OpenIDEO is an open innovation platform. Join our global community to solve big challenges for social good. Sign up, Login or Learn more

Myco (My Cognition) the voice assistant app for Mild Cognitive Impairment (MCI) tracking

VUI that interacts with a care-recipient, performing tests in order to provide the caregiver with reports on recipient's cognitive changes.

Photo of Molly Oberholtzer
28 11

Written by

Who is your idea designed for and how does it better support family caregivers as they care for a loved one with dementia?

Myco is for a technologically savvy, data-minded, (and totally overwhelmed) caregiver, who may begin their caregiving journey long distance, and is caring for someone who has just been diagnosed with the earliest stage of dementia. Myco delivers quantified reports for the caregiver; which help by providing evidence to eliminate second-guessing, prompt conversations with doctors, plan ahead for upcoming decisions, and prioritize current levels of care.


Caregiver's are expected, without knowledge or expertise, to manage a transitional period in loved one's life. Myco is a VUI app that supports them in this new role. By gathering data and providing synthesized reports, Myco helps the caregiver with decision making, doctor visits, increasing care responsibilities, and how to let family members know that it is time to act.

MYCO Value Propositions:

  • For the Caregiver: Myco offers peace of mind and less of a "roller coaster experience," by detecting small changes and processing them for  evidence of larger trends that signal what is around the corner.
  • For the Healthcare Provider: Myco provides evidence needed to justify further testing, and can act as longitudinal testing needed for a diagnosis. 
  • For the Bio Staticians/Scientists: Myco allows users to opt-into sharing their data critical to better understand the pre-clinical aspects, and focus on prevention strategies or treatments. 
  • For the Care-Recipient: Myco is simple brain tests that anyone can take to quantify their self and cognition. Myco can help them prove that they can take care of themselves, or help them to accept and face the facts that they can't. 

MYCO Details

Early Intervention

It is hard to pin point early signs of mild cognitive impairment, the precursor to various forms of dementia. It can be 2 years of depression and anxiety before any clinical prognosis of dementia occurs, usually because a loved one has noticed some changes in memory. When this does get noticed, it is likely to be at a later stage, and require immediate intervention and care. All of a sudden, lives are upended. The earliest phase is the best time to start their new role as a caregiver.  With a more smooth and gradual transition into caregiving, they can get used to the role, and prepare for the future changes with the participation and help of the care-recipient. Too often this time is lost because of late detection or denial.

Direction: What and When

The categories that are linked to the score and presented in the report are based off ADLs and IADLs, because geriatricians assess ADLs and IADLs as part of assessing an older person’s “function.” Generally, older adults need to be able to manage ADLs and IADLs in order to live independently without the assistance of another person.

Instrumental Activities of Daily Living (IADLs): These are the self-care tasks which require more complex thinking skills, including organizational skills. 

Activities of Daily Living (ADLs): These are the basic self-care tasks that we initially learn as very young children. 

The report detects changes and scores them, assigning a color code to the categories of IADLs and ADLs, so that the caregiver can quickly see if an area is threatened or not. If the financial planning category is in yellow, the caregiver knows that (warning!) it may be time to start thinking about it talking to the bank. 

Easy Integration

  • Voice-enabled smart speakers to reach 55% of U.S. households by 2022, says report. Combined with smartphones and smart speakers, the report forecasts the total number of voice assistant devices will reach 870 million in the U.S. by 2022 (TechCrunch) 
  • This solution ties into current quantified-self and data-tracking consumer trends (such as Fitbits.)
  • As people become more used to conversations with VUIs, the increase in interaction with the VUI means more accurate readings on their verbal cognition, and more advanced detection. 
  • The cost of the app could be mitigated by the value derived from the data provided by participants, and outside research funding.

Emotional Well-being

For the care-giver, the functionality of the VUI emphasizes the value and importance of team work at every possible moment. The easy-to-share buttons and prompts are to remind the caregiver to ask for help, to share the burden, to let those offering help figure out themselves, if possible, how they can do it, simply by reviewing the report. Sometimes when asked "how can I help" the caregiver is too occupied, busy, or stressed, to think clearly. The report does that for them, and makes team-decisions easier to make with concrete information to center a discussion around. 

For the care-recipient it is easier to confess concerns, fears, and forgetfulness to an app, rather than to a child, spouse, or friend. This circumnavigates the problematic area of role-reversal. The VUI personality acts as an intermediary step from self-care to shared-care, paving the way for a caregiver to assume new responsibilities. In the earliest stages, the reports feel like much-needed directions, when you don't know where to turn, or what to do. 

Connie and the VUI

Cognitive Testing

Most of the test that are approved for cognitive testing can be done quite simply by the physician. Some options are presented in the cognitive testing toolkit from the Alzheimer's Association, and are designed for clinical use during a yearly exam. But such testing can be more reliably conducted, via voice, in a familiar environment, when patients are relaxed, and the data can be calibrated over time and repeated tests. The voice assistant is already recording speech, and analyzing it. The framework of Myco would be to turn that into data reports that tracks cognitive status over time. 

"Speech sampling and analysis is by definition evaluating the cognitive domain of language (as opposed to problem solving or memory). To date it has been most effective in distinguishing controls from individuals with dementia. " 

Nick Bott, Research Scientist at Stanford School of Medicine

Specific tests performed by Myco include variations that are similar to classic games such as Scattegories and concentration. 

  • One-minute naming test/Verbal fluency test
  • Three Word Recall
  • Abstraction (similarity between X and Y) 

Experiments & Refinements

First user journey: Google document of feedback here

Second user journey: Google document of feedback here

Prototype Report (see below): Google document of feedback here

Benefits + VALUE 

(Quotes from caregiver feedback)

"Wow. It provides an absolutely neutral set of information based on neutral expertise.  In other words, it removes the "personal" from the decision.  You're not making a decision based on your best hunch, and you can fall back on the data to provide you confidence."

"It's like a second helper."

"With this information; It would give me something that would help me talk with relatives about how we could combine forces to help my gran get around."

This would help keep a clear record of daily progression, as a caregiver you are dealing with a lot so this would be helpful."

"It would be a less intrusive way to make decisions. My mother has to accompany my grandmother to all appointments to get the true story, having an app that tracks this would be beneficial over time. Less emotional, more facts-based. Something she could share with her siblings (they live far away)"

"I would use this app to better communicate with the PLWD's doctor on what their abilities and needs are as well as to better formulate my own questions before making decisions."


What early, lightweight experiment might you try out in your own community to find out if the idea will meet your expectations?

Team value: Get feedback from pcp, social workers on the (prototyped) report as used in discussion with a caregiver and care-recipient. Interaction value: Test the experience of interacting with the VUI and participating in the tests for a month. Data-driven decision-making value: Provide a report to a caregiver who is giving care to a loved one in an early stage, or to a professional who oversees transitions from assisted living to memory care units.

What skills, input, or guidance from the OpenIDEO community would be most helpful in building out or refining your idea?

The OpenIDEO has been very helpful with feedback and input. I appreciate everyone who has taken the time to comment and consider this idea. The idea has morphed greatly based on feedback, so I couldn't have done it without OpenIDEO. I hope to have the community's support in further iterations. Thank you!

How long has your idea existed?

  • 4 months - 1 year

This idea emerged from

  • A group brainstorm
  • An Individual

Tell us about your work experience:

I have a background in human-centered design, and studied service design at Parsons New School for Design. I have work experience in service design research for healthcare such as Memorial Sloan Kettering Cancer Center, and I have experience caregiving for my grandmother as she has slowly declined.

How would you describe this idea while in an elevator with someone?

Caregiver's are expected, without knowledge or expertise, to manage a transitional period in loved one's life. Myco is a VUI app that supports them in this new role. By gathering data and providing synthesized reports, Myco helps the caregiver with decision making, doctor visits, increasing care responsibilities, and how to let family members know that it is time to act.

How does your idea demonstrate our Criteria of Accessibility?

Insurance companies should see the viability and money-saving aspects of this app (with it's emphasis on preemptive screening measures) as most of healthcare costs are emergency based. The app itself would be inexpensive, and the data derived would provide immeasurable value for the science and research community.

How does your idea demonstrate or plan to demonstrate scalability?

Because it is a voice assistant applicaiton that situates itself between existing care-team members, it's ability to scale is unlimited, at least within english speaking areas. It could be translated to scale further, once the mechanisms were down it would be a matter of re-writing the language processing AI. At larger scale, the app is increasingly helpful as a resource for data for the scientific research/biostatician community.

How do you plan to measure the impact of your idea?

The impact could be measured in total downloads, active users, and health care professionals who approve of, endorse, or recommend it to their patients and patient's caregivers. At a different or alternate level, it could be used as a company tool for social workers who work with large groups of residents such as those as assisted living facilities.

What are your immediate next steps after the Challenge?

Developing a beta app that can be tested on a voice assistant device such as Google Home. Refining the report delivery and multi-platform functionality. Getting feedback from healthcare providers about what they would like to see in the app. Exploring options to use it in a B2B sense i.e. with assisted living homes.

Attachments (1)


my compiled research (as of Nov 10) on timeline, diagnosis, and UX for the elderly

This inspired (1)

An App for Mom


Join the conversation:

Photo of Obat Perangsang Wanita

Obat Perangsang Wanita Yang di Proses Secara Tradisional dengan Bahan Herbal Pilihan Berfungsi Untuk Meningkatkan Gairah Nafsu Wanita Secara Cepat

Photo of Yayan
Team _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Photo of hudangsubuh jahari

Photo of Molly Oberholtzer

Update! Feedback from a primary care physician: "I think if there was some sort of home testing done it would provide more data points and allow me to better treat my patients"
By helping the primary care physician in the early stages, and providing educational need-to-know information for the caregiver to discuss with the PCP, Myco facilities the care team's cohesion. With easy to distribute report touchpoints-- that encourage the growth of an informed, on-the-same-page team--Myco alleviates the pressure that would otherwise fall upon a solitary, isolated caregiver.

Photo of Brittany Margot

Hi Molly Oberholtzer - Great work on your User Journey Map! I love that you were inspired by Derrick from the OpenIDEO User Personas. As we approach the last week of Refinement, I have a few more questions for you:

1. Do you have experience with brain health games?
2. What sort of evidence exists that VUI technology can be used for monitoring?
3. Can you think of other ways in which VUI could a useful social engagement tool for patients with early-stage dementia?

Here are some thoughts from Nick Bott , Research Scientist at Stanford School of Medicine, who is on our Expert Advisory Panel:

"Having a VUI to help caregivers makes a lot of sense, assuming that there is an API that would allow Alexa to read in relevant cognitive data from wherever it was collected. Regarding the question of the evidence of VUI to track cognition and cognitive decline, there has been increasing interest in this. Speech sampling and analysis is by definition evaluating the cognitive domain of language (as opposed to problem solving or memory). To date it has been most effective in distinguishing controls from individuals with dementia. There hasn't been longitudinal tracking data of individuals using this paradigm but it stands to reason that it could. An adjacent possible is written narrative. Older studies have demonstrated that text analysis of structured writing samples track with cognitive status over time. As a virtual companion VUI could offer a number of other benefits to individuals with early-stage dementia. One of the challenges caretakers face is repetitive question asking or repetitive story telling due to memory difficulty. Alexa would be happy to answer the same question and would certainly not get bored hearing the same story again. Messages from friends and family could be forwarded to the patient and Alexa could provide a daily stream of verbal interaction from the individual's primary social support (think video log from Interstellar). A VUI could also break complex tasks down into more simple and manageable parts, whether cooking with a recipe, managing household chores, etc. It could also help with reminders for med-management or the delivery of other care protocols."

Photo of Molly Oberholtzer

Hello, Brittany Margot and Nick, thanks so much for your feedback and input. Many functions that are already part of Alexa's functionality make this idea easy to integrate into a daily routine, and allow it to better support the care team. Such as the reminders for medications, and answering questions that you mentioned--Alexa would not be bothered by being asked "What day is today" again and again!
The inherent social function of the VUI cannot be ignored-- as emotional support and as verbal (additional) cognitive testing. Regarding social engagement, I think there is potential, but that aspect of the VUI may be best developed in another iteration of this project. Patients with early stage dementia are still quite capable and cognizant. The VUI can gather verbal interactions, and summarize information and deliver it to the caregivers and team in an easy to share format so that they are better prepared and equipped, perhaps even enthused for their own conversations with the person living with dementia.
After the social interaction where the plwd is asked how they are, etc, the verbal challenges would be adaptations of the following:
- One-minute naming test/Verbal fluency test
- Three Word Recall
- Abstraction (similarity between X and Y)
These are simple and playful, and have additional clinical utility; they are part of recommendations from the Cognitive Assessment Toolkit from the Alzheimer's Association (
As-is, they are already similar to beloved/classic games, such as Scattegories, The Dictionary Game, or Concentration. I've tested them out with a few groups and they are quick, easy, and feel more like a challenge than a test. I've looked into games designed for data gathering, and I am greatly inspired by those at Sea Hero Quest ( and I think that these tests can be further augmented into a more playful and game-like format.

Primarily, the VUI is to deliver the cognitive testing, a shortened version of standard cognitive exams used by practitioners. They are a tool to better communicate with your doctor and the care team, and simply help answer the daily questions of "is this normal, are we ok?"

I hope to have some new refinement and recent work up soon, and to integrate your ideas and inquiries into it. I look forward to your thoughts, thanks again so for your input.

Photo of Wendy Carmical

Hi Molly - took a look at the user journey - love it! I really like cartoons. It communicates the idea really well. As you said, I do think this is forward thinking and will be great for the huge population that is aging. My Mom wouldn't have done games, but she isn't your target.

So a couple thoughts - it seems like you will need a companion device to gather data? It isn't clear yet if that is included. What I love is the aspect of support for the caregiver. If the person with dementia won't participate - you could have a manual version where the caretaker can check off changes in behavior. I can't tell you how many times I've gone online to recheck the stages and try to estimate where on the spectrum my Mom is.

There is a company called Miro Health you should check out.

: )

Photo of Layne Hubbard

Dear Molly,

I really appreciate your specific and detailed requests for input. It inspired me to think with more specificity when I reach out to my community as well.

I appreciate your ideas to use VUIs to mitigate some of the challenges of role-reversal—I think it can help maintain playful, joyful relationships by lessening the dependence on persistent directives, while also helping increase autonomy through goal-setting and follow-through. In this way, people can select areas where they want support and can be personally aligned with the interaction.

Best of luck in your quest!

Photo of Molly Oberholtzer

Hi Layne , thanks! I hope to get more specific as I go, maybe even add a link to a google survey I can share and so on. I think the target user, Derrick, at 35 and in IT/tech is a good representative of a growing population that relies on data for decision making. PS I love that you are in HCI and have a background in early childhood, are you participating in the other challenge, Early Childhood innovation PRize?

Photo of Layne Hubbard

Hi Molly,

Great idea to include a survey or Q&A, especially as people have just perused your content.

I previously worked as a Data Engineer at this data startup: There is certainly a strong need for people to be able to access data-driven information, though I'd argue that emotions are also carriers of data. We also need ways to decode emotional information as well as to harness it for benefit.

And yes! I've been following the Early Childhood prize closely and am hoping to contribute in one way or another. :)

Keep in touch,

Photo of Wendy Carmical

Congratulations! Please let me know if there is anything I can do to help you. I highly recommend It is a great prototyping tool, and free for three projects. Super easy to use and you end up with a link you can send.

Photo of Molly Oberholtzer

Thanks Wendy Carmical ! I'd love your feedback as a caregiver on a scenario that I am (almost) ready to post... Also, as a game designer, do you have any thoughts on the brain game side of things? Would you be interested in teaming up? Thanks again :)

Photo of Wendy Carmical

Absolutely - if you send to my gmail: first.last@gmail I will get back to you quickly.

I am disheartened that my idea didn't make it further, can't help that. But I'm happy to help you in any way I can. As to the brain game, I think it is important to keep your brain active, but my Mom was well read, healthy - she ate like the perfect diet: salmon, brown rice, veg, no sugar, no dairy, for 30 years - never smoked, didn't do drugs, did drink wine and coffee. Exercised daily. None of it mattered.

Her Mom also had dementia, but they say genetics is only part of the picture:

What I think is - you can do the preventive steps, but if you are going to get it - you're going to get it. How it progresses may be helped, but I don't really entirely buy it based on my experience. Personally, I fervently hope they find a cure.

My other suggestion is to realize the dementia patient has diminished capacities. My Mom first behaved as a teenager, then 10, then 4, now....she is present, aware, makes facial responses, but she can't do anything without help. That said, she doesn't seem like an's hard to explain.

Photo of Molly Oberholtzer

Thanks Wendy Carmical ! I too am sorry that your idea did not make it further. I hope you continue to work on it, and I am still happy to help. I just added the user journey. Your feedback would be great, and I'll have some specific questions soon. The idea has changed a bit, it is not so much preventative, but quantitative. The data would ideally contribute to personalized medicine and a cure in the future. Realizing the scope of diminished capacities is exactly what I want to focus on-- it is hard for a new caregiver to come to terms with the fact that they may have to watch their parent brush their teeth to ensure they do it, and so on! The visuals/history doesn't match, there is a disconnect that I hope the data reports can assist with.

Photo of Brittany Margot

Hi Molly Oberholtzer - welcome to Refinement!

We're eager to see updates to your idea and learn more. Don't forget there are a few additional questions now that you're in the Refinement Phase. You can see them by clicking "Edit Contribution" and scrolling to the questions that don't yet have answers from you. We also encourage you to add a user journey map and start thinking about a small experiment you could test in the next few weeks. Please reach out with questions! or feel free to email me at

Photo of Molly Oberholtzer

Thanks Brittany Margot , I've been on vacation but I am back and have watched the video call (thanks for sharing it!) that I missed. I've taken a stab at the questions, and I am working on a user-journey narrative, and hope to get feedback on it shortly.

Photo of Barbara Will

HI Molly,
I really like this idea. I work as a Family Caregiver Coordinator for seniors and their families and run a caregiver support group. Questions arise all the time, such as, is this behavior "normal'" how do I get them to stop driving, when are they unsafe to be home alone? As an outsider looking in, the answer can seem obvious, but when you are so close to the situation, it becomes a bit harder. So many emotions come into play for the caregiver spouse who isn't ready to accept the fact that changes are taking place and when you add the complexity of denial and the emotions from the care recipient it can compound a difficult situation. I have found more times than not a caregiver will do nothing until to an emergency type situation forces their hand. I think this could be a good tool to monitor the progression of cognitive decline, which when paired with professional medical advice can help prepare the caregiver for the upcoming challenges he or she will face.

Photo of Molly Oberholtzer

Thanks Barbara Will !! Im so encouraged that you get it, and see the value. (I may quote you on the feedback, is that OK!?) Feedback from caregivers so far has been really supportive as well, that this would help reinforce their decision making, and empower their choices. One of the visual I made to symbolize this idea is the "no rollercoaster" -- that this way, sudden changes and unexpected events are less likely, and the process is smoother, more manageable. Thanks again, and thanks for being a caregiver coordinator and being awesome, running a support group! Your efforts make such a difference, and I'm glad you are in the challenge.

Photo of Anuhya Tadepalli

This sounds like a great idea and I see several synergies between your approach and Digital Companion.

Photo of Molly Oberholtzer

Thanks Anuhya Tadepalli ! I'd love to get more of your thoughts and feedback on this, as it has evolved a bit since your comment. Let me know if you have time to chat! You can email me at

Photo of Kate Rushton

Hi Molly,

It is great to see the activity on your post.

How would this compare to an evaluation by a medical professional? How would you encourage users to seek medical advice?

How would the caregiver be directed to information that would help them in their new role?

Photo of Molly Oberholtzer

An evaluation by a medical professional is always to bottom line. But this would help the caregiver and medical professional PCP discuss the results, and reinforce decision making for both the caregiver and PCP, since the path to diagnosis often involves interviews or needs longitudinal testing. The VUI will prompt the user often to discuss with a doctor, and offer to share data/results with a doctor. The idea that it is positing data to be discussed by the entire team. The caregiver would be directed to information as it was needed, encouraging them to focus on problems at hand, and making them aware of what is around the corner.

Photo of Richard Caro

Hi Molly: I think this is an excellent idea. A concept like this has come up a number of times in discussions by our community of "Longevity Explorers" (older adults exploring the intersection of technology and aging). Quite a few of our explorers like the basic idea, which is good news. :)

I have the impression that the technology may not be quite there yet, but then again maybe it just needs to be pulled together. If you decide to proceed with this, we may be able to help.

You can see more about the Longevity Explorers here:

Photo of Molly Oberholtzer

Hi Richard, thank you! And thanks for sharing, I'd love to learn more about the Longevity Explorers, and where your discussions have taken you. I've reached out via the website contact and I hope we can be in touch. Thanks again, Molly

Photo of Wendy Carmical

This is an interesting idea - using data. The dementia diagnosis causes alot of anxiety. I like that you are targeting Baby Boomers, who are more likely to be tech-savvy.

Photo of Molly Oberholtzer

Thanks Wendy Carmical ! I know it is future-thinking, but this problem isn't going anywhere. Memory decline is an inevitable development, but it can be delayed or countered by activities that increase gray matter, and hopefully anxiety can be ameliorated with those actions and sense of agency. The first person to assume the caregiver role is the diagnosed individual--self-care. The data supports them and helps them to consider the trajectory and prepares them (or at least gives them the time to prepare) to hand it off when the time comes. It reminds me of this article/podcast-- "Bud Hammes works as a medical ethicist at a local hospital called Gundersen Health System. For years, he was called when someone's dad had a stroke, was in a coma, on machines. Bud would sit down with the family and try to help them figure out what to do next. And every time, he says, the discussion was excruciating."

Photo of Kate Rushton

CHi Molly,

Great to see you in the challenge!

Could this be linked to longer-term care planning i.e. CAREGIVING TRAJECTORIES ?

Photo of Molly Oberholtzer

Hi, yes! Thanks Kate Rushton   for linking me to Bob Kelly AIA, IDSA   's research post, I'd love to link the ideas. I think that in mapping a caregiving trajectory, it is absolutely essential to align it to the biometric data because caregiving tasks are inversely correlated to physical and cognitive abilities. Especially with unexpected events, and different types of dementia, biometrics are very helpful. If caught early enough, the patient with newly diagnosed dementia could also participate in the planning phase, using the trajectory map. Each timeline point as such would have a number of options that a caregiving team could consider, then decide upon according to their context. Timeline points of say, "failed cognition test for driving" would always be aligned to "needs designated person for rides," or "needs to hire a caregiving helper for 1-2 hours a week." Timing itself isn't assured, but increasing care needs are. Points may not even be sequential, but the trajectory doesn't have to be entirely linear... @Bob Kelly AIA, IDSA , IDSA what do you think?