Development and application of an optimised Bayesian shrinkage prior for spectroscopic biomedical diagnostics.
Development and application of an optimised Bayesian shrinkage prior for spectroscopic biomedical diagnostics.
复制标题
用于光谱生物医学诊断的优化贝叶斯收缩先验的开发和应用。
DOI：
10.1016/j.cmpb.2024.108014
复制
发表时间：
2024
影响因子：
6.1
通讯作者：
Chu HO
中科院分区：
文献类型：
--
作者：
Chu HO
背景与目的对于含有相似分子键的生物物质，振动光谱的分类往往具有挑战性，干扰光谱输出。为了解决这个问题，各种方法被广泛研究。然而，虽然提供了强大的估计，这些技术是计算广泛的，经常过拟合的数据。收缩先验，有利于模型与相对较少的预测变量，经常被应用在贝叶斯惩罚技术，以避免overfitting.MethodsUsing的logit-normal连续模拟的穗和板（LN-CASS）作为收缩先验建模，我们已经建立了分类准确的分析，与建立的系统被发现比传统的最小绝对收缩和选择算子，马蹄形或钉板形。基于线性回归模型和密度泛函理论计算产生的振动光谱对系数数据进行了检验。然后应用于拉曼光谱从唾液中分类sex.ResultsSubsequently应用于从唾液中获得的光谱，评估模型表现出高精度（AUC> 90%），即使参数的数量高于观察的数量。交叉验证后，所有贝叶斯模型的光谱分析产生了高的分类精度。此外，对于唾液传感，LN-CASS被发现是唯一的分类与100%的准确度在预测的输出的基础上留一出crossvalidation.ConclusionsWith潜在的应用，在帮助诊断从小光谱数据集和兼容的光谱数据格式的范围。如IR和拉曼光谱的分类所示。这些结果是非常有前途的新兴发展的生物医学诊断传感系统的光谱平台。
Background and objectiveClassification of vibrational spectra is often challenging for biological substances containing similar molecular bonds, interfering with spectral outputs. To address this, various approaches are widely studied. However, whilst providing powerful estimations, these techniques are computationally extensive and frequently overfit the data. Shrinkage priors, which favour models with relatively few predictor variables, are often applied in Bayesian penalisation techniques to avoid overfitting.MethodsUsing the logit-normal continuous analogue of the spike-and-slab (LN–CASS) as the shrinkagepriorand modelling, we have established classification for accurate analysis, with the established system found to be faster than conventional least absolute shrinkage and selection operator, horseshoe or spike-and-slab. These were examinedversuscoefficient data based on a linear regression model and vibrational spectra producedviadensity functional theory calculations. Then applied to Raman spectra from saliva to classify the sample sex.ResultsSubsequently applied to the acquired spectra from saliva, the evaluated models exhibited high accuracy (AUC>90 %) even when number of parameters was higher than the number of observations. Analyses of spectra for all Bayesian models yielded high-classification accuracy upon cross-validation. Further, for saliva sensing, LN–CASS was found to be the only classifier with 100 %-accuracy in predicting the output based on a leave-one-out cross validation.ConclusionsWith potential applications in aiding diagnosis from small spectroscopic datasets and are compatible with a range of spectroscopic data formats. As seen with the classification of IR and Raman spectra. These results are highly promising for emerging developments of spectroscopic platforms for biomedical diagnostic sensing systems.