Automated, scalable assessments of negative symptoms (e.g. anhedonia, alogia), cognitive impairment (CIAS), and psychosocial functioning. Precision psychiatry delivered through a few minutes of speech.
ONGOING — We’re currently seeking partnership opportunities for additional validation in schizophrenia.
At a glance
100+
features extracted per recording
~2 minute
fast-paced task protocol option
The Schizophrenia
Measurement Problem
25M
People affected by schizophrenia worldwide
~70%
Of trial variance attributed to rater subjectivity
Zero
Approved biomarkers for schizophrenia
High
Screen failure rates due to poor stratification
Traditional cognitive batteries (like the MCCB) are exhausting. Schizophrenia patients often score poorly due to confounding phenomena such as avolition. Analyzing linguistic complexity through a set of speech-based tasks measures cognitive processing, and completes the CIAS assessment, without requiring grueling, sustained attention.
Negative symptoms, including anhedonia blunted affect, alogia, etc., are among the hardest endpoints in psychiatry. They are subjective to rate, slow to change, and poorly captured by clinician observation alone.
ki:elements extracts over 100 acoustic, prosodic, and linguistic features from a 2-minute task automatically. It can support exploratory clinical research and longitudinal assessment.