Evaluation of Breast Cancer using Artificial Intelligence in various modalities

Breast cancer screening has been shown to significantly reduce mortality in women. The increasinguse of screening tests has increased the demand for fast and accurate diagnostic reports.Mammography her screening for breast cancer has been introduced in many countries in the last۳۰ years. Initially using an analog screen film based system, over the last ۲۰ years they haveswitched to using a fully digital system. AI has become a very useful tool in the field of cancercare. It has shown remarkable results and has the potential to change all current therapies. Bothimaging and computational achievements have increased the potential use of artificial intelligence(AI) in numerous breast imaging errands and in computer-aided discovery, including diagnosis,prognosis, treatment response, and risk assessment. It extends beyond its current use. Breast cancerscreening has significantly reduced mortality. AI has great potential to contribute to workplaceefficiency, results evaluation, and breast imaging quality measurement. Some algorithms usedclinically are still under development. Deep learning (DL) itself is a branch of ML that focuses onrepresenting data in the best possible way to simplify the learning task. The incorporation of AIinto the screening methods such as the examination of biopsy slides enhances the treatment successrate. There has been an increased interest in this area over recent years, and the field seems tohave a very promising future. In modern breast imaging centers, full-field digital mammography(FFDM) has substituted traditional analog mammography, and this has opened new chances fordeveloping computational outlines to mechanize detection and diagnosis. With the overview ofdigitization, the computer clarification of images has been a subject of penetrating interest, causingin the overview of computer-aided detection (CADe) and diagnosis (CADx) procedures in theearly ۲۰۰۰’s.The automatic competences of AI proposal the potential to improve the diagnosticskill of clinicians, including precise separation of tumor volume, removal of characteristic cancerphenotypes, conversion of tumoral phenotype topographies to clinical genotype insinuations, andrisk forecast. Traditional CAD systems in mammography screening have shadowed a rules-basedmethod, joining area information into hand-crafted topographies before using classical machinelearning techniques as a classifier. Finally, AI plays a role in image post-processing and qualityanalysis, such as image registration and volume segmentation in multiple image modalities. Furtherresearch is needed to clinically implement AI in breast cancer screening, but the results of thisstudy will help provide a basis for future studies, including prospective studies.

کلیدواژه ها:

Breast cancer ، MRI ، Artificial intelligence ، Computational Radiology ، Deep Learning ، CAD ، Mammography