Development and validation of a prediction model for postoperative intensive care unit admission in patients with non-cardiac surgery

Xu, Zhifeng; Yao, Shihua; Jiang, Zhonghe; Hu, Linhui; Huang, Zhihui; Zeng, Qun; Liu, Xueyan

doi:10.1016/j.hrtlng.2023.08.001

Cited by 2 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…When strictly comparing metrics such as F1 score, LLMs still underperform dedicated classification models utilizing tabular features. [45][46][47][48][49][50][51][52][53][54][55][56][57][58] Traditional machine learning models are rarely utilized in the clinical setting because of difficulty in interpreting a model's predictions. In contrast, LLMs present natural language explanations understandable to human clinicians, and can develop a rationale for each outcome variable of interest.…”

Section: Discussionmentioning

confidence: 99%

“…The LLM in this study exhibited very good performance at ASA-PS classification prediction, ICU admission prediction, and hospital mortality prediction. When strictly comparing metrics such as F1 score, LLMs still underperform dedicated classification models utilizing tabular features . Traditional machine learning models are rarely utilized in the clinical setting because of difficulty in interpreting a model’s predictions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication

Chung,

Fong,

Walters

et al. 2024

JAMA Surg

View full text Add to dashboard Cite

ImportanceGeneral-domain large language models may be able to perform risk stratification and predict postoperative outcome measures using a description of the procedure and a patient’s electronic health record notes.ObjectiveTo examine predictive performance on 8 different tasks: prediction of American Society of Anesthesiologists Physical Status (ASA-PS), hospital admission, intensive care unit (ICU) admission, unplanned admission, hospital mortality, postanesthesia care unit (PACU) phase 1 duration, hospital duration, and ICU duration.Design, Setting, and ParticipantsThis prognostic study included task-specific datasets constructed from 2 years of retrospective electronic health records data collected during routine clinical care. Case and note data were formatted into prompts and given to the large language model GPT-4 Turbo (OpenAI) to generate a prediction and explanation. The setting included a quaternary care center comprising 3 academic hospitals and affiliated clinics in a single metropolitan area. Patients who had a surgery or procedure with anesthesia and at least 1 clinician-written note filed in the electronic health record before surgery were included in the study. Data were analyzed from November to December 2023.ExposuresCompared original notes, note summaries, few-shot prompting, and chain-of-thought prompting strategies.Main Outcomes and MeasuresF1 score for binary and categorical outcomes. Mean absolute error for numerical duration outcomes.ResultsStudy results were measured on task-specific datasets, each with 1000 cases with the exception of unplanned admission, which had 949 cases, and hospital mortality, which had 576 cases. The best results for each task included an F1 score of 0.50 (95% CI, 0.47-0.53) for ASA-PS, 0.64 (95% CI, 0.61-0.67) for hospital admission, 0.81 (95% CI, 0.78-0.83) for ICU admission, 0.61 (95% CI, 0.58-0.64) for unplanned admission, and 0.86 (95% CI, 0.83-0.89) for hospital mortality prediction. Performance on duration prediction tasks was universally poor across all prompt strategies for which the large language model achieved a mean absolute error of 49 minutes (95% CI, 46-51 minutes) for PACU phase 1 duration, 4.5 days (95% CI, 4.2-5.0 days) for hospital duration, and 1.1 days (95% CI, 0.9-1.3 days) for ICU duration prediction.Conclusions and RelevanceCurrent general-domain large language models may assist clinicians in perioperative risk stratification on classification tasks but are inadequate for numerical duration predictions. Their ability to produce high-quality natural language explanations for the predictions may make them useful tools in clinical workflows and may be complementary to traditional risk prediction models.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication

Chung,

Fong,

Walters

et al. 2024

JAMA Surg

View full text Add to dashboard Cite

show abstract

“…Z. Xu и соавт. разработана и проверена модель, объединяющая пред-и интраоперационные переменные для прогнозирования поступления в ОИТ после операции [19]. Потенциальные предикторы включали возраст, класс по классификации ASA (American Society of Anesthesiologists), тип операции, неотложность операции, предоперационный уровень альбумина и мочевины, интраоперационное введение…”

Section: оценка риска - какие новые подходы?unclassified

“…У больных стабильной кристаллоидов, переливание крови и катетеризацию, а также время операции. Авторы создали онлайнкалькулятор для клинического применения, который может улучшить результаты лечения и помочь в распределении дорогостоящих и ограниченных ресурсов ОИТ [19].…”

Section: кардиальные осложнения при несердечных операцияхunclassified

Assessment and correction of the risk of cardiac complications during non-cardiac surgery

Sumin

2024

Racionalʹnaâ farmakoterapiâ v kardiologii

View full text Add to dashboard Cite

About 300 million non-cardiac operations are performed annually worldwide. In developed countries, an increasing number of such operations are performed in older patients with comorbid cardiac pathology. The leading cause of perioperative mortality is cardiovascular complications. The national and international guidelines published in the last year on cardiovascular assessment and correction of the risk of cardiac complications during non-cardiac operations contain several contradictory provisions. First of all, this refers to the methods of preoperative risk assessment: which is better — to base on scales, functional status assessment or biomarkers (BNP/ NT-pro-BNP) determination? Questions remain about the most appropriate algorithm for preoperative evaluation of patients with suspected cardiac pathology (the need for preoperative non-invasive tests or limited to biomarkers evaluation), some of these algorithms have been validated in recent studies. Several publications address the problem of drug therapy in non-cardiac surgery, both previously studied drugs (statins, beta-blockers) and new ones (colchicine, nicorandil). Perioperative myocardial injury syndrome and its possible drug prevention continue to be studied. This review considers recent publications on this topic, which have not yet been reflected in the guidelines and may influence their correction in the future. The article also discusses the problems of implementing existing guidelines in clinical practice. The publications of the last year presented in the review help researchers and practitioners to form their views on existing controversial issues. Ultimately, this will promote the use of an individualized approach in the management of this category of patients.

show abstract

Development and validation of a prediction model for postoperative intensive care unit admission in patients with non-cardiac surgery

Cited by 2 publications

References 31 publications

Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication

Large Language Model Capabilities in Perioperative Risk Prediction and Prognostication

Assessment and correction of the risk of cardiac complications during non-cardiac surgery

Contact Info

Product

Resources

About