In an initial study, Canary analyzed speech recordings with AI and identified early stage Alzheimer’s with 96 percent accuracy.
Adobe Stock
Dave Arnold retired in his 60s and began spending time volunteering in local schools. But then he started misplacing items, forgetting appointments and losing his sense of direction. Eventually he was diagnosed with early stage Alzheimer’s.
“Hearing the diagnosis made me very emotional and tearful,” he said. “I immediately thought of all my mom had experienced.” His mother suffered with the condition for years before passing away. Over the last year, Arnold has worked for the Alzheimer’s Association as one of its early stage advisors, sharing his insights to help others in the initial stages of the disease.
Arnold was diagnosed sooner than many others. It's important to find out early, when interventions can make the most difference. One promising avenue is looking at how people talk. Research has shown that Alzheimer’s affects a part of the brain that controls speech, resulting in small changes before people show other signs of the disease.
Now, Canary Speech, a company based in Utah, is using AI to examine elements like the pitch of a person’s voice and their pauses. In an initial study, Canary analyzed speech recordings with AI and identified early stage Alzheimer’s with 96 percent accuracy.
Developing the AI model
Canary Speech’s CEO, Henry O’Connell, met cofounder Jeff Adams about 40 years before they started the company. Back when they first crossed paths, they were both living in Bethesda, Maryland; O’Connell was a research fellow at the National Institutes of Health studying rare neurological diseases, while Adams was working to decode spy messages. Later on, Adams would specialize in building mathematical models to analyze speech and sound as a team leader in developing Amazon's Alexa.
It wasn't until 2015 that they decided to make use of the fit between their backgrounds. ““We established Canary Speech in 2017 to build a product that could be used in multiple languages in clinical environments,” O'Connell says.
The need is growing. About 55 million people worldwide currently live with Alzheimer’s, a number that is expected to double by 2050. Some scientists think the disease results from a buildup of plaque in the brain. It causes mild memory loss at first and, over time, this issue get worse while other symptoms, such as disorientation and hallucinations, can develop. Treatment to manage the disease is more effective in the earlier stages, but detection is difficult since mild symptoms are often attributed to the normal aging process.
O’Connell and Adams specialize in the complex ways that Alzheimer’s effects how people speak. Using AI, their mathematical model analyzes 15 million data points every minute, focusing on certain features of speech such as pitch, pauses and elongation of words. It also pays attention to how the vibrations of vocal cords change in different stages of the disease.
To create their model, the team used a type of machine learning called deep neural nets, which looks at multiple layers of data - in this case, the multiple features of a person’s speech patterns.
“Deep neural nets allow us to look at much, much larger data sets built out of millions of elements,” O’Connell explained. “Through machine learning and AI, we’ve identified features that are very sensitive to an Alzheimer’s patient versus [people without the disease] and also very sensitive to mild cognitive impairment, early stage and moderate Alzheimer's.” Based on their learnings, Canary is able to classify the disease stage very quickly, O’Connell said.
“When we’re listening to sublanguage elements, we’re really analyzing the direct result of changes in the brain in the physical body,” O’Connell said. “The brain controls your vocal cords: how fast they vibrate, the expansion of them, the contraction.” These factors, along with where people put their tongues when talking, function subconsciously and result in subtle changes in the sounds of speech.
Further testing is needed
In an initial trial, Canary analyzed speech recordings from phone calls to a large U.S. health insurer. They looked at the audio recordings of 651 policyholders who had early stage Alzheimer’s and 1018 who did not have the condition, aiming for a representative sample of age, gender and race. They used this data to create their first diagnostic model and found that it was 96 percent accurate in identifying Alzheimer’s.
Christian Herff, an assistant professor of neuroscience at Maastricht University in the Netherlands, praised this approach while adding that further testing is needed to assess its effectiveness.
“I think the general idea of identifying increased risk for cognitive impairment based on speech characteristics is very feasible, particularly when change in a user’s voice is monitored, for example, by recording speech every year,” Herff said. He noted that this can only be a first indication, not a full diagnosis. The accuracy still needs to be validated in studies that follows individuals over a period of time, he said.
Toby Walsh, a professor of artificial intelligence at the University of New South Wales, also thinks Canary’s tool has potential but highlights that Canary could diagnose some people who don’t really have the disease. “This is an interesting and promising application of AI,” he said, “but these tools need to be used carefully. Imagine the anxiety of being misdiagnosed with Alzheimer’s.”
As with many other AI tools, privacy and bias are additional issues to monitor closely, Walsh said.
Other languages
A related issue is that not everyone is fluent in English. Mahnaz Arvaneh, a senior lecturer in automatic control and systems engineering at the University of Sheffield, said this could be a blind spot.
“The system may not be very accurate for those who have English as their second language as their speaking patterns would be different, and any issue might be because of language deficiency rather than cognitive issues,” Arvaneh said.
The team is expanding to multiple languages starting with Japanese and Spanish. The elements of the model that make up the algorithm are very similar, but they need to be validated and retrained in a different language, which will require access to more data.
Recently, Canary analyzed the phone calls of 233 Japanese patients who had mild cognitive impairment and 704 healthy people. Using an English model they were able to identify the Japanese patients who had mild cognitive impairment with 78 percent accuracy. They also developed a model in Japanese that was 45 percent accurate, and they’re continuing to train it with more data.
The future
Canary is using their model to look at other diseases like Huntington’s and Parkinson’s. They’re also collaborating with pharmaceuticals to validate potential therapies for Alzheimer’s. By looking at speech patterns over time, Canary can get an indication of how well these drugs are working.
Dave Arnold and his wife dance at his nephew’s wedding in Rochester, New York, ten years ago, before his Alzheimer's diagnosis.
Dave Arnold
Ultimately, they want to integrate their tool into everyday life. “We want it to be used in a smartphone, or a teleconference call so that individuals could be examined in their home,” O’Connell said. “We could follow them over time and work with clinical teams and hospitals to improve the evaluation of patients and contribute towards an accurate diagnosis.”
Arnold, the patient with early stage Alzheimer’s, sees great promise. “The process of getting a diagnosis is already filled with so much anxiety,” he said. “Anything that can be done to make it easier and less stressful would be a good thing, as long as it’s proven accurate.”
Certain gases used for anesthesia are 3,000 times more damaging for the climate than CO2. Some anesthesiologists are pointing to other solutions.
Adobe Stock
On the whole, the medical sector is responsible for a disproportionally large percentage of greenhouse gas emissions, with the U.S. as the biggest culprit. According to a key paper published in 2020 in Health Affairs, the health industry “is among the most carbon-intensive service sectors in the industrialized world,” accounting for between 4.4 percent and 4.6 percent of greenhouse gas emissions. “It’s not just anesthesia but health care that has a problem,” says Jodi Sherman, anesthesiology professor and Medical Director of the Program on Healthcare Environmental Sustainability at Yale University as well as co-director of the Lancet Planetary Health Commission on Sustainable Healthcare. In the U.S., health care greenhouse gas emissions make up about 8.5 percent of domestic greenhouse gas emissions. They rose 6 percent from 2010 to 2018, to nearly 1,700 kilograms per person, more than in any other nation.
Of course, patients worry primarily about safety, not sustainability. Yet, Koch emphasizes that “as doctors, we have the responsibility to do no harm, and this includes making sure that we use resources as sustainably as possible.” Studies show that 2018 greenhouse gas and toxic air pollutant emissions resulted in the loss of 388,000 disability-adjusted life years in the U.S. alone. “Disease burden from health care pollution is of the same order of magnitude as deaths from preventable medical errors, and should be taken just as seriously,” Sherman cautions.
When Koch, the anesthesiologist, started discussing sustainable options with colleagues, the topic was immediately met with plenty of interest. Her experience is consistent with the latest representative poll of the nonprofit Foundation Health in Germany. Nine out of ten doctors were interested in urgently finding sustainable solutions for medical services but lacked knowhow and resources. For teaching purposes, Sherman and her team have developed the Yale Gassing Greener app that allows anesthesiologists to compare how much pollution they can avoid through choosing different anesthesia methods. Sherman also published professional guidelines intended to help her colleagues better understand how various methods affect carbon emissions.
Significant traces of inhalation gases have been found in Antarctica and the Himalayas, far from the vast majority of surgery rooms.
A solution espoused by both Sherman and Koch is comparatively simple: They stopped using desflurane, which is by far the most damaging of all inhalation gases to the climate. Its greenhouse effect is 2,590 times stronger than carbon dioxide. The Yale New Haven Hospital already stopped using desflurane in 2013, becoming the first known healthcare organization to eliminate a drug based on environmental grounds. Sherman points out that this resulted in saving more than $1.2 million in costs and 1,600 tons of CO2 equivalents, about the same as the exhaust from 360 passenger vehicles per year.
At the Charité, Koch claims that switching to other anesthesiology choices, such as propofol, has eliminated 90 percent of the climate gas emissions in the anesthesiology department since 2016. Young anesthesiologists are still taught to use desflurane as the standard because desflurane is absorbed less into the patients’ bodies, and they wake up faster. However, Koch who has worked as an anesthesiologist since 2006, says that with a little bit of experience, you can learn when to stop giving the propofol so it's timed just as well with a person’s wake-up process. In addition, “patients are less likely to feel nauseous after being given propofol,” Koch says. Intravenous drugs might require more skill, she adds, "but there is nothing unique to the drug desflurane that cannot be accomplished with other medications.”
Desflurane isn’t the only gas to be concerned about. Nitrous oxide is the second most damaging because it’s extremely long-lived in the environment, and it depletes the ozone layer. Climate-conscious anesthesiologists are phasing out this gas, too, or have implemented measures to decrease leaks.
Internationally, 192 governments agreed in the Kyoto protocol of 2005 to reduce halogenated hydrocarbons – resulting from inhalation gases, including desflurane and nitrous oxide – because of their immense climate-warming potential, and in 2016, they pledged to eliminate them by 2035. However, the use of inhalation anesthetics continues to increase worldwide, not least because more people access healthcare in developing countries, and because people in industrialized countries live longer and therefore need more surgeries. Significant traces of inhalation gases have been found in Antarctica and the Himalayas, far from the vast majority of surgery rooms.
Certain companies are now pushing new technology to capture inhalation gases before they are released into the atmosphere, but both Sherman and Koch believe marketing claims of 99 percent efficiency amount to greenwashing. After investigating the technology first-hand and visiting the company that is producing such filters in Germany, Koch concluded that such technology only reduces emissions by 25 percent. And Sherman believes such initiatives are akin to the fallacy of recycling plastic. In addition to questioning their efficiency, Sherman fears such technology “gives the illusion there is a magical solution that means I don’t need to change my behavior, reduce my waste and choose less harmful options.”
Financial interests are at play, too. “Desflurane is the most expensive inhalation gas, and some think, the most expensive must be the best,” Koch says. Both Koch and Sherman lament that efforts to increase sustainability in the medical sector are entirely voluntary in their countries and led by a few dedicated individual professionals while industry-wide standards and transparency are needed, a notion expressed in the American Hospital Association’s Sustainability Roadmap.
Susanne Koch, an anesthesiologist in Berlin, wants her colleagues to stop using a gas called desflurane, which is by far the most damaging of all inhalation gases to the climate.
Adobe Stock
Other countries have done more. The European Union recommends reducing inhalation gases and even contemplated a ban of desflurane, except in medical emergencies. In 2008, the National Health Service (NHS) created a Sustainable Development Unit, which measures CO2 emissions in the U.K. health sector. NHS is the first national health service that pledged to reach net zero carbon by 2040. The carbon footprint of the NHS fell by 26 percent from 1990 to 2019, mostly due to reduced use of certain inhalers and the switch to renewable energy for heat and power. “The evidence that the climate emergency is a health emergency is overwhelming,” said Nick Watts, the NHS Chief Sustainability Officer, in a press release, “with health professionals already needing to manage its symptoms.”
Sherman is a leading voice in demanding action in the U.S. To her, comprehensive solutions start with the mandatory, transparent measurement of emissions in the health sector to tackle the biggest sources of pollution. While the Biden administration highlighted its efforts to reduce these kinds of emissions during the United Nations Climate Conference (COP27) in November 2022 and U.S. delegates announced that more than 100 health care organizations signed the voluntary Health Sector Climate Pledge, with the aim to reduce emissions by 50 percent in the next eight years, Sherman is convinced that voluntary pledges are not enough. “Voluntary measures are insufficient,” she testified in congress. “The vast majority of U.S. health care organizations remain uncommitted to timely action. Those that are committed lack policies and knowledge to support necessary changes; even worse, existing policies drive inappropriate consumption of resources and pollution.”
Both Sherman and Koch look at the larger picture. “Health care organizations have an obligation to their communities to protect public health,” Sherman says. “We must lead by example. That includes setting ambitious, science-based carbon reduction targets to achieve net zero emissions before 2050. We must quantify current emissions and their sources, particularly throughout the health care supply chains.”
