ObjectiveTo review the progress of radiomics in the field of colorectal cancer in recent years and summarize its value in the imaging diagnosis of colorectal cancer.MethodsEighty English and seven Chinese articles were retrieved through PUBMED, OVID, CNKI, Weipu and Wanfang. The structure and content of these literatures were classified and analyzed.ResultsIn five studies predicting the preoperative stages of colorectal cancer based on CT radiomics, the area under curve (AUC) ranged from 0.736 to 0.817; in two studies predicting the preoperative stages of colorectal cancer based on MRI radiomics, the AUC were 0.87 and 0.827 respectively. In two studies about radiomics analysis for evaluation of pathological complete response to neoadjuvant chemoradiotherapy based on CT, the AUC were 0.79 and 0.72 respectively; in four studies about radiomics analysis for evaluation of pathological complete response to neoadjuvant chemoradiotherapy based on MRI, the AUC ranged from 0.84 to 0.979. In one study evaluating the sensitivity of neoadjuvant chemotherapy based on MRI radiomics, the AUC was 0.79. In one study predicting the postoperative survival rate based on MRI radiomics, the AUC value of the final model was 0.827. In one study, the accuracy of the model based on PET/CT radiomics in 4-year disease-free survival (DSS), progression-free survival (DFS) and overall survival (OS) were 0.87, 0.79 and 0.79 respectively.ConclusionAt present, radiomics has a valuable impact on preoperative staging, neoadjuvant therapy evaluation, and survival analysis of colorectal cancer.
Thyroid nodules are a common endocrine disorder, and their early detection and accurate diagnosis are crucial for the prevention of thyroid cancer. However, the highly heterogeneous morphology and boundaries of thyroid nodules pose significant challenges to their precise identification and classification. Traditional diagnostic approaches rely heavily on physicians’ experience, which increases the risk of misdiagnosis and missed diagnoses. With the rapid advancement of computer-aided diagnosis (CAD) technologies, applying deep learning algorithms to the analysis of thyroid nodule ultrasound images has shown great potential. This paper reviews the latest research progress on deep learning-based CAD methods for thyroid nodules, with a focus on their applications in image preprocessing, segmentation and classification. The advantages and limitations of current techniques are analyzed, and potential future directions are discussed. This review aims to highlight the potential of deep learning in thyroid nodule diagnosis and to provide a foundation for selecting feasible pathways for future clinical applications.