目的:我们旨在开发一种易于部署的人工智能(AI)驱动模型,用于快速预测尿液培养测试结果。
方法:我们使用了一个训练数据集(n=34,584个尿液样本)和两个单独的,看不见的测试集(n=10083和9289个样本)。比较了各种机器学习模型的诊断性能。预测参数包括尿液分析结果(试纸和流式细胞术),患者人口统计学(年龄和性别),和样品采集方法。
结果:尽管更复杂的模型在预测阳性培养物中获得了最高的AUC(最高:多层感知器(MLP),AUC为0.884,95%CI0.878-0.89),仅使用流式细胞术参数的多元逻辑回归(MLR)获得了非常好的AUC(0.858,95%CI0.852-0.865)。为了帮助解释,MLP和MLR模型的预测结果根据似然比(LR)对阳性进行分类:极不可能(LR0.1),不太可能(LR0.3),灰色区域(LR0.9),可能(LR5.0),很有可能(LR40)。这导致了17%,28%,32%,10%,13%的样本属于MLR模型的每个类别,20%,27%,31%,7%,MLP模型为16%。
结论:结论:这种稳健的模型有可能通过在大量样本中获得尿培养结果之前提供见解来协助临床医生的决策过程(~2/3).
OBJECTIVE: We aimed to develop an easily deployable artificial intelligence (AI)-driven model for rapid prediction of urine culture test results.
METHODS: We utilized a training dataset (n = 34,584 urine samples) and two separate, unseen test sets (n = 10,083 and 9,289 samples). Various machine learning models were compared for diagnostic performance. Predictive parameters included urinalysis results (dipstick and flow cytometry), patient demographics (age and gender), and sample collection method.
RESULTS: Although more complex models achieved the highest AUCs for predicting positive cultures (highest: multilayer perceptron (MLP) with AUC of 0.884, 95% CI 0.878-0.89), multiple logistic regression (MLR) using only flow cytometry parameters achieved a very good AUC (0.858, 95% CI 0.852-0.865). To aid interpretation, prediction results of the MLP and MLR models were categorized based on likelihood ratio (LR) for positivity: highly unlikely (LR 0.1), unlikely (LR 0.3), grey zone (LR 0.9), likely (LR 5.0), and highly likely (LR 40). This resulted in 17%, 28%, 34%, 9%, and 13% of samples falling into each respective category for the MLR model and 20%, 26%, 31%, 7%, and 16% for the MLP model.
CONCLUSIONS: In conclusion, this robust model has the potential to assist clinicians in their decision-making process by providing insights prior to the availability of urine culture results in a significant portion of samples (∼2/3rd).