Predicting PANSS symptoms in schizophrenia spectrum disorders using speech only: an international, multi-centre, computational secondary analysis across multiple languages

Background Speech carries cues to variation in mental state in schizophrenia, typically indexed with clinician-rated scales such as Positive and Negative Syndrome Scale (PANSS). In this study, we aggregate multi-centre recordings to assemble a large corpus and assess symptom-prediction models at scale, to enable rater-independent, efficient, and reproducible assessments for early detection of relapse-related signals from speech. Methods In this computational secondary analysis, we compiled data from 453 patients with schizophrenia spectrum disorders, recruited between 2005 and 2026 from ten global sites in the Czech Republic, United States, Spain, Chile, France, Switzerland, the Netherlands, Türkiye, Germany, and China, and clipped their speech recordings into 6664 segments. Across three featuresets, acoustic-prosodic profile, pretrained multilingual embeddings, and their concatenation, we compared 16 algorithms in separate item-specific models to predict eight relapse-related PANSS items, including three positive (P1, P2, P3), three negative (N1, N4, N6), and two general (G5, G9) items, on speaker-disjoint splits. Performance was assessed by root-mean-squared-error (RMSE) at both segment and participant (median-aggregation) levels. Best model per item underwent bias checks for age, sex, education, and symptom severity. Findings Best-performing models predicted symptoms with prediction errors of 1.5 or lower: P1 1.544/1.527, P2 1.318/1.107, P3 1.487/1.542, N1 1.071/1.030, N4 1.521/1.430, N6 0.869/0.855, G5 0.951/0.882, G9 1.298/1.282 (segment/participant). Performance of pretrained embeddings surpassed acoustic-prosodic features and concatenation. Results were comparable in languages or language variants with fewer available speech and language computational resources. We found no bias by age, sex, or education, aside from reduced N4 accuracy in males; but performance degraded with higher symptom severity. Interpretation Speech can support automatic assessment of schizophrenia symptoms using pretrained embeddings, even without transcripts. Such models show promise as clinically meaningful, efficient, and low-burden tools for real-time monitoring of symptom trajectories. Future work should validate these models prospectively in higher-severity samples and improve clinical interpretability. Funding European Union Horizon.

Authors

Institutions

Publication Details

Journal
EClinicalMedicine
Published
2026-10-28
DOI
https://doi.org/10.1016/j.eclinm.2026.104265
Primary Topic
Schizophrenia research and treatment
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Predicting PANSS symptoms in schizophrenia spectrum disorders using speech only: an international, multi-centre, computational secondary analysis across multiple languages

Arndis Simonsen, Amandine Lecomte, José Tomás García Molina, Philipp Homan et al.
4 links
EClinicalMedicine
Schizophrenia research and treatment
article

Predicting PANSS symptoms in schizophrenia spectrum disorders using speech only: an international, multi-centre, computational secondary analysis across multiple languages

Arndis Simonsen, Amandine Lecomte, José Tomás García Molina, Philipp Homan, Filip Španiel, Katja Koelkebeck, Burcu Verim, Riccardo Fusaroli, Janna N. de Boer, Yuan Zhou, Sophie Barthélémy, Vincent Tourneur, Wolfram Hinzen, Iris E. Sommer, Renxiang Qiu, Berna Yalınçetin, Silvia Ciampelli, Cemal Demirlek, Alberto Parola, Alicia I. Figueroa-Barra, Sanne Koops, Michel Musiol, Sunny X. Tang, Maxime Amblard, Amandine Decker, Lena K. Palaniyappan, Rui He, Guy Gimenez, Maryia Kirdun, Bora Emre, Simran Bhola, Roya Melanie Hüppi, Sandy Yin, Rosa Ayesa Arriola, Lara Navarrete Orejudo, Claudio Palominos, Huiling Wang
article en

Abstract

Background Speech carries cues to variation in mental state in schizophrenia, typically indexed with clinician-rated scales such as Positive and Negative Syndrome Scale (PANSS). In this study, we aggregate multi-centre recordings to assemble a large corpus and assess symptom-prediction models at scale, to enable rater-independent, efficient, and reproducible assessments for early detection of relapse-related signals from speech. Methods In this computational secondary analysis, we compiled data from 453 patients with schizophrenia spectrum disorders, recruited between 2005 and 2026 from ten global sites in the Czech Republic, United States, Spain, Chile, France, Switzerland, the Netherlands, Türkiye, Germany, and China, and clipped their speech recordings into 6664 segments. Across three featuresets, acoustic-prosodic profile, pretrained multilingual embeddings, and their concatenation, we compared 16 algorithms in separate item-specific models to predict eight relapse-related PANSS items, including three positive (P1, P2, P3), three negative (N1, N4, N6), and two general (G5, G9) items, on speaker-disjoint splits. Performance was assessed by root-mean-squared-error (RMSE) at both segment and participant (median-aggregation) levels. Best model per item underwent bias checks for age, sex, education, and symptom severity. Findings Best-performing models predicted symptoms with prediction errors of 1.5 or lower: P1 1.544/1.527, P2 1.318/1.107, P3 1.487/1.542, N1 1.071/1.030, N4 1.521/1.430, N6 0.869/0.855, G5 0.951/0.882, G9 1.298/1.282 (segment/participant). Performance of pretrained embeddings surpassed acoustic-prosodic features and concatenation. Results were comparable in languages or language variants with fewer available speech and language computational resources. We found no bias by age, sex, or education, aside from reduced N4 accuracy in males; but performance degraded with higher symptom severity. Interpretation Speech can support automatic assessment of schizophrenia symptoms using pretrained embeddings, even without transcripts. Such models show promise as clinically meaningful, efficient, and low-burden tools for real-time monitoring of symptom trajectories. Future work should validate these models prospectively in higher-severity samples and improve clinical interpretability. Funding European Union Horizon.

EClinicalMedicineVol. 101
Institució Catalana de Recerca i Estudis Avançats (ES), University of Copenhagen (DK), Western University (CA), Feinstein Institute for Medical Research (US), University Medical Center Groningen (NL), Harvard University (US), University of Groningen (NL), Universitat Pompeu Fabra (ES), Dokuz Eylül University (TR), Chinese Academy of Sciences (CN), Bielefeld University (DE), University of Zurich (CH), Aarhus University (DK), Aix-Marseille Université (FR), Douglas Mental Health University Institute (CA), McLean Hospital (US), Instituto de Salud Carlos III (ES), Universidad Alberto Hurtado (CL), ETH Zurich (CH), Centro de Investigación Biomédica en Red de Salud Mental (ES), Instituto de Investigación Marqués de Valdecilla (ES), Max Planck Institute for Human Cognitive and Brain Sciences (DE), Institute of Psychology, Chinese Academy of Sciences (CN), Renmin Hospital of Wuhan University (CN), University of Chinese Academy of Sciences (CN), Zucker Hillside Hospital (US), National Institute of Mental Health (CZ), Robarts Research Institute (CA), Laboratoire de Psychologie Clinique, de Psychopathologie et de Psychanalyse (FR), McGill University (CA), University of Chile (CL), University of Pennsylvania (US), Université de Lorraine (FR)
Brain and Behavior Research Foundation, OPO-Stiftung, Ministerio de Ciencia, Tecnología, Conocimiento e Innovación, European Commission, Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, ZonMw, Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, China Scholarship Council, Instituto de Salud Carlos III, HORIZON EUROPE Reforming and enhancing the European Research and Innovation system
Good health and well-being
Openalex Percentile: Top 13%
Schizophrenia research and treatment
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon