A review on AI in radiology in Iran

Background and aims: Radiology is one of the fields of medicine that uses various imagingtechniques such as CT scan, ultrasound, MRI, simple x-ray to diagnose and follow up with patientsand respond to treatment. With the introduction of artificial intelligence, especially machinelearning, the hopes to obtain knowledge and undiscovered rules between data have become moreintense. Among the different fields of artificial intelligence, computer vision or image processinghas a unique position and popularity. We conducted a study to review articles written in Iran usingradiological methods and artificial intelligence & nbsp; to Show their results and to suggest somesolutions to enhance the quality of such studiesMethodology: We searched google scholar, SID, and PubMed and found ۴۰۰ articles studied, andwe separated related articles until December ۱, ۲۰۲۲.Results: Among the ۹۳ reviewed articles, ۳۹ articles met the inclusion criteria, including: Iranianauthor, use of artificial intelligence methods, activity in the field of radiology We excluded ۵۴ articlesfrom the study according to the exclusion criteria, including the following: Studies conductedoutside of Iran, review articles from foreign articles / not having enough connection with the topicof this article Among these articles, ۹ cases used chest CT images, ۱۷ used MRI and CT images,۳ used simple X-rays, and ۱۰ used ultrasounds. Among these studies, ۱۸ cases used external, and۲۱ used internal databases. Twenty-six cases investigated tumor and malignancy, five investigatedand diagnosed covid, and eight dealt with topics such as fatty liver, Alzheimer’s, age estimation,and MS, ...Conclusion: The studies conducted in Iran in the field of radiology and used image processingshow the desire of researchers to study tumors and malignancies. Unfortunately, a high percentageof researchers conducted on foreign databases. Our results show the necessity of creating andadequately storing data to provide internal resources for the optimal use of researchers to producepractical knowledge for the country. Machine learning for the prediction of inflammatory boweldiseases: a systematic reviewBackground and aims: Inflammatory bowel disease (IBD) are gastrointestinal (GI) disordersthat are chronic, and debilitating, and diminish the quality of life. In recent years, with a betterunderstanding of the pathophysiology of the disease, diagnosis and treatments, as well as technologicaladvances such as artificial intelligence, we have had new strategies and approaches in themanagement of IBD. Artificial intelligence (AI) is a new discipline that aims to simulate, extend,and expand human intelligence and integrates theory, method, and application research and development.The purpose of this study is to review articles about Machine learning (ML) that wereapplied to the field of inflammatory bowel diseases (IBD).Method: The study protocol adopted the PRISMA (Preferred Reporting Items for SystematicReviews and Meta-Analyses) guidelines. A systematic review was performed using PubMed andScopus to identify articles using machine learning in English literature and published from January۲۰۱۷ to November ۲۰۲۲. Based on the predefined selection criteria, ۲ levels of screening wereperformed: title and abstract review, and full review of the articles. Data extraction was performedindependently by all investigators and included algorithms, risk factors, sample size, purpose,evaluation index, deep learning, system developed, imaging modality, and limitation.Results: The study protocol adopted the PRISMA (Preferred Reporting Items for SystematicReviews and Meta-Analyses) guidelines. A systematic review was performed using PubMed andScopus to identify articles using machine learning in English literature and published from January ۲۰۱۷ to November ۲۰۲۲. Based on the predefined selection criteria, ۲ levels of screening wereperformed: title and abstract review, and full review of the articles. Data extraction was performedindependently by all investigators and included algorithms, risk factors, sample size, purpose,evaluation index, deep learning, system developed, imaging modality, and limitation.Conclusion: IBD public datasets need to be constructed and data standardization is necessaryfor clinical application of machine learning in digestive field. It is noteworthy that the purpose ofusing ML in the management of IBD is not to replace it with the physician, but as a tool to supporthuman-led decision-making and delivery of care.