Research Essay (Main Body) Machine Learning in the healthcare sector

The ability of computers to mimic intelligent human behaviour is known as machine learning, which is a particular branch of artificial intelligence (Brown, 2021). There are numerous ways that machine learning and artificial intelligence can be used in medicine, and these applications have many positive effects and advantages. 

The first use is for identifying, diagnosing, and treating diseases. A client of IBM's created a predictive AI model that can identify sepsis in infants, and it was proven to be 75% accurate. Additionally, machine learning models may monitor the vital signs of a critically ill patient and can issue a warning if any danger factors appear. Because AI is can learn and store preferences, personalised therapies are now possible. A 24/7 AI virtual assistant that can offer advice or address any questions or concerns a patient might have based on their prior medical history or propensity can therefore be made available to them by the healthcare system. (IBM).

Support for decision-making is another application. The doctor's knowledge and the machine's capacity for huge data computation can work together to provide the best possible medical care. It's possible that a doctor won't be able to analyse every patient's heartbeat, data, or awareness of their unique conditions. However, the machine would carry out these tasks, and the doctor would be shown the results for approval. Additionally, by using computers to manage fees, insurance, and potential reimbursements, this can be extended to the financial side of the healthcare system (Shailaja et al. (2018) pp. 911–914).

Drug development is the third application. Drug development is expensive and can take a long period, as well as being time-consuming. However, this can be lessened by developing new, better medication designs and utilising machinery in combination. Alternatively, IBM Watson Health clients claim that AI might reduce the number of machine search codes by more than 70%, speeding up clinical trials by more effectively searching for machine codes (IBM).

Although machine learning has improved healthcare, some would claim that further research is needed. The person collecting the data may be biased or may not take into account particular categories of people (Chen et al., 2021, pp. 123–144). There is also the issue of explainability as experienced medical personnel would be interested in how the machine arrived at a certain diagnosis (Ahmad et al. (2018) pp. 559–560).