• info@icbcongress.com
• English
• Persian

• Login

Create Free Account

INTERNATIONAL CONGRESS OF

CancerGenetics

• The role of Artificial Intelligence (AI) in breast cancer diagnosis

• Ali Rezaeian,¹ Atefeh Kamran,² Zahra Amirkhani,^3,*
  1. Student Research Committee, Larestan University of Medical Sciences, Larestan, Iran.
  2. Student Research Committee, Larestan University of Medical Sciences, Larestan, Iran.
  3. Student Research Committee, Larestan University of Medical Sciences, Larestan, Iran.
  
  
  
• Introduction: Breast cancer is a condition in which abnormal breast cells get out of control and form tumors. If left unchecked, tumors can spread throughout the body and become deadly. Breast cancer cells begin inside the milk ducts and/or lobules that produce breast milk. According to our studies, breast cancer screening with mammography is very effective in reducing breast cancer-related deaths. In addition, it is important to minimize errors and misinterpretations of lesions visible in digital mammography, which contributes to at least 25% of detectable cancers. Computer-aided diagnostic systems (cads) were introduced as assisting radiologists who tried to improve human diagnostic performance. Although some studies have shown that single plus CAD reading can replace double reading, few, if any, have identified the real benefits of using single plus CAD versus single reading alone (e.g., the real benefits of radiologists ' on-screen performance). In general, the benefits of using CAD in screening are still unclear. However, significant advances in gene (AI) AI with deep complex neural networks (com manly known as deep learning algorithms) reduce the performance difference between humans and com putters in many medical imaging applications, including breast cancer diagnoses.
• Methods: Our search strategies and data sources following the preferred reporting cases for systematic reviews and meta-analysis guidelines (PRISMA) for systematic review and meta-analysis, we conducted a systematic literature review to integrate the findings of quantitative studies. Our systematic literature research reviewed 7 electronic databases: Scopus, CINAHL, Medline via PubMed, Web of Science, EMBASE, Cochrane Library (review and central). Using grid and manual search terms, the keywords being explored are: " Breast cancer ", "", " Artificial Intelligence (AI) ", "cancer survivors*," post-treatment*, in combination with "risk factors", "outcomes" and " Other eligible studies were identified by examining cited sources from published studies obtained.
• Results: Breast cancer is one of the leading causes of death among women. Early detection, proper control mechanism and treatment of breast cancer can significantly improve the lives of millions of women worldwide. Several popular imaging methods such as MGS, US, MRI and HP images are used among many others to detect breast cancer. Given the importance of finding a solution/framework for early detection and detection, recently many AI researchers are focusing on automating this task. The various imaging methods used by researchers to automate the work of breast cancer diagnosis are mammography, ultrasound, magnetic resonance imaging, histopathological images, or any combination thereof. . Automated AI capabilities provide the potential to enhance the diagnostic expertise of physicians, including accurate determination of tumor volume, extraction of characteristic cancer phenotypes, translation of tumor phenotype characteristics into clinical genotype implications, and risk prediction. Combining image-specific findings with underlying genomic, pathological and clinical characteristics is increasing in value in breast cancer. The simultaneous emergence of newer imaging techniques has provided radiologists with diagnostic tools and image datasets for analysis and interpretation. Integrating an AI based workflow into breast imaging enables the integration of multiple data streams into powerful multidisciplinary applications that may guide the path to specific patient's personal medicine.
• Conclusion: Most of our research focused on breast cancer diagnosis and subtype classification. This leaves room for future research to address various related topics such as identifying risk levels and predicting the likelihood of relapse. One direction for future research relates to the implementation of multi-class predictors using genetic data. Most research papers used genetic sequencing data only by binary classification, with the primary focus being breast cancer diagnosis and the likelihood of survival. As a result, radiologists improved their diagnosis in the diagnosis of breast cancer in mammography using an AI computer system for support without the need for additional reading time. However as promising as these findings may be Studies in a screening scenario need to be done to verify them.
• Keywords: Breast cancer, Artificial Intelligence (AI), post-treatment