Natural speech algorithm applied to baseline interview data can predict which patients will respond to psilocybin for treatment-resistant depression.
Facundo Carrillo, Mariano Sigman, Diego Fernández Slezak, Philip Ashton, Lily Fitzgerald, Jack Stroud, David Nutt, Robin Carhart-Harris
Journal of Affective Disorders April 1, 2018 DOI: 10.1016/j.jad.2018.01.006 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Observational cohort with machine learning classification Peer reviewed |
|---|---|
| Sample size | 35 |
| Population | 17 patients with treatment-resistant depression and 18 untreated age-matched healthy control subjects |
| Interventions | Psilocybin psychological support |
| Dose | 10 mg and 25 mg, 7 days apart |
| Duration | 2 doses 7 days apart; baseline interview before treatment |
| Topics | Psychedelic-assisted therapy Depression Psilocybin |
| Keywords | Psilocybin therapy Psilocybin treatment Depression treatment Mental health treatment Speech biomarkers Speech analysis Vocal patterns Voice analysis Linguistic features Acoustic markers Speech features Treatment prediction Predictive analytics Outcome forecasting Prognosis Patient response Personalized medicine Precision medicine Algorithms for depression Digital psychiatry |
| Citations | 57 |
| Key findings | Machine learning applied to baseline speech differentiated depressed patients from healthy controls and identified psilocybin treatment responders from non-responders with 85% accuracy and 75% precision. The authors suggest natural language analysis could be a cost-effective screening tool for treatment suitability, but note the small sample requires replication. |
Abstract
Background: Natural speech analytics has seen some improvements over recent years, and this has opened a window for objective and quantitative diagnosis in psychiatry. Here, we used a machine learning algorithm applied to natural speech to ask whether language properties measured before psilocybin for treatment-resistant can predict for which patients it will be effective and for which it will not.
Methods: A baseline autobiographical memory interview was conducted and transcribed. Patients with treatment-resistant depression received 2 doses of psilocybin, 10 mg and 25 mg, 7 days apart. Psychological support was provided before, during and after all dosing sessions. Quantitative speech measures were applied to the interview data from 17 patients and 18 untreated age-matched healthy control subjects. A machine learning algorithm was used to classify between controls and patients and predict treatment response.
Results: Speech analytics and machine learning successfully differentiated depressed patients from healthy controls and identified treatment responders from non-responders with a significant level of 85% of accuracy (75% precision).
Conclusions: Automatic natural language analysis was used to predict effective response to treatment with psilocybin, suggesting that these tools offer a highly cost-effective facility for screening individuals for treatment suitability and sensitivity.
Limitations: The sample size was small and replication is required to strengthen inferences on these results.
Comparable studies
Other observational and cohort studies on psilocybin and psychedelic-assisted therapy, most cited first.
| Study | Year | Design | Participants |
|---|---|---|---|
| Acute subjective effects in LSD- and MDMA-assisted psychotherapy Patients with psychiatric disorders (posttraumatic stress disorder and major... | 2020 | Observational study | n = 18 |
| Predicting the outcome of psilocybin treatment for depression from baseline fMRI functional connectivity. Patients with treatment-resistant depression or moderate-to-severe major depression | 2024 | Observational cohort | n = 38 |
| The Promise of Therapeutic Psilocybin: An Evaluation of the 134 Clinical Trials, 54 Potential Indications, and 0 Marketing Approvals on ClinicalTrials.gov. Psilocybin clinical trials listed on ClinicalTrials.gov | 2024 | Observational study | n = 134 |
| Naturalistic psychedelic therapy: The role of relaxation and subjective drug effects in antidepressant response 28 PAT patients and 28 healthy volunteers | 2024 | Observational cohort with comparison group | n = 56 |
| Body mass index (BMI) does not predict responses to psilocybin | 2022 | Observational cohort |