week, Professor Dinesh Kant Kumar is celebrating a significant achievement: an app developed by his team that could enable paramedics to identify strokes within seconds, a first-of-its-kind innovation globally.

Leading a team of biomedical engineers at RMIT University, Kumar spearheaded the development of the artificial intelligence (AI) technology powering this groundbreaking software. Designed to operate seamlessly as a smartphone app, its primary goal is to enhance early stroke recognition, crucial for minimising brain damage and improving patient outcomes.

“The technology isn’t overly complex,” Kumar explains. “It’s something that was waiting to be developed. We were surprised that no one had attempted it before,” he tells The Indian Sun.

Professor Dinesh Kumar from RMIT University. Credit: Seamus Daniel, RMIT University

Currently, emergency clinicians assess stroke symptoms like facial droop, voice changes, and hand numbness. Reflecting on their approach, Kumar notes, “We considered examining all these symptoms but found hand assessment impractical without an EMG device available in ambulances. Facial expressions emerged as the most practical focus. We steered clear of the ‘black box’ approach used by other AI engines, prioritising transparency in diagnosis.”

“We utilised action units to pinpoint facial differences, particularly around the mouth’s edges. By integrating this data into a straightforward AI system, we successfully differentiated stroke patients from others,” Kumar shares.

Like any other AI tool, Kumar and his team have developed a model with 82 percent accuracy, with plans to enhance it further by adding more data. Video recordings of facial expression examinations of 14 people with post-stroke and 11 healthy controls were used in this study.

PhD scholar Guilherme Camargo de Oliveira (left), from RMIT and São Paulo State University, and Professor Dinesh Kumar. Credit: Seamus Daniel, RMIT University

Next, registered clinical trials are planned to obtain TGA approval, after which the app will be ready for rollout. “If all goes according to plan, we should be ready in 24 months. Securing funding sources and partnerships with hospitals and industry are essential steps for us to move forward,” says Kumar.

Discussing how paramedics could integrate this tool into their emergency protocols and its potential impact on patient outcomes, Kumar explains, “Imagine an emergency scenario with paramedics attending to a person on a triple zero call. Amidst the chaos, they can’t focus on identifying changes in the person’s facial expression.”

“The bigger problem is that the changes are often very small. Additionally, women and people of colour are more likely to be underdiagnosed compared to white men. This happens because paramedics are often better at noticing subtle changes in people who look like them due to familiarity and common appearance. As a result, there is a lower diagnostic rate for people of colour. This has been shown in a study from the US. However, an app can overcome such biases in the system, as it doesn’t get stressed or biased. Those are the things we are overcoming by providing this screening tool.”

PhD scholar Guilherme Camargo de Oliveira (right) demonstrates the face screening tool with Visiting Associate Professor Nemuel Daniel Pah from RMIT University. Credit: Seamus Daniel, RMIT University

However, developing such an app has its challenges. “Obtaining facial data ethically is difficult due to consent requirements. Moreover, resource constraints were a challenge as this project was not externally funded. We were fortunate to have Guilherme Camargo de Oliveira, a bright young PhD scholar from Brazil, as the first author of this paper. He believed in the project and contributed significantly,” says Kumar, an alumnus of IIT Madras and IIT Delhi.

Kumar had been working in neurodegenerative disease diagnostics for about 10 years before focusing on stroke diagnostics and video analysis for the past three years, which led to the development of the app. He leveraged his association with São Paulo University, where one of the key ideas that emerged from this collaboration was the creation of the app.

PhD scholar Guilherme Camargo de Oliveira (right) demonstrates the face screening tool with Visiting Associate Professor Nemuel Daniel Pah from RMIT University. Credit: Seamus Daniel, RMIT University

As the app serves as a screening tool, it is crucial to incorporate training data from diverse global sources to mitigate biases. Kumar adds, “We are hoping more media exposure will amplify this part so that we can get partners from all parts of the world to be a part of this growth.”

Looking ahead, Kumar is seeking partnerships from India and beyond, including entrepreneurs and healthcare providers with deep resources. Considering the impact of stroke-related disabilities, Kumar anticipates widespread adoption of this technology in global healthcare settings. He expects healthcare systems, including stroke foundations and similar organisations, to promote its integration into hospital systems.

The Indian Sun acknowledges the support of the Victorian Government.

Connect with Indira Laisram on Twitter

Support independent community journalism. Support The Indian Sun.

Follow The Indian Sun on Twitter | InstagramFacebook


Donate To The Indian Sun

Dear Reader,

The Indian Sun is an independent organisation committed to community journalism. We have, through the years, been able to reach a wide audience especially with the growth of social media, where we also have a strong presence. With platforms such as YouTube videos, we have been able to engage in different forms of storytelling. However, the past few years, like many media organisations around the world, it has not been an easy path. We have a greater challenge. We believe community journalism is very important for a multicultural country like Australia. We’re not able to do everything, but we aim for some of the most interesting stories and journalism of quality. We call upon readers like you to support us and make any contribution. Do make a DONATION NOW so we can continue with the volume and quality journalism that we are able to practice.

Thank you for your support.

Best wishes,
Team The Indian Sun