This project investigated the potential of generative AI models in aiding health sciences librarians with collection development. Researchers at Chapman University’s Harry and Diane Rinker Health Science campus evaluated four generative AI models—ChatGPT 4.0, Google Gemini, Perplexity, and Microsoft Copilot—over six months starting in March 2024. Two prompts were used: one to generate recent eBook titles in specific health sciences fields and another to identify subject gaps in the existing collection. The first prompt revealed inconsistencies across models, with Copilot and Perplexity providing sources but also inaccuracies. The second prompt yielded more useful results, with all models offering helpful analysis and accurate Library of Congress call numbers. The findings suggest that Large Language Models (LLMs) are not yet reliable as primary tools for collection development due to inaccuracies and hallucinations. However, they can serve as supplementary tools for analyzing subject coverage and identifying gaps in health sciences collections.

Background: Primary malignant brain tumors (PBT) impose substantial burdens on patients and caregivers. Caregivers are essential in the delivery of outpatient care for patients with PBT but experience high levels of fatigue, distress, and health decline. Although exercise is known to improve outcomes in cancer patients, interventions tailored specifically to the PBT patient–caregiver dyad remain limited. Dyadic intervention, as well as exercise oncology, are emerging areas of active research in neuro-oncology. This scoping review incorporates both principles to evaluate the existing literature on exercise interventions on primary brain tumor patient–caregiver dyads. Methods: We conducted a comprehensive search of MEDLINE (PubMed), Embase, CINAHL (EBSCO), Rehabilitation & Sports Medicine (EBSCO), and Cochrane Central (Ovid) in December 2025 for studies involving exercise interventions that included adult PBT patients and caregivers. Results: Of the 1126 records screened, eight studies were included: four yoga-based interventions (three feasibility trials and one ongoing multicenter RCT), one pilot ski-based intervention, and three aerobic and resistance training-based interventions (two qualitative and one ongoing trial). The interventions were safe and feasible, with high adherence and retention. The preliminary reported benefits included improvements in fatigue, sleep, quality of life, and caregiver distress for the dyads. Videoconference delivery was effective, particularly during the COVID-19 pandemic. The eight included studies comprised 5–67 dyads, with four being single-arm feasibility studies. Conclusions: Current literature on dyadic exercise intervention in neuro-oncology consists primarily of small-scale feasibility and pilot studies. Initial findings have demonstrated that such interventions are safe. However, preliminary efficacy remains limited due to the risk of bias and lack of statistical power. Larger randomized clinical trials with objective endpoints are needed to define efficacy and guide evidence-based protocols.