Happy Nice Mall

Since 2013 Ieso has focused on depression and generalized anxiety disorder and used data-driven techniques of which NLP is a core part. recovery rates for these conditions dramatically. According to Ieso, the recovery rate for depression in 2021 is 62%. 50% national average– and 73% for generalized anxiety disorder – compared to the national average of 58%.

Ieso says she focuses in part on anxiety and depression because these are two of the most common conditions. But they also respond better to CBT than others, such as obsessive-compulsive disorder. It’s not yet clear how much the clinic can improve its success, but it plans to focus on more conditions.

In theory, using AI to monitor quality allows clinicians to see more clients because better therapy means fewer unproductive sessions, but Ieso has yet to examine the direct impact of NLP on the efficiency of care.

“We’re currently able to treat 80 to 90 patients with 1,000 hours of therapy,” says Freer.

Unlike Ieso, Lyssn does not offer therapy itself. Instead, it provides its software to other clinics and universities in the UK and USA for quality control and training.

In the US, Lyssn’s customers include a telehealth opioid treatment program in California that wants to monitor the quality of care provided by its providers. The company is also working with the University of Pennsylvania to establish CBT therapists across Philadelphia with its technology.

In the UK, Lyssn works with three organisations, including Ieso, the Trent Psychological Therapies Service, an independent clinic commissioned by the NHS to provide mental health care. Trent PTS is still testing the software. Because the NLP model was manufactured in the USA, the clinic had to work with Lyssn to recognize British regional accents.

Dean Repper, director of clinical services at Trent PTS, believes the software can help therapists standardize best practices. “You would think for years that therapists who did this would get the best results,” she says. “But they’re not necessarily.” Repper compares this to driving: “When you learn to drive, you learn to do a number of safe things,” he says. “But after a while you stop doing some of these safe things and maybe get speeding tickets.”

Improve, not change

The purpose of artificial intelligence is to improve human care, not replace it. The lack of quality mental health care will not be resolved by short-term quick fixes. Addressing this issue will also require reducing stigma, increasing funding and improving education. Blackwell in particular denies most of the claims made for AI. “There is a dangerous amount of hype,” he says.

For example, there’s a lot of buzz by chatbot therapists and apps about things like 24-hour monitoring – often billed as Fitbits for the mind. But most of this technology falls somewhere between “years from now” and “never will be”.

“It’s not about welfare practices and things like that,” Blackwell says. “Putting an app in someone’s hands that says it will cure their depression probably only serves to inoculate them against seeking help.”

However, one problem with making psychotherapy more evidence-based is that it means asking therapists and clients to open up in private conversations. Will therapists object to such monitoring of their professional performance?

Repper expects some reluctance. “This technology represents a challenge for therapists,” he says. “They’re copying everything they say, as if someone else had entered the room for the first time.” To begin with, Trent PTS uses Lyssn’s software only with trainees waiting to be watched. Repper thinks that when these therapists are proficient, they might agree to watch because they’re used to it. More experienced therapists may need to be convinced of its benefits.

Former therapist Imel says it’s all about using technology as a support, not a stick. He thinks many would welcome the extra information. “It’s hard to be alone with your customers,” he says. “If all you do is sit in a private room with another person 20 or 30 hours a week without feedback from your colleagues, it can be really hard to improve.”

Freer agrees. At Ieso, therapists discuss AI-generated feedback with their supervisors. The idea is to allow therapists to take control of their professional development, to show them what they are good at – what other therapists can learn – and not so well – they might want to work on.

Ieso and Lyssn are just getting started, but there is clear potential for learning about therapy that is only revealed by mining large enough datasets. Atkins is a Meta-analysis published in 2018 This combined nearly 1,000 hours of therapy without the aid of AI. “Lyssn processes it in a day,” he says. New research published by both Ieso and Lyssn analyzes tens of thousands of sessions.

For example, in a 2019 paper in JAMA Psychiatry, Ieso researchers described a deep learning NLP model trained to categorize statements from therapists More than 90,000 hours of CBT sessions with approximately 14,000 clients. The algorithm learned to distinguish whether different phrases and short speech segments are examples of certain types of CBT-based conversations, such as checking the client’s mood, identifying and reviewing homework (where clients practice skills they learned in a session), discussing change methods. , planning for the future, etc.—or non-CBT-related conversations, such as general chat.