For the near-infrared (NIR) spectral analysis of the concentration of blood glucose, the calibration accuracy can be affected because of the existing of outlier samples. In this research, a Monte-Carlo cross validation (MCCV) method is constructed for eliminating outlier samples. The human blood plasma experiment in vitro and the human body experiment in vivo were introduced to evaluate the MCCV method for its application effect in NIR spectral analysis of blood glucose. And the uninformative sample elimination method based on modified uninformative variable elimination (MUVE-USE) was employed in this study for the comparison with MCCV. The results indicated that, like the MUVE-USE method, the outlier samples elimination method based on MCCV could be used to eliminate the outlier samples which came from gross errors (such as bad sample) or system errors (such as baseline drift). In addition, the outlier samples from the random errors of uncertain causes which affect model accuracy can be eliminated simultaneously by MCCV. The elimination of multiple outlier samples is beneficial to the improvement of prediction accuracy of calibration model.
A method was proposed to detect pulmonary nodules in low-dose computed tomography (CT) images by two-dimensional convolutional neural network under the condition of fine image preprocessing. Firstly, CT image preprocessing was carried out by image clipping, normalization and other algorithms. Then the positive samples were expanded to balance the number of positive and negative samples in convolutional neural network. Finally, the model with the best performance was obtained by training two-dimensional convolutional neural network and constantly optimizing network parameters. The model was evaluated in Lung Nodule Analysis 2016(LUNA16) dataset by means of five-fold cross validation, and each group's average model experiment results were obtained with the final accuracy of 92.3%, sensitivity of 92.1% and specificity of 92.6%.Compared with other existing automatic detection and classification methods for pulmonary nodules, all indexes were improved. Subsequently, the model perturbation experiment was carried out on this basis. The experimental results showed that the model is stable and has certain anti-interference ability, which could effectively identify pulmonary nodules and provide auxiliary diagnostic advice for early screening of lung cancer.