مشخصات پژوهش

عنوان	Hidden Pattern Discovery on Clinical Data: an Approach based on Data Mining Techniques
نوع پژوهش	مقاله چاپ شده
کلیدواژه‌ها	Clinical Data, Data Mining, Unsupervised Learning, Association Rule Mining, Clustering
چکیده	In this study, we sought to minimize the need for redundant blood tests in diagnosing common diseases by leveraging unsupervised data mining techniques on a large-scale dataset of over one million patients' blood test results. We excluded non-numeric and subjective data to ensure precision. In order to identify relationships between attributes, we applied a suite of unsupervised methods including preprocessing, clustering, and association rule mining. Our approach uncovered correlations that enable healthcare professionals to detect potential acute diseases early, improving patient outcomes, and reducing costs. The reliability of our extracted patterns also suggests that this approach can lead to significant time and cost savings while reducing the workload for laboratory personnel. Our study highlights the importance of big data analytics and unsupervised learning techniques in increasing efficiency in healthcare centers.
پژوهشگران	راضیه میدانشاهی (نفر دوم)، میثم روستائی (نفر اول)