Automated detection of breast cancer in resected specimens with fluorescence lifetime imaging.

摘要: Re-excision rates for breast cancer lumpectomy procedures are currently nearly 25% due to surgeons relying on inaccurate or incomplete methods of evaluating specimen margins. The objective this study was determine if could be automatically detected in specimens from mastectomy and by a classification algorithm that incorporated parameters derived fluorescence lifetime imaging (FLIm). This generated database co-registered histologic sections FLIm data (N = 20) support vector machine (SVM) able detect cancerous, fibrous, adipose tissue. Classification accuracies were greater than 97% automated detection tissue specimens. worked equally well scanned hand with mechanical stage, demonstrating the system used during surgery excised ability technique simply discriminate between cancerous normal tissue, particular distinguish fibrous tumor, which is notoriously challenging optical techniques, leads conclusion has great potential assess Identification positive margins before waiting complete analysis significantly reduce re-excision rates.