A Multimodal Clinical Framework for Mental Health Assessment Integrating Psychometric and Emotion-Based Textual Features

Screening methods for depression and anxiety, which use static self-reported data as their primary assessment method, encounter difficulties in identifying these two mental health conditions because they cannot observe how people express their emotions over time. The research introduces a hybrid framework that combines clinically validated psychometric assessments with sentiment-based analysis of user-generated textual data. The study involved participants who completed questionnaires and wrote reflections on their emotions. The team used natural language processing techniques to extract sentiment and emotion features, which they then studied together with psychometric results. The study results show that there exists a strong inverse relationship between sentiment scores and clinical measurements, which indicates that people with higher depression and anxiety levels present their emotions in a more negative manner. The study demonstrated that the hybrid solution achieved better results than both psychometric testing and sentiment analysis methods, which required separate assessments. The study results demonstrate that structured clinical assessment, combined with dynamic emotional indicators, yields better detection of mental health than other assessment methods. The proposed framework helps connect traditional clinical methods with contemporary computational techniques by establishing a basis for mental health screening systems that require greater sensitivity, contextual understanding, and scalability. The upcoming research will concentrate on validating large scale implementations, while advanced language models will be used to enhance clinical functions and system stability.