Better prevention and cure thanks to artificial intelligence

(Photo: 123RF)

ARTIFICIAL INTELLIGENCE. With a healthcare system stretched thin, tools powered by artificial intelligence (AI) are worth their weight in gold, as they promise to save professionals time, facilitate diagnosis, make it easier to find the appropriate treatment and improve prevention. Here are two Montreal start-ups that offer innovations that improve care through AI.

David Landry’s grandmother met a tragic end following a fall. Left on the ground for 12 hours before she was discovered, she succumbed to a hemorrhage, which marked the entrepreneur who was only 11 years old at the time.

Seeing that technology had not advanced so that personnel or loved ones could intervene quickly in the event of a fall of an elder, David Landry co-founded LivingSafe in 2020. His company uses sensors emitting electromagnetic waves which are by subsequently received by the same device. “It’s a bit the same principle as ultrasound in bats,” he explains in a telephone interview. Waves hit objects and come back. »

This information is transmitted to AI which is able to eliminate objects or pets to follow only humans in real time. In the event of a fall, an alert is sent to designated persons.

“It’s a continuous surveillance system that does not affect privacy in any way,” says the mechanical engineer. There are no words or captured images, as there are no microphones or cameras. The eldest has nothing to wear, no watch or bracelet. In addition, the system is autonomous. No staff or user intervention is required. »

The predictive power of AI

David Landry believes that the raw information gathered is perfect for AI. “The data captured is fairly standard and in large numbers,” he says. Since they are not so variable, it allows to train the AI ​​in an efficient way. »

In addition to quickly detecting falls and alerting rescuers, AI will allow LivingSafe to go further. “The core of our business is data analysis,” he recalls. What is wonderful is that we can retrieve data before and after the incident. »

So, by dint of accumulating them, the company wants to establish correlations to better understand why and how falls occur and thus act accordingly to minimize them. “Our ultimate goal is prevention and ensuring that seniors can stay in their homes as long as possible. »

Helping healthcare professionals

The lack of data to make AI work is a big problem for start-ups or those that are innovating. So you have to be patient.

“Often, you don’t need AI at first, but then it can come and refine your solution,” explains Philippe Chartrand, founder of Empego, who has developed questionnaires to which the patient responds in order to describe their symptoms. before meeting with a doctor or pharmacist.

“The health professional will be entitled to a complete report in hand before the appointment, affirms the president. He is therefore better prepared and this will allow him not to omit important questions to be examined. »

Empego uses AI to properly align the questions asked of the patient with what drives him to go for a consultation. “We want the questions to correspond to the health problem described by the patient in the software,” he continues. Therefore, the system learns based on relevance. It becomes more and more precise over time, according to the errors and successes achieved. »

Within three years, the young shoot also plans to use AI to predict the diagnosis and suggest therapy. “By following up with the treatment chosen by the professionals, we can cross-check this with the answers provided at the start by the patient,” specifies Philippe Chartrand. We will thus be able to know what the patient is suffering from according to the answers to the questionnaire. »

Until then, there is work. “We have objectives from a technical point of view, but also from a commercial point of view, explains the pharmacist. It is worth structuring our data to later maximize the AI. We get ready right away. »

This is important, because for AI to work well, he points out that you need quality data, in large quantities and well organized. “In health, it’s often narrative data, a story, which is an obstacle for AI, says the entrepreneur. If you don’t put the right “label” in the right place, it distorts the results and in medicine, you can’t afford that. »

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *