Exploratory Study of Artificial Intelligence in Healthcare (original) (raw)

The rise of artificial intelligence has brought a positive shift in the sector by providing accurate data-driven decisions. The data from large systems is used for the early detection of chronic illnesses. These illnesses include cancer, diabetes, and cardiovascular diseases, etc. Existing technology is limited in terms of medical diagnosis etc. With the advent of ML/AI in the healthcare system, we expect to see much automation in clinical decision-making. We illustrate popular machine learning algorithms, their applications followed by methodology. This research will focus on the impact of Artificial Intelligence applications on the healthcare sector, its history, challenges, and concerns in the medical field. BACKGROUND INFORMATION Fostering trust in AI systems is a tremendous obstacle to bringing the most transformative AI technologies into reality, such as large-scale integration of machine intelligence in medicine. The challenge is to implement guiding ethical principles and aspirations and make the responsible practice of AI accessible, reproducible, and achievable for all who engage with the AI system. Meeting this challenge is critical to ensuring that medical professionals are prepared to correctly leverage AI in their practice and, ultimately, save lives. This research will concentrate on the influence AI applications have on the healthcare sector, its need, and its history in the medical field. Artificial intelligence models will assist doctors in various applications like patient care and administrative operations. (2011, March) Plant, R. et. al. According to the National Academies of Science, Engineering-diagnostic mistakes lead to roughly 10% of patient fatalities and 6 to 17% of hospital problems. It's crucial to remember that diagnostic errors aren't always caused by poor physician performance. Diagnostic mistakes, according to experts, are caused by a number of causes, including: • Collaboration and integration of health information technology are inefficient (Health IT) • Communication breakdowns between physicians, patients, and their families • A healthcare work system that is designed to be insufficiently supportive of diagnostic procedures. LITERATURE REVIEW Machine learning is being increasingly and frequently utilized in the healthcare field in various ways, like automating medical billing, clinical decision support, and establishing clinical care standards. Friedman, C., & Elhadad, N. (2014) et al. There are several significant applications of machine learning and healthcare ideas in medicine. The first medical machine learning system to diagnose acute toxicities in patients getting radiation treatment for head and neck malignancies has been created by researchers. In radiology, deep learning in healthcare automatically detects complicated patterns and assists radiologists in making informed judgments when analyzing pictures such as traditional radiography, CT, MRI, PET scans, and radiology reports. Machine learning-based automated detection and diagnostic systems have been demonstrated to perform as well as an expert radiologist. Google is creating a machine learning platform to identify breast cancer from images. These are only a handful of the numerous applications of machine learning in healthcare Jack Jr, C(2013) et.al. Natural Language Processing Nearly 80% of the information kept or "locked" in electronic health record systems is unstructured healthcare data for machine learning. These are papers or text files, not data components that could not previously be evaluated without a human viewing the information. Unfortunately, human language, often known as "natural language," is extremely complicated, lacks consistency, and contains many ambiguities, jargon, and vagueness. Therefore, machine learning in healthcare frequently uses natural language processing (NLP) tools to transform these papers into more usable and analyzable data.