基于Transformer的前列腺癌血清拉曼光谱辅助诊断算法  

Transformer-based Assisted Diagnosis Algorithm of Prostate Cancer

作　　者：刘昌海 张恒[1] 邵晓光[2] 潘家骅[2] 陆小锋[1] 刘书朋[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200444 [2]上海交通大学医学院附属仁济医院,上海200127

出　　处：《工业控制计算机》2025年第2期68-69,72,共3页Industrial Control Computer

摘　　要：前列腺癌是全球范围内男性高发的恶性肿瘤之一,提高前列腺癌患者的早期诊断准确性对于提升治愈率和存活率至关重要。拉曼光谱作为肿瘤非侵入式诊断研究领域的一项前沿技术,已经显示出在区分癌症病人与正常人的多种生物样本方面具备很大潜力。设计了基于Transformer的前列腺癌血清拉曼光谱辅助诊断算法,旨在以克服传统算法的局限性,提高前列腺癌辅助诊断的准确性。该算法将拉曼光谱数据点分割成块,并使用Transformer编码捕捉这些块之间的内部关联,能够有效区分早期前列腺癌患者和前列腺增生患者的血清拉曼光谱,且经过对比实验分析,准确率优于其他常用算法模型,为非侵入式前列腺癌早期诊断提供一种更可靠的方法。In this paper,a prostate cancer serum Raman spectroscopy-assisted diagnostic algorithm based on Transformer is designed with the aim of overcoming the limitations of traditional algorithms and improving the accuracy of prostate cancer assisted diagnosis.This algorithm segments Raman spectroscopy data points into blocks and uses Transformer encoding to capture internal correlations between these blocks,effectively distinguishing serum Raman spectra from early-stage prostate cancer patients and benign prostatic hyperplasia patients.Through comparative experimental analysis,it was found that the accuracy of this algorithm is superior to other commonly used algorithm models,providing a more reliable method for non-invasive early diagnosis of prostate cancer.

关 键 词：前列腺癌 拉曼光谱 TRANSFORMER 早期诊断 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程] R737.25[医药卫生—肿瘤]