@iftmuniversity.ac.in
Assistant Professor, Department of Computer Science & Engineering
IFTM University
Computer Science
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Ankur Jain, Muddada Murali Krishna, Sai Nitisha Tadiboina, Kapil Joshi, Yerrolla Chanti, and K. Sai Krishna
IEEE
The use of AI models in health care system and the life sciences is expanding. In this paper, we will take a look at the present state of the art and address the unanswered issues regarding the development of Ai technologies as clinical decision support tools. A review, which included a critical examination of papers published from 1990 and 2022, led the study's most challenging aspects.First, we demonstrate the structural distinction between ML and DL methods. Methods for training, validating, and testing ML models, as well as feature extraction, are described. In DL, models are provided as multi-layered artificial neural networks for direct image analysis. Data management includes technical stages like as image labelling, picture annotation, data standardization, and federated learning. After that, we divide the following into their own subsections: sample size computation, including frequent trials in AI methods; data augmentation strategies for coping with limited or unequal data; and the understandability of AI models. Finally, the advantages and disadvantages of ML and DL in introducing AI applications to diagnostic imaging are compared and contrasted. Biomedical and healthcare systems rank high on the list of important topics for AI applications, with medical imaging ranking as one of the most relevant and promising fields in which to apply such technology. Gaining insight into the specific difficulties associated with developing and deploying such systems in healthcare situations is helpful.
Ankur Jain, K K Ramachandran, Shikha Sharma, Trishu Sharma, Prakash Pareek, and Bhasker Pant
IEEE
The study explores with Machine learning (ML), which is a type of neural network (AI) that empowers software programmers to start increasing prediction without being done with full to do so. Because data is so valuable, improving strategies for intelligently having to manage the now-ubiquitous content infrastructures is a necessary part of the process toward completely autonomous agents. Computer vision and computer vision have improved a wide range of industries, including medical diagnoses, data display and procedures, science and research, and so. Just as preparing for a sport may be risky for individuals who are prone to injury, learning from contaminated or erroneous data can be costly. As described in the article Approaching Data Science, incorrectly trained algorithms result in expenses for a corporation rather than savings. Because incorrectly labeled, missing, or irrelevant data might impair the accuracy of any algorithm, organizations must be able to vouch for the quality and integrity of any data sets, along with their sources.