This observational study evaluates the accuracy of the Okaya AI platform in detecting fatigue and depression in cardiology patients, comparing its assessments to PHQ-9 and Fatigue Assessment Scale scores.

Patients frequently experience fatigue and depression, which are often underdiagnosed due to limitations in traditional screening tools. This study introduces the Okaya platform, a browser-based AI system that analyzes facial and vocal biomarkers collected during conversational check-ins. The platform uses computer vision and natural language processing to extract features such as eye contact, facial affect, pitch, volume, and speech patterns. These features are processed through regression models to generate a composite AI based score. The study aims to validate this score against PHQ-9 and FAS assessments. Participants will complete a single baseline check-in using the Okaya platform and complete standard questionnaires. No interventions will be provided.

Inclusion Criteria:

• Age ≥18, English-speaking, able to consent

Exclusion Criteria:

• Active substance use, nonverbal, cognitive disability, active suicidal/homicidal ideation

The participants will be selected from both inpatient and outpatient setting.