Review of Medical Diagnostic Imaging Using Machine Learning Research and Challenges
##plugins.themes.bootstrap3.article.main##
الملخص
Background: Recent years have seen continuous progress in the medical industry to enable correct medical diagnosis. Machine learning is another key area which defines the capability of a system to learn from large-scale data entities within the field of medicine to diagnose illnesses.
Purpose: To this end, the review sought to assess the part that is played by machine learning in identifying certain diseases. This was in a bid to find commonalities and disparities of various supervised learning techniques that may be helpful in the diagnosis of illnesses.
Materials and Methods: Some of the machine learning techniques compared within the study included Supervised, Unsupervised, and Semi-supervised Machine Learning, Active Learning, Reinforcement Learning, Evolutionary Learning, as well as Deep Learning. The authors analyzed various approaches to machine learning used in healthcare to diagnose images, where preference was given to various medical specialties.
Results: After going through the various methods, the researcher suggested that deep learning should be implemented in Machine Learning processes since it has various functions in medical diagnosis. Moreover, the most compelling argument raised as to why one should adopt multiple approaches to learning indicated that learning improves the accuracy of medical diagnostic imaging.
Conclusion: Homomorphic Filtering (MHFIL) is applied to enhance the images contrast in medical diagnostic imaging hence is the recommendation. This method allows for the simultaneous overcoming of issues that low contrast machine learning poses in efficient diagnosis.
التنزيلات
##plugins.themes.bootstrap3.article.details##
كيفية الاقتباس
هذا العمل مرخص بموجب Creative Commons Attribution 4.0 International License.