Sabir Shahwal, Israr Ayesha, Faheem Muhammad, Sani Ghulam Rasool, Khalid Aqsa, Bashir Sajid, Jabbar Tania, Jamil Yasir
Laser Spectroscopy Lab, Department of Physics, University of Agriculture, Faisalabad, Pakistan.
PINUM Cancer Hospital, Faisalabad, Pakistan.
Mikrochim Acta. 2025 Sep 11;192(10):660. doi: 10.1007/s00604-025-07517-y.
Early detection of cancer in low income countries is still a major health problem. This study discovers a novel diagnostic technique based on Nanoparticle-Enhanced Laser-Induced Breakdown Spectroscopy (NE-LIBS), which combines machine learning, nanotechnology, and laser-based elemental analysis as a potential early screening tool reported for the first time to our knowledge. We have explored that the incorporation of silver and copper oxide nanoparticles significantly enhances the intensity of emission signals of laser induced blood plasma, particularly of metallic biomarkers sodium and calcium, which have previously been known to reflect cancer-related metabolic changes. The use of advanced machine learning models to analyze these improved spectral features enables the accurate classification of cancerous and non-cancerous samples with an accuracy of nearly 95%. In low-income countries where conventional methods are still unavailable, machine learning assisted NE-LIBS has the potential to develop into a future platform that would enable scalable, reasonably priced initial clinical cancer screening when combined with machine learning.
在低收入国家，癌症的早期检测仍然是一个重大的健康问题。本研究发现了一种基于纳米颗粒增强激光诱导击穿光谱技术（NE-LIBS）的新型诊断技术，该技术首次将机器学习、纳米技术和基于激光的元素分析结合起来，作为一种潜在的早期筛查工具。据我们所知，我们已经探索发现，银和氧化铜纳米颗粒的加入显著增强了激光诱导血浆发射信号的强度，特别是金属生物标志物钠和钙的信号强度，此前已知这些生物标志物能反映癌症相关的代谢变化。使用先进的机器学习模型来分析这些改进后的光谱特征，能够以近95%的准确率对癌性和非癌性样本进行准确分类。在传统方法仍然无法获得的低收入国家，机器学习辅助的NE-LIBS有潜力发展成为一个未来平台，当与机器学习相结合时，能够实现可扩展、价格合理的初始临床癌症筛查。
