Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.

Considering the great potential of iron chelators at inhibiting the proliferation of tumor cells, in order to determine the molecular and biological basis for the effects of iron chelator in oral cancer, we investigated the effects of iron chelator, desferrioxamine (DFO), on the gene profiling analysis of immortalized human oral keratinocytes (IHOK), and oral cancer cells (HN12), using the cDNA microarray. We identified 46 clones cDNA exhibiting more than 2 fold overexpression in DFO treated IHOK and HN12 cells, and 94 cDNA reveal more than 2 fold down-regulated expression. Examination of gene expression that differs between DFO treated vs. control IHOK and HN12 cells apprear to be related to : cell cycle regulator, cell growth and apoptosis, signal transduction and stress. p21 for cell cell cycle factor was upregualted, and cyclin-cdk gene was decreased expression, so we observed cell cycle arrest in DFO treated IHOK and HN12 cells. In tumor growth, we have identified downregulation of hemidesmosomal protein (bullous pemphigoid antigen 1) and epiregulin expression in DFO treated IHOK and oral cancer cells. Signal transducers including mitogen-activated protein kinase-activated protein kinase 5, serine/thereonine kinase 6 were downregulated with DFO treated cells, suggesting the DFO regulates the p38 MAP kianse pathway in immortalized and maignant oral keratincytes. In conclusion, we have demonstrated the high-throughput utility of cDNA array hybridization in parallel to the gene expression analysis to identify genes that are expressed differentially in DFO treated with immortalized and malignant oral keratinocytes. The differentially expressed genes identified here should be informative in DFO-induced anti-cancer effects.