Using Large Language Models to Create Personalized Networks From Therapy Sessions

TL;DR

The paper tackles the challenge of personalizing psychotherapy without relying on extensive longitudinal data by building an end-to-end pipeline that generates session-level personalized networks from therapy transcripts. It combines three stages—process detection with EEMM dimensions, two-step clustering into clinically meaningful themes, and ensemble-based, explainable links between themes—guided by expert feedback. Across a 77-session dataset, the approach yields clinically meaningful themes and connections, outperforming a direct prompting baseline in innovativeness, usefulness, and treatment-planning utility, with evaluators rating networks as clinically relevant and novel. The work demonstrates a scalable, bottom-up method for case conceptualization and treatment planning, while outlining limitations and promising directions for validating clinical impact and extending to dynamic, therapist-centric representations.

Abstract

Recent advances in psychotherapy have focused on treatment personalization, such as by selecting treatment modules based on personalized networks. However, estimating personalized networks typically requires intensive longitudinal data, which is not always feasible. A solution to facilitate scalability of network-driven treatment personalization is leveraging LLMs. In this study, we present an end-to-end pipeline for automatically generating client networks from 77 therapy transcripts to support case conceptualization and treatment planning. We annotated 3364 psychological processes and their corresponding dimensions in therapy transcripts. Using these data, we applied in-context learning to jointly identify psychological processes and their dimensions. The method achieved high performance even with a few training examples. To organize the processes into networks, we introduced a two-step method that grouped them into clinically meaningful clusters. We then generated explanation-augmented relationships between clusters. Experts found that networks produced by our multi-step approach outperformed those built with direct prompting for clinical utility and interpretability, with up to 90% preferring our approach. In addition, the networks were rated favorably by experts, with scores for clinical relevance, novelty, and usefulness ranging from 72-75%. Our findings provide a proof of concept for using LLMs to create clinically relevant networks from therapy transcripts. Advantages of our approach include bottom-up case conceptualization from client utterances in therapy sessions and identification of latent themes. Networks generated from our pipeline may be used in clinical settings and supervision and training. Future research should examine whether these networks improve treatment outcomes relative to other methods of treatment personalization, including statistically estimated networks.

Figures (22)

• Figure 1: Sample Network Generated From a Therapy Transcript
• Figure 2: Overview of Our Multi-Stage Pipeline
• Figure 3: LLM Performance on Process Detection and Dimension Assignment
• Figure 4: Results of the Evaluation Comparing the Three Ensemble Prompting Strategies: Prompt-Based, Temperature-Based, and Model-Based
• Figure 5: Comparison of Our Multi-Step Pipeline With the Direct Prompting Baseline