The Extent to Which Artificial Intelligence Can Help Fulfill Metastatic Breast Cancer Patient Healthcare Needs: A...

Area of Science:

• Oncology
• Artificial Intelligence
• Digital Health

Background:

• Metastatic breast cancer (MBC) patients with HR+/HER2- subtypes require psychosocial and supportive care.
• Existing digital tools may not fully address the complex needs of these patients.

Purpose of the Study:

• To assess the impact of the Artificial Intelligence Patient Librarian (AIPL) on the ability of HR+/HER2- MBC patients to manage their advanced disease.
• To evaluate patient activation and user experience with the AIPL chatbot.

Main Methods:

• Phase 1: Adapted educational content for chatbot delivery and annotated online resources using a Convolutional Neural Network (CNN).
• Phase 2: Conducted pre- and post-surveys measuring Patient Activation Measure (PAM) and System Usability Scale (SUS) in 42 participants over two weeks.
• Phase 3: Explored user experiences through focus groups with 10 participants.

Main Results:

• No significant difference in PAM scores (pre: 59.33, post: 59.22) was observed, potentially due to high baseline activation.
• System Usability Scale (SUS) scores indicated good usability of the AIPL chatbot.
• Thematic analysis revealed AIPL provided basic wellness guidance but limited support for relationships and condition-specific information, and could not offer hope.

Conclusions:

• The AIPL demonstrated good usability and met basic information needs for MBC patients, especially those newly diagnosed.
• The chatbot's impact on patient activation was not significant, possibly due to high baseline levels.
• Future development will integrate a large language model (LLM) for enhanced, personalized patient support.