Machine learning for analysis and fourier transform infrared spectroscopy characterization of titanium hydroxyapatite composite

With the development of artificial materials, their application has extended to all fields including human implants. These artificial materials are synthesized or prepared to interact with biological systems of the human body for the diagnosis or treatment of diseases, ailments, and are known as bio materials. The bio materials thus synthesized must possess characteristics like bio compatibility, mechanical stability and strength, inertness, and ease of fabrication. The research study on bio materials revealed that hydroxyapatite is the primary material that is generally used, which contains calcium phosphate that has been used in complex tissue reconstruction for more than six decades [1]. The mechanical properties of pure Hydroxyapatite are relatively low and are weak in load-bearing applications due to their brittle structure. Various studies [1,2] revealed that titanium hydroxyapatite can give excellent mechanical properties with required biological properties. Amongst various methods to prepare Titanium Hydroxyapatite, for current research the wet Precipitation method is adopted for mass production of Titanium Hydroxyapatite. This is with the reason that, the only by-product during mass production will be water alone, and in addition to that the process is simple and economical. By employing the wet precipitation method, titanium hydroxyapatite material prepared and tested for chemical as well as mechanical properties such as tensile strength, hardness, brittleness etc. The overall properties of the synthesized powders decide their effectiveness for bio-implant applications. Some of the important properties are particle size, specific surface area, purity, composition, particle-size distribution, particle morphology, and density. FTIR Testing is one of the important tests carried out to provide information about the presence of functional groups in the composites. Further, this test can assist in optimizing the formulations by evaluating the effects of different inputs on composites' chemical structure and properties for various applications. To explore this optimization of chemical combinations, there is a possibility to explore machine Learning methods to save time as well as to find avenues) for solving problems in different industries such as manufacturing, physics and chemical engineering, material Science & metallurgy. In Biomaterials, we can use machine learning to analyse data points of experimental testing to suggest outcomes. Various researchers have worked on the synthesis of Titanium hydroxyapatite and a research gap was identified, where no systematic approach is used in the synthesis of titanium-hydroxyapatite composite and its characterization. In this paper, actual FTIR testing results are used for the Regression Analysis tool in MATLAB using Machine learning to predict outcomes based on input features. The results revealed that machine learning has yielded in the optimization of the synthesis Process.