Identification of Neuro-Blastoma and Pediatric Cancer Using Time Series Analysis

Abstract

This chapter has studies that attempt to assess the detection of neuroblastoma and pediatric cancer via time series evaluation. The look took a retrospective method using five patient datasets, combining gadget learning and traditional statistical techniques to detect chance tiers of tumors inside the datasets. The machine-getting-to-know techniques include random forests, decision trees, assist vector machines, k-nearest friends, and convolutional neural nets. Statistical strategies used included co-occurrence matrix evaluation, principal element evaluation, and Shannon entropy. The observation outcomes demonstrate that time series analysis can offer high accuracy in predicting cancer threat levels and may be carried out on diverse cancers for well-timed and reliable prognosis. Eventually, the chapter offers various hints to enhance detection, incorporating extra statistics capabilities and designing a custom reference index.