TIPPS: Comparative Analysis of Preprocessing and Postprocessing Techniques for Tumor Images Across Different Modalities

Tumor image analysis is essential for diagnosing and treating cancer. This study conducts a comparative analysis of preprocessing and postprocessing techniques applied to tumor images from different modalities: CT scan, MRI, and histopathological images. It investigates how these techniques enhance image quality, improve feature extraction, and aid accurate tumor classification and diagnosis. Preprocessing involves techniques like denoising techniques, unsharp masking, contrast stretching, and contrast limited adaptive histogram equalization. These techniques are applied to original dataset and new enhanced version of datasets are produced then. Further, classification algorithms such as support vector machine, decision tree, random forest, and logistic regression, enable accurate tumor categorization based on extracted features. Visualization and interpretation methods help clinicians understand complex image data. Finally, voting classifier is applied over original and new enhanced version of produced datasets acting as the post-processing technique. Through evaluating various preprocessing and postprocessing method(s) across modalities, this study aims to optimize tumor image analysis for clinical use. The findings of this study can contribute to the development of robust and effective pipelines for tumor image analysis, ultimately supporting improved diagnosis, treatment planning, and patient care.