hadar fisher

Post-doc | Harvard Medical School/McLean Hospital

[npj Digital Medicine @npjDigitalMed]

AI may detect depression from language alone.

Across 40,000+ text samples, AI models identified depression from language with ~80% accuracy, highlighting both the potential and the risks of language-based mental health screening.

https://t.co/4gncqVHtQa

[Yoni Ashar @YoniAshar]

NOW HIRING:

Clinical role as a VA pain psychologist, delivering PRT, EAET, and other neuroscience-based pain treatments!

And opportunities to work on grant-funded research trials

https://t.co/KUfJGMEHu7

[hadar fisher @hadarfisher]

An AI-generated podcast featuring our work. So cool!

[NPP-Digital Psychiatry and Neuroscience (DPN) @dpn_journal]

Our latest Deep Dive covers work by @hadarfisher @JustinBakerMD @DiegoPizzagalli @NigelJaffe @ChrisWebbPhD exploring whether smartphone passive sensors & LLMs can track daily activation in adolescents w/ anhedonia receiving behavioral activation therapy

https://t.co/tSlXOzX8As

[hadar fisher @hadarfisher]

We also identified important limitations:
• Overreliance on a single benchmark dataset
• Incomplete reporting of performance metrics
• Limited clinical integration

Huge thanks to
@ChrisWebbPhD , @NigelJaffe, Kristina Pidvirny, Anna Tierney, Mia Vaidean, and Poorvesh Dongre.

[hadar fisher @hadarfisher]

Now out in npj Digital Medicine 🎉

https://t.co/1fgQ7YD7yn

Our systematic review and meta-analysis examines how well language-based models detect depression from text.

We reviewed 123 studies (40,000 + observations) using NLP and machine learning.

[hadar fisher @hadarfisher]

Key findings:
• Overall accuracy ≈ 80%, but balanced accuracy ≈ 70% when accounting for class imbalance
• Text type matters: structured clinical interviews perform best
• Simple linguistic features + traditional ML often perform comparably to more complex transformer models

[hadar fisher @hadarfisher]

Grateful to see our paper included in this respected list. In this work, we showed that passive data from phone use, together with ChatGPT-based text analysis, can help clinicians track behavioral activation, a core treatment target in depression
🔗 https://t.co/0VHzmUl1eW

[Christian Webb @ChrisWebbPhD]

Honored to have our work recognized by @BBRFoundation, alongside many colleagues doing exciting work in the field

[hadar fisher @hadarfisher]

@ChrisWebbPhD Thanks Christian, appreciate it! 🙏 Couldn’t have done it without you and the team.

[hadar fisher @hadarfisher]

A huge thanks to @ChrisWebbPhD @NigelJaffe , Kristina Pidvirny, Anna Tierney, Mia Vaidean, and Poorvesh Dongre who went with me through all stages of meta-analysis grief from horror, despair and self-blame (what were we thinking?) all the way to hope and pride in this work.

[hadar fisher @hadarfisher]

💫 Excited to share new preprint a systematic review & meta-analysis of 123 studies (40k+ ppl) on how well language-based models detect depression from text.

https://t.co/uB0FLzKQgj

[hadar fisher @hadarfisher]

Language-based detection could make early screening more scalable and accessible, but how well do these tools actually work?

Our goal was to bring clarity to the field, synthesize evidence, and highlight what works & what needs work to build better tools for early detection.

[Shirley B. Wang @ShirleyBWang]

Thrilled to share a new paper in @JAMAPsych on path asymmetry in complex dynamic systems of psychopathology! https://t.co/3731z5Ghig

With amazing collaborators Tessa Blanken, Han van der Maas, & @BorsboomDenny 🥳

[hadar fisher @hadarfisher]

Huge thanks to @NigelJaffe @ChrisWebbPhD and our amazing team Habiballah Rahimi-Eichi, Erika Forbes, @DiegoPizzagalli and Justin Baker!

[hadar fisher @hadarfisher]

Now out in NPP – Digital Psychiatry and Neuroscience! 📱🧠

Can LLMs and smartphone sensing help us track therapy progress in real time?

[NPP-Digital Psychiatry and Neuroscience (DPN) @dpn_journal]

🆕 study by @hadarfisher @JustinBakerMD @DiegoPizzagalli @NigelJaffe @ChrisWebbPhD et al tested whether smartphone mobility sensing and LLM ratings of text can track behavioral activation in teens to boost their engagement in rewarding activities

https://t.co/jlVfMoxUbx

[hadar fisher @hadarfisher]

These patterns appeared at the individual level, showing the potential for personalized, real-time monitoring of therapy progress. In the future, such tools could help clinicians track clients’ progress outside therapy and deliver just-in-time interventions that improve treatment

[hadar fisher @hadarfisher]

All we’ve wanted and needed to hear for the past two years 🥹

[Eylon Levy @EylonALevy]

ALL THE SURVIVING HOSTAGES ARE FREE