Radiomics is an emerging field in quantitative imaging that uses advanced imaging features to objectively and quantitatively describe tumour phenotypes. It has potential for predicting treatment outcomes and cancer genetics, with applications in personalized medicine. This review discusses the applications and challenges of radiomics, including its use in predicting treatment response, tumour staging, tissue identification, and cancer genetics. Radiomic features, extracted from medical images using advanced algorithms, can provide objective information about tumour characteristics that may not be visible to the naked eye. These features have shown promise in differentiating between benign and malignant tumours, predicting treatment outcomes, and assessing tumour heterogeneity. However, radiomic feature quantification is sensitive to technical factors such as acquisition modes, reconstruction parameters, and image discretization schemes. Additionally, the reproducibility of radiomic features is important for reliable results. Studies have shown that some radiomic features are more stable than others, and that the number of features used in a study can affect the results. Proper study design, including the use of external validation datasets and statistical correction methods, is essential to avoid false positive results. Despite its potential, radiomics faces challenges such as the need for standardization and the impact of technical factors on feature quantification. Future research should focus on improving the reliability and reproducibility of radiomic features to enhance their clinical utility.Radiomics is an emerging field in quantitative imaging that uses advanced imaging features to objectively and quantitatively describe tumour phenotypes. It has potential for predicting treatment outcomes and cancer genetics, with applications in personalized medicine. This review discusses the applications and challenges of radiomics, including its use in predicting treatment response, tumour staging, tissue identification, and cancer genetics. Radiomic features, extracted from medical images using advanced algorithms, can provide objective information about tumour characteristics that may not be visible to the naked eye. These features have shown promise in differentiating between benign and malignant tumours, predicting treatment outcomes, and assessing tumour heterogeneity. However, radiomic feature quantification is sensitive to technical factors such as acquisition modes, reconstruction parameters, and image discretization schemes. Additionally, the reproducibility of radiomic features is important for reliable results. Studies have shown that some radiomic features are more stable than others, and that the number of features used in a study can affect the results. Proper study design, including the use of external validation datasets and statistical correction methods, is essential to avoid false positive results. Despite its potential, radiomics faces challenges such as the need for standardization and the impact of technical factors on feature quantification. Future research should focus on improving the reliability and reproducibility of radiomic features to enhance their clinical utility.