Breast cancer (BC) is one of the most common types of cancer in women, and its prevalence is on the rise. The diagnosis of this disease in the first steps can be highly challenging. Hence, early and rapid diagnosis of this disease in its early stages increases the likelihood of a patient's recovery and survival. This study presents a systematic and detailed analysis of the various ML approaches and mechanisms employed during the BC diagnosis process. Further, this study provides a comprehensive and accurate overview of techniques, approaches, challenges, solutions, and important concepts related to this process in order to provide healthcare professionals and technologists with a deeper understanding of new screening and diagnostic tools and approaches, as well as identify new challenges and popular approaches in this field. Therefore, this study has attempted to provide a comprehensive taxonomy of applying ML techniques to BC diagnosis, focusing on the data obtained from the clinical methods diagnosis. The taxonomy presented in this study has two major components. Clinical diagnostic methods such as MRI, mammography, and hybrid methods are presented in the first part of the taxonomy. The second part involves implementing machine learning approaches such as neural networks (NN), deep learning (DL), and hybrid on the dataset in the first part. Then, the taxonomy will be analyzed based on implementing ML approaches in clinical diagnosis methods. The findings of the study demonstrated that the approaches based on NN and DL are the most accurate and widely used models for BC diagnosis compared to other diagnostic techniques, and accuracy (ACC), sensitivity (SEN), and specificity (SPE) are the most commonly used performance evaluation criteria. Additionally, factors such as the advantages and disadvantages of using machine learning techniques, as well as the objectives of each research, separately for ML technology and BC detection, as well as evaluation criteria, are discussed in this study. Lastly, this study provides an overview of open and unresolved issues related to using ML for BC diagnosis, along with a proposal to resolve each issue to assist researchers and healthcare professionals.

MicroRNAs (miRNAs) are a broad class of non-coding RNAs nearly 21 nucleotides length, which play crucial functions in posttranscriptional gene regulation. These molecules are associated with many developmental and cellular processes in eukaryotic organisms. Current investigation has reported major factors contributing to miRNA biogenesis and has constituted basic principles of miRNA function. More recently, it was confirmed that various miRNAs are clearly implicated in human malignancies, such as lung, breast, ovarian, bladder, colon cancer and other kinds of carcinoma. In addition, dysregulation in the miRNA machinery elements such as Dicer, Drosha, DGCR8, Argonaut, and TRBP could be involved in the progress of many tumor types. The purpose of the current review was to compile growing information besides how miRNA biogenesis and gene silencing are modified to develop cancer.