Small changes in how a person talks could reveal Alzheimer’s earlier
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.”
Gene Transfer Leads to Longer Life and Healthspan
The naked mole rat won’t win any beauty contests, but it could possibly win in the talent category. Its superpower: fighting the aging process to live several times longer than other animals its size, in a state of youthful vigor.
It’s believed that naked mole rats experience all the normal processes of wear and tear over their lifespan, but that they’re exceptionally good at repairing the damage from oxygen free radicals and the DNA errors that accumulate over time. Even though they possess genes that make them vulnerable to cancer, they rarely develop the disease, or any other age-related disease, for that matter. Naked mole rats are known to live for over 40 years without any signs of aging, whereas mice live on average about two years and are highly prone to cancer.
Now, these remarkable animals may be able to share their superpower with other species. In August, a study provided what may be the first proof-of-principle that genetic material transferred from one species can increase both longevity and healthspan in a recipient animal.
There are several theories to explain the naked mole rat’s longevity, but the one explored in the study, published in Nature, is based on the abundance of large-molecule high-molecular mass hyaluronic acid (HMM-HA).
A small molecule version of hyaluronic acid is commonly added to skin moisturizers and cosmetics that are marketed as ways to keep skin youthful, but this version, just applied to the skin, won’t have a dramatic anti-aging effect. The naked mole rat has an abundance of the much-larger molecule, HMM-HA, in the chemical-rich solution between cells throughout its body. But does the HMM-HA actually govern the extraordinary longevity and healthspan of the naked mole rat?
To answer this question, Dr. Vera Gorbunova, a professor of biology and oncology at the University of Rochester, and her team created a mouse model containing the naked mole rat gene hyaluronic acid synthase 2, or nmrHas2. It turned out that the mice receiving this gene during their early developmental stage also expressed HMM-HA.
The researchers found that the effects of the HMM-HA molecule in the mice were marked and diverse, exceeding the expectations of the study’s co-authors. High-molecular mass hyaluronic acid was more abundant in kidneys, muscles and other organs of the Has2 mice compared to control mice.
In addition, the altered mice had a much lower incidence of cancer. Seventy percent of the control mice eventually developed cancer, compared to only 57 percent of the altered mice, even after several techniques were used to induce the disease. The biggest difference occurred in the oldest mice, where the cancer incidence for the Has2 mice and the controls was 47 percent and 83 percent, respectively.
With regard to longevity, Has2 males increased their lifespan by more than 16 percent and the females added 9 percent. “Somehow the effect is much more pronounced in male mice, and we don’t have a perfect answer as to why,” says Dr. Gorbunova. Another improvement was in the healthspan of the altered mice: the number of years they spent in a state of relative youth. There’s a frailty index for mice, which includes body weight, mobility, grip strength, vision and hearing, in addition to overall conditions such as the health of the coat and body temperature. The Has2 mice scored lower in frailty than the controls by all measures. They also performed better in tests of locomotion and coordination, and in bone density.
Gorbunova’s results show that a gene artificially transferred from one species can have a beneficial effect on another species for longevity, something that had never been demonstrated before. This finding is “quite spectacular,” said Steven Austad, a biologist at the University of Alabama at Birmingham, who was not involved in the study.
Just as in lifespan, the effects in various organs and systems varied between the sexes, a common occurrence in longevity research, according to Austad, who authored the book Methuselah’s Zoo and specializes in the biological differences between species. “We have ten drugs that we can give to mice to make them live longer,” he says, “and all of them work better in one sex than in the other.” This suggests that more attention needs to be paid to the different effects of anti-aging strategies between the sexes, as well as gender differences in healthspan.
According to the study authors, the HMM-HA molecule delivered these benefits by reducing inflammation and senescence (cell dysfunction and death). The molecule also caused a variety of other benefits, including an upregulation of genes involved in the function of mitochondria, the powerhouses of the cells. These mechanisms are implicated in the aging process, and in human disease. In humans, virtually all noncommunicable diseases entail an acceleration of the aging process.
So, would the gene that creates HMM-HA have similar benefits for longevity in humans? “We think about these questions a lot,” Gorbunova says. “It’s been done by injections in certain patients, but it has a local effect in the treatment of organs affected by disease,” which could offer some benefits, she added.
“Mice are very short-lived and cancer-prone, and the effects are small,” says Steven Austad, a biologist at the University of Alabama at Birmingham. “But they did live longer and stay healthy longer, which is remarkable.”
As for a gene therapy to introduce the nmrHas2 gene into humans to obtain a global result, she’s skeptical because of the complexity involved. Gorbunova notes that there are potential dangers in introducing an animal gene into humans, such as immune responses or allergic reactions.
Austad is equally cautious about a gene therapy. “What this study says is that you can take something a species does well and transfer at least some of that into a new species. It opens up the way, but you may need to transfer six or eight or ten genes into a human” to get the large effect desired. Humans are much more complex and contain many more genes than mice, and all systems in a biological organism are intricately connected. One naked mole rat gene may not make a big difference when it interacts with human genes, metabolism and physiology.
Still, Austad thinks the possibilities are tantalizing. “Mice are very short-lived and cancer-prone, and the effects are small,” he says. “But they did live longer and stay healthy longer, which is remarkable.”
As for further research, says Austad, “The first place to look is the skin” to see if the nmrHas2 gene and the HMM-HA it produces can reduce the chance of cancer. Austad added that it would be straightforward to use the gene to try to prevent cancer in skin cells in a dish to see if it prevents cancer. It would not be hard to do. “We don’t know of any downsides to hyaluronic acid in skin, because it’s already used in skin products, and you could look at this fairly quickly.”
“Aging mechanisms evolved over a long time,” says Gorbunova, “so in aging there are multiple mechanisms working together that affect each other.” All of these processes could play a part and almost certainly differ from one species to the next.
“HMM-HA molecules are large, but we’re now looking for a small-molecule drug that would slow it’s breakdown,” she says. “And we’re looking for inhibitors, now being tested in mice, that would hinder the breakdown of hyaluronic acid.” Gorbunova has found a natural, plant-based product that acts as an inhibitor and could potentially be taken as a supplement. Ultimately, though, she thinks that drug development will be the safest and most effective approach to delivering HMM-HA for anti-aging.
In recent years, researchers of Alzheimer’s have made progress in figuring out the complex factors that lead to the disease. Yet, the root cause, or causes, of Alzheimer’s are still pretty much a mystery.
In fact, many people get Alzheimer’s even though they lack the gene variant we know can play a role in the disease. This is a critical knowledge gap for research to address because the vast majority of Alzheimer’s patients don’t have this variant.
A new study provides key insights into what’s causing the disease. The research, published in Nature Communications, points to a breakdown over time in the brain’s system for clearing waste, an issue that seems to happen in some people as they get older.
Michael Glickman, a biologist at Technion – Israel Institute of Technology, helped lead this research. I asked him to tell me about his approach to studying how this breakdown occurs in the brain, and how he tested a treatment that has potential to fix the problem at its earliest stages.
Dr. Michael Glickman is internationally renowned for his research on the ubiquitin-proteasome system (UPS), the brain's system for clearing the waste that is involved in diseases such as Huntington's, Alzheimer's, and Parkinson's. He is the head of the Lab for Protein Characterization in the Faculty of Biology at the Technion – Israel Institute of Technology. In the lab, Michael and his team focus on protein recycling and the ubiquitin-proteasome system, which protects against serious diseases like Alzheimer’s, Parkinson’s, cystic fibrosis, and diabetes. After earning his PhD at the University of California at Berkeley in 1994, Michael joined the Technion as a Senior Lecturer in 1998 and has served as a full professor since 2009.
Dr. Michael Glickman