Altoida: startup using AI and ML for early detection of Alzheimer’s

Alzheimer’s is a devastating disease – it’s something that I’m personally terrified of.  Not only is it the 6th leading cause of death in the US – the progression of the disease is just a horrendous, drawn-out process for everyone involved.

On the plus side, there seems to be an increasing amount of academic funding and venture capital interest in Alzheimer’s-related research.

One area I’m especially interested in is the early detection of Alzheimer’s – which is a focus also shared by the AI-driven biotech startup, Altoida.


The approach:

Altoida is essentially a tablet app using augmented reality and AI to help screen for Alzheimer’s earlier.  The app is designed to be low-friction for the user (aka ‘gamification’), so people are less intimidated by the process of getting themselves screened for Alzheimer’s – more on that below.

The app captures movements of the person as they are directed to do everyday tasks around the environment, like ‘use your finger to trace this line on the tablet’s screen’ or maybe ‘point to that picture on the wall,’

During these actions, biomarkers (such as such as pupil dilation and hand micromovements) are recorded and then analyzed by Altoida’s AI; the processed results are then sent to the patient’s medical provider.


The science:

Note: for reference, I did my undergrad in biomedical engineering at UIUC, and during that time was involved with a startup using neural networks to process EMG signals.

As far as relatively earlier-stage startups in the medical sector go, I was pretty impressed with Altoida’s science.  They’ve received more than $35 million in research & development grant money so far, which is a relatively impressive sum for where they’re at.

Through collecting data on these Alzheimer’s-screening biomarkers from each patient, Altoida is generating about 250 features to feed into it’s AI/ML logic – which they call their ‘Digital Biomarker Platform.’

The features they’re using are backed by both independent and in-house research (they have an impressive team full of PhD’s).   In particular, they mention using longitudinal studies involving 5,000 patients over 8+ years.  Again, for a non-late-stage biotech startup, that’s quite impressive.

In terms of the algorithms analyzing these features, it’s also not a one-size-fits-all approach – the analysis will adapt based on contextual factors (such as the local patient population and new biomarkers collected), and as new relevant Alzheimer’s research becomes available.

Focus on ease-of-use

If you’re anything like me, you might have some deeply-ingrained aversion about engaging with the medical industry.  It just seems the entire industry is increasingly designed to needlessly drain your energy through endless forms, appeals, bureaucracy, stress, confusion – and then, financially bleed you dry (assuming you can pay at all).

And to top it off, the whole system is apparently designed to ‘manage’ your symptoms – not cure them.  A ‘managed’ patient is much more profitable than a ‘cured’ patient.  All while healthcare costs are skyrocketing, and no one seems to have any idea of how to actually address it.

So put that all together, and you have more and more people skipping medical treatments.  It’s largely due to financial concerns, but for how much it just sucks to deal with much of today’s healthcare industry – many people are just choosing to stay away from it, as much as they possibly can.

Unfortunately, this means people without any medical training are increasingly making life-or-death decisions about whether it’s worth it for them to get that medical treatment (or screening).

Anyways, what I like about Altoida is they’re trying to make the early-detection process as non-intimidating and frictionless as possible.  A tall order for pretty much anything in the health industry these days, but with an easy-to-use app and coverage through Medicare, it’s a big step in the right direction relative to most other Alzheimer’s screening methods.

Altoida understands the growing importance of early detection, and how aggressively streamlining the screening process will make it way more likely that the people who need it most will get the help they need the earliest.


Wrapping up

Startups like Altoida are refreshing for me to look into, as we’re seeing at least an attempt to use AI and machine learning for something that would actually benefit all of society – and without me needing to use convoluted logic to convince myself why it would actually be beneficial.

So many of our best and brightest minds are using their data science or ML skills to optimize e-commerce click-through rates, figure out what the most effective messaging for getting people to buy more chips, or (like me…) working in finance.

I’ll be keeping an eye on these guys, and even if they don’t ultimately succeed, I’m hopeful that their work will at least be helpful in getting us closer to figuring out Alzheimer’s.

What do you think?  Am I right, wrong, way off?  Let me know – feel free to email me or connect on LinkedIn.



The views expressed on this site are my own and do not represent the views of any current or former employer or client. 

Leave a reply:

Your email address will not be published.

Sliding Sidebar