Trivedi, Shubham

Education-level or socioeconomic background of patients may dictate their ability to understand medical jargon. Inability to understand primary findings from a radiology report may lead to unnecessary anxiety among patients or missed follow up. We aim to meet this challenge by developing a patient-sensitive summarization model for radiology reports. We selected computed tomography (CT) exams of chest as a use-case and collected 7,000 studies from Mayo Clinic. Summarization model was built on top of the T5 large language model (LLM) as our experiments indicated that its text-to-text transfer architecture was suited for abstractive text summarization, resulting in a model with 0.77B trainable parameters. Noisy ground truth for model training was collected by prompting LLaMA-13B model. We recruited experts (board-certified radiologists) and laymen to manually evaluate model-generated summaries generated by model. Our model rarely missed information as marked by majority opinion of radiologists. Laymen indicated 63% improvement in their understanding by reading model-generated layman summaries. Comparison with zero-shot performance of ChatGPT indicated that the proposed model reduced the rate of hallucination by half and rate of missing important information by fivefold. The proposed model can generate reliable summaries for radiology reports understandable by patients with vastly different levels of medical knowledge.

BACKGROUND While most hypertriglyceridemia is asymptomatic, hypertriglyceridemia-associated acute pancreatitis (HTG-AP) can be more severe than AP of other etiologies. The reasons underlying this are unclear. We thus examined whether lipolytic generation of nonesterified fatty acids (NEFAs) from circulating triglycerides (TGs) could worsen clinical outcomes. METHODS Admission serum TGs, NEFA composition, and concentrations were analyzed prospectively for 269 patients with AP. These parameters, demographics, and clinical outcomes were compared between HTG-AP (TGs >500 mg/dL; American Heart Association [AHA] 2018 guidelines) and AP of other etiologies. Serum NEFAs were correlated with serum TG fatty acids (TGFAs) alone and with the product of TGFA serum lipase (NEFAs – TGFAs × lipase). Studies in mice and rats were conducted to understand the role of HTG lipolysis in organ failure and to interpret the NEFA-TGFA correlations. RESULTS Patients with HTG-AP had higher serum NEFA and TG levels and more severe AP (19% vs. 7%; P < 0.03) than did individuals with AP of other etiologies. Correlations of long-chain unsaturated NEFAs with corresponding TGFAs increased with TG concentrations up to 500 mg/dL and declined thereafter. However, NEFA – TGFA × lipase correlations became stronger with TGs above 500 mg/dL. AP and intravenous lipase infusion in rodents caused lipolysis of circulating TGs to NEFAs. This led to multisystem organ failure, which was prevented by pancreatic TG lipase deletion or lipase inhibition. CONCLUSIONS HTG-AP is made severe by the NEFAs generated from intravascular lipolysis of circulating TGs. Strategies that prevent TG lipolysis may be effective in improving clinical outcomes for patients with HTG-AP. FUNDING National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK, NIH) (RO1DK092460 and R01DK119646); Department of Defense (PR191945 under W81XWH-20-1-0400); National Institute on Alcohol Abuse and Alcoholism (NIAAA), NIH (R01AA031257).

PURPOSE Patient-centered outcomes (PCOs) are pivotal in cancer treatment, as they directly reflect patients' quality of life. Although multiple studies suggest that factors affecting breast cancer–related morbidity and survival are influenced by treatment side effects and adherence to long-term treatment, such data are generally only available on a smaller scale or from a single center. The primary challenge with collecting these data is that the outcomes are captured as free text in clinical narratives written by clinicians. MATERIALS AND METHODS Given the complexity of PCO documentation in these narratives, computerized methods are necessary to unlock the wealth of information buried in unstructured text notes that often document PCOs. Inspired by the success of large language models (LLMs), we examined the adaptability of three LLMs, GPT-2, BioGPT, and PMC-LLaMA, on PCO tasks across three institutions, Mayo Clinic, Emory University Hospital, and Stanford University. We developed an open-source framework for fine-tuning LLM that can directly extract the five different categories of PCO from the clinic notes. RESULTS We found that these LLMs without fine-tuning (zero-shot) struggle with challenging PCO extraction tasks, displaying almost random performance, even with some task-specific examples (few-shot learning). The performance of our fine-tuned, task-specific models is notably superior compared with their non–fine-tuned LLM models. Moreover, the fine-tuned GPT-2 model has demonstrated a significantly better performance than the other two larger LLMs. CONCLUSION Our discovery indicates that although LLMs serve as effective general-purpose models for tasks across various domains, they require fine-tuning when applied to the clinician domain. Our proposed approach has the potential to lead more efficient, adaptable models for PCO information extraction, reducing reliance on extensive computational resources while still delivering superior performance for specific tasks.

Abstract Objective To develop a natural language processing (NLP) algorithm that can accurately extract headache frequency from free‐text clinical notes. Background Headache frequency, defined as the number of days with any headache in a month (or 4 weeks), remains a key parameter in the evaluation of treatment response to migraine preventive medications. However, due to the variations and inconsistencies in documentation by clinicians, significant challenges exist to accurately extract headache frequency from the electronic health record (EHR) by traditional NLP algorithms. Methods This was a retrospective cross‐sectional study with patients identified from two tertiary headache referral centers, Mayo Clinic Arizona and Mayo Clinic Rochester. All neurology consultation notes written by 15 specialized clinicians (11 headache specialists and 4 nurse practitioners) between 2012 and 2022 were extracted and 1915 notes were used for model fine‐tuning (90%) and testing (10%). We employed four different NLP frameworks: (1) ClinicalBERT (Bidirectional Encoder Representations from Transformers) regression model, (2) Generative Pre‐Trained Transformer‐2 (GPT‐2) Question Answering (QA) model zero‐shot, (3) GPT‐2 QA model few‐shot training fine‐tuned on clinical notes, and (4) GPT‐2 generative model few‐shot training fine‐tuned on clinical notes to generate the answer by considering the context of included text. Results The mean (standard deviation) headache frequency of our training and testing datasets were 13.4 (10.9) and 14.4 (11.2), respectively. The GPT‐2 generative model was the best‐performing model with an accuracy of 0.92 (0.91, 0.93, 95% confidence interval [CI]) and R 2 score of 0.89 (0.87, 0.90, 95% CI), and all GPT‐2–based models outperformed the ClinicalBERT model in terms of exact matching accuracy. Although the ClinicalBERT regression model had the lowest accuracy of 0.27 (0.26, 0.28), it demonstrated a high R 2 score of 0.88 (0.85, 0.89), suggesting the ClinicalBERT model can reasonably predict the headache frequency within a range of ≤ ± 3 days, and the R 2 score was higher than the GPT‐2 QA zero‐shot model or GPT‐2 QA model few‐shot training fine‐tuned model. Conclusion We developed a robust information extraction model based on a state‐of‐the‐art large language model, a GPT‐2 generative model that can extract headache frequency from EHR free‐text clinical notes with high accuracy and R 2 score. It overcame several challenges related to different ways clinicians document headache frequency that were not easily achieved by traditional NLP models. We also showed that GPT‐2–based frameworks outperformed ClinicalBERT in terms of accuracy in extracting headache frequency from clinical notes. To facilitate research in the field, we released the GPT‐2 generative model and inference code with open‐source license of community use in GitHub. Additional fine‐tuning of the algorithm might be required when applied to different health‐care systems for various clinical use cases.

OBJECTIVES: The pathophysiology of delirium is complex and incompletely understood. Inflammation is hypothesized to be integral to its development due to effects on blood brain barrier integrity, facilitation of leukocyte extravasation into brain parenchyma, and propagation of neuroinflammation. Septic shock is the prototypical condition associated with ICU delirium; however, the relative contribution of resultant hypotension and systemic inflammation to the development of delirium is unknown. DESIGN: This was a prospective exploratory study. SETTING: A multidisciplinary ICU at an academic medical center in Phoenix, AZ. PATIENTS: Critically ill patients older than or equal to 18 years old admitted to the ICU. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Screening for delirium was performed using the Confusion Assessment Method for the ICU tool. The levels of C-C motif ligand 2 (CCL2), C-C motif ligand 3, C-X-C motif chemokine ligand 1, C-X-C motif chemokine ligand 10, and interleukin-8 were measured in serum samples obtained within 12 hours of ICU admission. Univariate and multivariate analyses were performed to assess the association of delirium with patient data pertaining to hospital course, laboratory values, vital signs, medication administration, and levels of the aforementioned chemokines. Forty-one of 119 patients (34.5%) in the study cohort developed ICU delirium. Each chemokine studied was associated with delirium on univariate analyses; however, CCL2 was the only chemokine found to be independently associated with the development of delirium on multivariable analysis. The association of increased CCL2 levels with delirium remained robust in various models controlling for age, presence of shock, Sequential Organ Failure Assessment score, Acute Physiology and Chronic Health Evaluation IV score, mean arterial pressure at presentation, lowest mean arterial pressure, and total opioid, midazolam, propofol, and dexmedetomidine exposure. CONCLUSIONS: The demonstrated relationship between CCL2 and delirium suggests this chemokine may play a role in the development of delirium and warrants further investigation.