AI can tell how you are feeling, sometimes more accurately than you can...

Interview with Grace Chang from Kintsugi about the possibilities of AI in helping diagnose mental health and book giveaway.

Now for something both scary and hopeful. This week, I got the pleasure of chatting with Grace Chang - she is the co-founder and CEO of Kintsugi, a tech start-up developing ways that AI can recognize biomarkers to help clinicians in diagnosing mental health issues.

I met Grace last summer at the SF Art Book Fair, and when I asked her what she spent her time doing, I got extremely excited when I heard her response. As someone, like Grace, who found the mental health system super difficult to navigate, I spent years trying to find a therapist that could help me alter my meds. I was told several times that all good therapists don’t take health insurance. That made everything cost prohibitive. And I am lucky.

For most, access to mental health services is a battle. This is not new news. I remember in 2016 when I had a mental breakdown, my mother was the only one who could save me. She took a leave from work, flew out from Minnesota, and stayed with us, caring for me for two weeks until I was not suicidal. But just wait, this is not the interesting part.

In 2018, after Grace dealt with her own burnout from work, she was introduced to the work of Jim Glass, a senior research scientist at MIT’s Computer Science and Artificial Intelligence Laboratory:

“In a paper being presented at the Interspeech conference, MIT researchers detail a neural-network model that can be unleashed on raw text and audio data from interviews to discover speech patterns indicative of depression. Given a new subject, it can accurately predict if the individual is depressed, without needing any other information about the questions and answers.” Link to article.

This AI takes speech biomarkers such as pitch, speed, and frequency from voice samples, compares them to a robust data set, and can detect depression. It doesn’t matter what you are talking about, or what language you are speaking, which is so amazing.

I want to say that I know myself. I think we all assume that we know ourselves better than other people can. However, I am turning 40 this year, and I will say that lots of my patterns, good and bad, are often imperceptible to me. Often it takes being in the right place at the right time with the right person to get me to actually understand something about myself that is destructive. Like for years, I told myself that I actually don’t worry more than other people. That I am within the spectrum of normal. But this is actually not true. This also reminds me of last week, talking to Ayelet about her work and her research that we actually learn from other people’s failure more than we learn from our own. That our own biases on the topic of failure often distort our reality, distancing us from what is really going on.

Have you taken this questionnaire at the doctor’s office before:

I definitely have. And I remember thinking every time I took it that this was silly. The first time I took it there were ONLY two questions. I guess you could say that if this is the gold standard of mental health assessment, we need help. And I am so grateful to people like Grace who are working on new ways to serve people in their darkest hours.

This week, I am also giving away copies of my book, It’s Okay to Feel Things Deeply. Comment here if you want a copy:

Next week, I would love to invite you to meet Gabrielle Blair and me for her Ejaculate Responsibly book tour!

IN PERSON.

Book Passage (SF Ferry Building). Free.

Thurs., March 9th • 6:00pm PT. More details here.

If you agree or not, Gabrielle Blair is interesting. I am SO excited to talk to her about the book and to you. Gabby also has a substack, if you wanted to check her out but are not ready to commit to reading a whole book.

Doing this is my favorite part of the week. Thanks for making it possible. Let me know how you are. Did you know that we grow most in the springtime?

XO, Carissa