HJNO Sep/Oct 2023

48 SEP / OCT 2023 I  HEALTHCARE JOURNAL OF NEW ORLEANS COLUMN MEDICAID A range of potential Recent applications of machine learning (ML) and additional tools of AI applied to electronic health record (EHR) data have greatly improved the ability of healthcare providers to focus their attention on un- foreseen issues indicated by specific se- rological markers, identifying patients as ideal candidates for specific need-based services and care. ML is the process by which an algorithm is used to encode in- stances or entries — from a database of examples — into weighted parameters or outputs for predicting future outcomes. The larger the database the better, as it pro- vides more instances to train the MLmod- el. For example, some healthcare networks have begun working with Google to build out ML prediction models that use big data to alert clinicians of high-risk conditions through hospital stays such as sepsis and DYNAMIC disease patterns and increas- ing costs make effective prevention, edu- cation, and population health an ongoing challenge for health systems. Individual screenings and medically documenting preexisting conditions are tradition- al methods for progressing patient care. However, we currently possess the techni- cal capacity to pursue a new type of “preci- sion”health: big data. Big data plays a pow- erful role in advancing healthcare delivery, research, and innovations that address barriers and improve the quality of care. Through big data and the application of artificial intelligence (AI) methods, we are able to drill down tomore tailored needs of individuals. To date, big data has been used tomake dramatic improvements in disease diagnosis, resource allocation, treatment candidacy, screening, and linkage to care, and even goes as far as influencing ad- ministrative processes and workflows. The first influence of AI on healthcare process- es started as early as 1970 when “MYCIN” was developed at Stanford for use with diagnosing blood-borne bacterial infec- tions but not adopted into clinical practice. Since then, though, many other programs, platforms, and people have contributed to the field of precision medicine. Through the association of computer science, algorithms, machine learning, and data science, AI can help solve prevailing healthcare challenges such as detecting gaps in screening, enhancing decision support, and quantifying treatment or re- source response. In understanding utiliza- tion, AI can also go beyond a clinical care application. Most often in healthcare, ad- dressing these challenges relates back to quality of life, cost-savings in care delivery, and predicting engagement in health sys- tems at multiple entry points. ARTIFICIAL INTELLIGENCE: Synthesizing Big Data to Improve Individual Care