# Machine Learning- Data Science Insight

The lack of available doctors and other medical staff is certainly an issue in impoverished areas of the world. However, the data science method named machine learning may be there to help this situation. Machine learning utilizes data from previous patients to help predict what ailment patients currently have, and how to treat it. With the advancement of the collection of data, (as machine learning has been around since the 1950s), this strategy of having essentially a robot doctor could save resources, time, and lives in areas where physicians are scarce.

The available resources of data in the healthcare industry are there. According to healthcatalyst.com, “30 percent of the world’s warehoused data comes from the healthcare industry, and an estimated $300 billion dollars in annual cost savings could be saved while utilizing this data.” This can be done using machine learning models. This machine learning essentially uses previous outcomes of diagnosis and imagery and selects data points in order to create statistical models. These models can be used for prediction in new patients and can help guide a physician to care for the patient more efficiently.

For example, researchers at Stanford have recently created a machine learning model that can “diagnose irregular heart rhythms (arrhythmias) from single-lead ECG signals better than a cardiologist,” according to healthcatalyst.com. This website also makes it important to note that “Clinicians record more than 300 million ECGs annually, so the data needed for improved arrhythmia diagnosis already exists.” Another study conducted at Stanford has found a machine learning model for different types of skin cancer. Their model looks at skin tissue damages and sees if those marks are cancerous or not. Seeing that dermatologists largely conduct visual exams, this can make care both more efficient and can sometimes help point them in the direction of harmful skin.

There are so many ways for data science and machine learning to be able to be used in the healthcare industry. It can only grow exponentially as more patients add to the stockpile of data. The possibilities of machine learning can be used in so many areas of healthcare, the future is bright. In developing worlds, these models of machine learning can be made cheap to use in the future as they evolve and become more accurate. Companies like KenSci and PathAI are notable in leading the way for machine learning in healthcare. In the poorest of countries, the automation of data can free up physicians to care for patients that need their attention more, so they can give more efficient care. A final hurdle for these developing areas is the high cost of machine learning. The cost needs to be driven downwards before these countries can truly be helped.