When we make a call for help during critical situations, we trust that our request for assistance will be answered rapidly and with accuracy. There is no room for error! That is why the expert team at Corti has utilized AI technology to create a system that analyzes voice patterns over the phone to help emergency personnel recognize typical patient symptoms. Corti CEO, Andreas Cleve, knows that every moment is crucial, so precision is key. Andreas and his team are taking the guesswork out of patient care just in the nick of time.
Tamara: Describe your company and the AI/predictive analytics/data analytics products/services you offer.
Andreas: Corti is a Copenhagen-based artificial intelligence (AI) platform that uses machine learning to provide emergency medical personnel with life-saving diagnostic assistance.
By applying speech recognition and analyzing audio, language, dialect, and background noise during a live emergency 911 call, Corti’s AI technology detects critical illnesses — such as cardiac arrests — in real-time, accelerating the diagnostic process and response strategies for emergency dispatchers.
How Corti Works:
- During an emergency call, data is processed through Corti’s deep neural networks in real-time. Corti analyzes symptom descriptions, tone of voice, breathing patterns, and everything a patient says and shares, and finds important patterns by comparing it to the millions of emergency calls that the platform has already analyzed, meaning it continually learns over time
- Corti acts as a digital assistant for the dispatcher handling the call, providing them with clear-cut assistance, suggesting the best questions to ask, and making recommendations for triaging patients
- Corti increases efficiency and accuracy, allowing emergency departments to elevate their call-handling and achieve better results
Tamara: How do you see the AI/data analytics/predictive analysis industry evolving in the future?
Andreas: We hope to see some big applied AI use cases for core AI / machine learning companies, that move research from academia to a concrete, widespread use case.
Tamara: How do you see your products/services evolving going forward?
Andreas: Corti is in ongoing collaborations with other potential partners, with expansion into new markets to be announced later this year.
Tamara: How does AI, particularly your product/service, bring goodness to the world? Can you explain how you help people?
Andreas: With state-of-art machine learning technology, which enhances over time, Corti wants to advance the healthcare sector — and support emergency service call centers — by improving the speed and accuracy of diagnostics. This is achieved even when there is background noise, and regardless of whether the call is made by the patient themselves or another individual close to them.
The issue has previously been handled by emergency dispatchers on the phone, where they have to rely only on their own knowledge, handling more or less the worst days of people’s lives with no tools to do so. This is often made even more difficult when having to make sense of symptoms relayed by a panicked friend or relative. Emergency dispatchers have a tough job assuring callers while also trying to ask questions that could save the patient’s life. On top of this, they must remember to ask whoever is on the phone for the address of the incident to ensure the ambulance en route is headed to the right place.
Copenhagen EMS (Emergency Medical Services), which provides immediate care to people with acute illness or injury, has implemented Corti on the official emergency number 112, utilizing its AI to detect cardiac arrest cases faster and more accurately. Previously, dispatchers could only recognize cardiac arrest from descriptions over the phone around 73% of the time but, in an early, small-scale study, the machine learning model knew the calls were reporting cardiac arrest 95% of the time.
Tamara: What are the 3–5 things that most excite you about AI? Why? (industry specific)
Andreas: We are most excited about the possibilities for voice integrations in the emergency medical space. According to research, less than 50% of verbal information delivered at handoffs by ambulance crews was accurately retained by emergency dispatch staff. Voice-activated AI assistants can help staff retain information to get patients into the right treatment, faster, and alleviate the pitfalls of misdiagnosis early on.
Tamara: Over the next three years, name at least one thing that we can expect in the future related to AI?
Andreas: We look forward to seeing the amazing applications of AI augmentation done in real-time, to assist people as they perform a task.