Photo provided by: Jeomgwang
Below is an overview of “Risk stratification of interstitial cystitis/bladder pain syndrome (IC/BPS) using machine learning-based prediction” published in the April 2024 issue of the journal. Urology According to Lamb et al.
To enhance the diagnosis of interstitial cystitis/bladder pain syndrome (IC/BPS), researchers introduce a new approach that utilizes machine learning algorithms to develop an improved risk classification system for IC . Using a national crowdsourcing effort, researchers amassed a dataset of 1,264 urine samples. They included 536 samples from IC patients (513 women, 21 men, 2 unspecified) and 728 age-matched controls (318 women, 402 men, 8 unspecified). Accompanied by corresponding patient-reported outcome (PRO) pain and symptom scores.
In addition, 296 urine samples were collected from three academic centers, including 78 IC cases (71 women, 7 men) and 218 controls (148 women, 68 men, 2 unknown). did. Urinary cytokine biomarker levels were quantified using Luminex assay. The research group developed the Interstitial Cystitis Individual Inflammatory Symptoms (IC-PIS) score, a machine learning predictive classification model that integrates PROs and cytokine levels, and compared its performance with a challenger model. Results show that the best performing model utilizing biomarker measurements and PRO achieved an AUC of 0.87, outperforming PRO alone (AUC = 0.83).
The biomarker alone showed moderate predictive performance (AUC=0.58), but when combined with PRO, the predictive accuracy improved significantly. Therefore, the IC-PIS model represents a promising advance in IC/BPS diagnosis, providing a comprehensive approach that integrates both patient-reported symptoms and objective biomarkers. Additionally, this study highlights the robustness and scalability of study results through innovative sample collection logistics, including one of the largest crowdsourced biomarker development studies utilizing ambient delivery methods across the United States. doing.
sauce: sciencedirect.com/science/article/abs/pii/S0090429524002851
