Chaves, Juan Manuel Zambrano

Abstract Critical findings in radiology reports are life threatening conditions that need to be communicated promptly to physicians (“critical findings”) for timely man-agement of patients. Flagging radiology reports of such incidents could facilitate opportune communication of critical findings. With advancements in natural language processing (NLP), large language models (LLMs) can be trained with task-specific instructions and examples to mine information from narrative texts. We believe that similar methods can be applied to radiology reports to identify and extract critical findings from these reports. However, due to the rarity of such critical events, there is a dearth of manually labeled datasets of critical findings in radiology reports. To overcome this limitation, we train instruction-tuned MISTRAL-based language models in a two-phase weakly supervised fine-tuning setup on unlabeled radiology reports from Mayo Clinic (n=15000). The weakly fine-tuned model is then used to automatically extract critical terms from both internal and external test datasets - Mayo Clinic (n=80) and MIMIC-III (n=123) 1 respectively against the expert-annotation. We also evaluated performance of the models on a large-scale MIMIC-IV reports (n=5000) using automated LLM-aided evaluation metrics - G-eval and Prometheus. We observe that, for both manual and LLM-based evaluations, weakly supervised fine-tuning improves model performance, showing successful task-specific alignment. For community use, we released the trained model with open-source academic license∗

Cervical spinal hemisection at C2 (SH) paralyzes the diaphragm muscle by interrupting descending bulbospinal pathways to phrenic motoneurons. Signaling via the neurotrophin receptor TrkB in phrenic motoneurons enhances functional recovery after SH. However, it is not clear if the delay in gene expression resulting from adeno‐associated virus (AAV)‐mediated delivery would blunt the therapeutic efficacy of AAV delivered after SH. The goal of the present study was to examine the efficacy of intrapleurally‐delivered AAV7 encoding the full‐length TrkB receptor at 3 days post‐SH. Chronic diaphragm EMG recordings were conducted in adult male Sprague‐Dawley rats confirming absence of ipsilateral activity 3 days after SH. The proportion of animals displaying recovery of ipsilateral diaphragm EMG activity increased over time post‐SH and was greater in those treated with AAV‐TrkB (by 14 days post‐SH: AAV‐TrkB: 10/15 vs. AAV‐GFP: 2/11; p=0.01). Furthermore, peak root‐mean‐square amplitude of the diaphragm EMG activity increased progressively over time post‐SH. During eupnea, animals displaying recovery of ipsilateral diaphragm EMG activity that were treated with AAV‐TrkB achieved 40% of pre‐injury values (vs. 10% in those treated with AAV‐GFP). Targeting phrenic motoneurons via intrapleural delivery of AAV7‐TrkB promotes recovery of diaphragm activity when administered 3 days post‐injury. Grant Funding Source : Supported by R01 HL96750, T32 HL105355 and Mayo Clinic Center for Regenerative Medicine