Vallamchetla, SaiKrishna

e24196 Background: Radiation-induced edema and necrosis are common, dose-limiting toxicities. Corticosteroids remain first-line therapy but are associated with significant morbidity and limited long-term tolerability. Dietary and herbal agents with antioxidant, anti-inflammatory, and neurovascular protective properties are suggested as adjunct or steroid-sparing agents, yet systematic analysis of their overall efficacy is unclear. Methods: A systematic search of PubMed, Embase, Scopus, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov was conducted from inception through December 2025. Eligible studies included randomized trials and observational studies evaluating non-traditional dietary or herbal interventions for radiation-induced cerebral edema or injury. The primary outcome was radiographic improvement of cerebral edema on MRI, harmonized as any MRI-documented reduction or resolution of edema or necrosis. A random-effects comparative meta-analysis was performed for controlled studies with extractable binary outcomes, excluding case reports. A secondary random-effects, single-arm meta-analysis of proportions was conducted for Boswellia serrata , including observational cohorts and case series. Studies without MRI-based cerebral toxicity endpoints were synthesized qualitatively. Results: Eleven studies encompassing more than 400 patients met the inclusion criteria. Boswellia serrata was the most frequently evaluated intervention, followed by dl-3-n-butylphthalide, vitamin E, omega-3 fatty acids, melatonin, multi-herbal formulations, and integrative multi-compound regimens. In a comparative meta-analysis of two controlled studies, dietary and herbal interventions were associated with a significantly higher likelihood of radiographic improvement in cerebral edema compared with standard therapy or placebo (pooled risk ratio 1.60, 95% CI 1.12–2.30; I² = 0%). In a secondary, single-arm, meta-analysis of five Boswellia serrata studies, the pooled radiographic response rate was 56.2% (95% CI 45.0%–67.1%). Across other studies, neurologic and cognitive improvement and steroid-sparing effects were commonly reported, while serious treatment-related adverse events were uncommon. Integrative oncology approaches combining Boswellia serrata with other natural compounds were also associated with improved survival and reduced steroid dependence in longitudinal observational studies. Conclusions: Dietary and herbal interventions, particularly Boswellia serrata , are associated with meaningful radiographic and clinical improvement in radiation-induced cerebral edema and brain injury, with favorable safety profiles. These findings support further prospective randomized trials and suggest a potential adjunctive role for non-pharmacologic therapies in managing radiation-related neurotoxicity.

Abstract Introduction While Large Language Models (LLMs) offer generative capabilities, generic models often lack domain-specific grounding, leading to hallucinations in high-stakes contexts. However, LLMs can serve as clinical decision support tools if augmented with specific domain data. This study evaluates the utility of Retrieval-Augmented Generation (RAG) applied to open-source LLMs, utilizing a curated knowledge base of authoritative sleep medicine textbooks, to quantify improvement in performance. Methods We evaluated four open-source models—Llama-70B, Llama-8B, Qwen-14B, and Qwen-235B, linked to a RAG system indexing five standard textbooks, including ICSD-3 TR. Models were tested on sleep board–style MCQs and clinical vignettes diagnostic accuracy. Performance was assessed across nine configurations derived from three factors: RAG pipeline complexity (Plain vs. Complicated), textbook preprocessing (Uncleaned vs. Cleaned/TOC-aligned), and retriever type (Dense-only vs. Hybrid). Metrics included MCQ Accuracy and the rates at which the correct diagnosis appeared as the top result or within the top 5 differential diagnosis lists for vignettes. Results RAG consistently outperformed the no-RAG baseline, yielding absolute performance gains of 5.6%–10.7% for MCQs and 8.1%–10.2% for the top diagnosis in cases across all models. For MCQs, Qwen-235B achieved the highest accuracy of 87.3% (vs. 81.7% baseline) using the Plain RAG configuration with Cleaned (TOC-aligned) Hybrid retrieval. Llama-70B followed with 83.6% in the same configuration. For case vignettes, Qwen-235B achieved a correct top diagnosis rate of 66.3% and a correct diagnosis rate within the top 5 differential diagnosis lists of 90.8% using Cleaned Hybrid retrieval configurations. Hybrid retrieval (Dense + Sparse) consistently surpassed Dense-only methods. Notably, while larger models benefited from Cleaned text, smaller models (e.g., Qwen-14B) achieved higher top diagnosis rates with Uncleaned text (57.1% vs. 51.0%), suggesting a dependency on redundant context in lower-parameter settings. Conclusion RAG significantly enhances knowledge performance of open-source LLMs in sleep medicine. Hybrid retrieval and curated, TOC-aligned knowledge bases yield optimal results for large models, whereas smaller models benefit from the redundancy of uncleaned text. These findings suggest that RAG systems can augment LLMs in domain specific knowledge for clinical decision support. Support (if any)