Cellular-Level Fluorescence Imaging for Intraoperative Margin Assessment in Breast Cancer

This study is a multicenter, prospective, self-controlled study. Basis for Sample Size Calculation: According to the results of the first phase, the accuracy of frozen pathology is 98.2%. Assuming that the detection interpretation results of ES are not inferior to intraoperative frozen pathology, a one-sided test will be conducted, with α=0.02, 1-β=0.8, and the non-inferiority margin is 0.02. The estimated required sample size is 593. Considering a 10% exclusion rate, the total sample size will be 659. Each patient is expected to have at least 4 samples taken, totaling 165 patients. Data Collection:1)Basic patient information: age, gender; 2)Tumor diagnosis (including imaging diagnosis): location, size, and staging; 3)Intraoperative collected information: surgery start time, surgery end time; 4)Training and testing records of the physicians in the ES interpretation group; 5)Interpretation records: a.ES start time, ES interpretation end time, ES interpretation results, ES image satisfaction; b.Frozen pathology submission time, report receipt time, report results; c.Paraffin pathology submission time, report time, and report results. The statistical analysis will be performed using SAS statistical software, version 9.4 or higher. All statistical tests will be two-sided. A P value less than or equal to 0.05 will be considered statistically significant. The description of quantitative indicators will include the mean, standard deviation, median, minimum, maximum, and interquartile range. The description of categorical indicators will use the number and percentage of each category. The comparison of general conditions between the two groups will be analyzed using appropriate methods based on the type of indicators. The consistency evaluation of quantitative data will use the Bland-Altman method and ICC.