ENG800 Assignment Literature Review Partial Draft 1

ENG800 Assignment Literature Review Partial Draft 1 Student Name Franklin University ENG 800 Professor Name Date Literature Review: Machine Learning in Healthcare Machine Learning (ML) is a branch of AI that discovered the use of algorithms and statistical models to get computers to learn from the data and improve their operations without being individually programmed. The system works by using data, recognizing patterns, and even making decisions with little or no input from a human agent. ML is being used in every field like finance, retail, and automotive. In finance, the ML algorithms determine future trends of the market and fraud. In the retail domain, they are applied for recommendations of products, and stock control. The application of ML in the automotive industry is in the self-driving segment and vehicle diagnostics. In the healthcare sector, ML can help improve the delivery of services to patients through improved diagnostic capability, individualized treatments, patient prognosis and also in the management of healthcare records. The application of ML in the healthcare sector has the potential of effectively meeting some of the paramount issues affecting the healthcare industry including; costs, productivity, and quality outcomes. The following literature review seeks to understand how the healthcare sector has changed due to ML adoption, and the advantages and disadvantages of the use of ML to improve the healthcare delivery systems. Thus, this review focuses on the presentation of the current state of ML in healthcare, the analysis of its influence on healthcare results, and the identification of the weaknesses and the line of further development of ML within healthcare. Applications of Machine Learning in Healthcare Both supervised and unsupervised ML techniques have yielded outstanding results in diagnosing many diseases such as cancer, cardiovascular diseases, and diabetes. For instance, CNNs are utilized for image analysis to diagnose cancer in the early stages, whereas decision trees, as well as SVMs, are employed to identify heart diseases based on the patient’s data. Several research has shown that, by using mammograms, ML models have shown great potential to diagnose breast cancer as effectively as radiologists. An example is Google AI which helped to minimize false-negative and false-positive in breast cancer screening (Freeman et al., 2021). Healthcare applications have also incorporated logistic regression and random forests to forecast possible cardiovascular occurrences in order to intervene and assist patients with better treatment plans (Seetharam et al., 2019). Personalized medicine is a concept of practicing medicine that takes into consideration all attributes of a patient. Using data such as genetics, patient behaviors, and medical history, ML models come up with healthcare recommendations for patients. In oncology, some ML algorithms have been applied for the analysis of patient’s reactions to particular types of chemotherapy and, therefore, to select the most efficient treatment for each patient (Gambardella et al., 2020). Also, more advanced approaches are being studied to create the ML model that estimates what reaction patients will have to certain drugs in consideration of their specific genes, which could lower the occurrence of adverse drug reactions and increase the effectiveness of treatments (Secinaro et al., 2021). ML makes the medical imaging such as MRI, CT scans, and X-rays better. It does this by increasing the image quality, and resolution, but most importantly it can recognize diseases or disorders that might not be detected by any radiologist. For instance, deep learning models have been trained in such ways as to enable the acceleration of MRI scans so that it can produce high signals in a shorter time than what is applied in traditional methods (Waddington et al., 2023). Signs of diseases such as lung cancer can also be diagnosed at an emulation stage by the use of the ML algorithms, in CT scan images with better precision than conventional approaches (Asuntha & Srinivasan, 2020). Predictive analytics includes the use of ML models to analyze future trends in patient provisioning, for instance, the number of patients likely to be readmitted, their progression to other stages of the disease, or their likely recovery period. These predictions assist the various healthcare providers in the right deployment of their resources besides facilitating the putting in place of preventive measures. For instance, the development of an ML model can help estimate the probability of a patient’s readmission within 30 days of being discharged with the possibility of introducing intervention to lower readmission incidences (Sharma et al., 2022). ML has also been used to predict patients possibly to develop sepsis, which is a life-threatening condition and allows for the provision of timely treatment to enhance the patient’s survival rate (Yuan et al., 2020). Benefits of Machine Learning in Healthcare ML is beneficial since it involves relying on large data sets, by finding hidden patterns that help in providing accurate diagnosis. This helps minimize some cases of wrong diagnosis and delay in the treatment. Through virtual performance of such tasks and formulation of assisting decisions, ML minimizes human interference, hence enhancing the efficiency and precision of human-related medical tasks. Predictive analytics is useful in the screening of high-risk cases and preventing more costly complications, and the efficient use of resources hence cutting down on over expenses on healthcare. Businesses also are able to have ML algorithms handle tasks such as appointment making, invoicing, and handling of claims hence reducing the costs and time of operations. Patients want their treatments to be as unique as they are, and individualized treatment models can raise the chances of success, in addition to making patients happier with the treatments they are getting. Further, remote monitoring with the help of AI technologies monitors patients’ health in real-time and allows for follow-up treatment promptly. Challenges and Limitations Patient privacy is also a problem where patients’ information is at risk of being stolen and shared with the public. Security has to be implemented and in compliance with the existing laws such as the HIPAA laws. Adopting ML in healthcare has challenges such as the status of regulation and ethical considerations in use of data and AI decisions. Anyway, the […]