Exploratory data analysis of near-infrared spectrum from rats with biodegradable magnesium alloy implant

Abstract

Biodegradable Magnesium (Mg) based implants were recently developed to replace non-degradable metallic implants. The optical probe can capture variations in vivo from biological tissues using the near-infrared spectrum. This paper uses the spectrum from rats with biodegradable magnesium alloy "WE43" implants. This piolet study covers the connection between in vitro works (lab-based experiments) at OsloMet to support the in vivo (animal-based) experiments conducted by the MgSafe researcher's team in Italy. The lab experiments capture spectroscopic data over four days to study the combined effect of biodegradable magnesium implant disks with the living tissues/artificial medium. The primary challenge of in vivo data analysis is the complexity of the interaction of implant near interface summed up with the effect of light scattering and absorption in a biological medium. This paper verifies the feasibility of using the optical spectrum to relate the progress of implant surgery. Principal component analysis (PCA) is a popular exploratory data analysis technique. The principal component analysis helps to reduce dimensions and identify the significant features or wavelengths. The work relates the trend observed in the in vitro experiment compared to in vivo. The focus is on the first two weeks. PCA-based score plot identifies healthy rats from unhealthy ones. The preprocessing of optical data into derivative spectra highlights the wavelengths that influence unhealthy rats in the post-surgery period. The paper discusses biochemical reactions and chemical deposits on the surface of the in vitro implant to help relate the reason for the optical spectra changes on different days.