UNAIR NEWS – Recently, the term machine learning in the medical world has been used often in public. As a part of statistics, artificial intelligence, and computer science, machine learning is also known as predictive analytics or statistical learning.
The role of machine learning in medical sector is very diverse, such as identification, diagnosis, prediction of a disease, smart health records, medical imaging, and so on. So it’s no wonder that many people consider machine learning to be very helpful.
As expressed by Ratih Ardiati Ningrum, M.S., M.Stat , as a lecturer in Data Science Technology (TSD), Faculty of Advanced Technology and Multidiscipline (FTMM), if machine learning algorithms can be applied optimally, then a doctor can diagnose a disease earlier and more accurately.
“Some doctors have started to look at machine learning algorithms to be applied to data processing. This will certainly provide high efficiency and accuracy, ”said the lecturer who graduated from National Chiao Tung University, Taiwan on Tuesday, April 27, 2021.
Acts like human brain
Basically, machine learning is the study of how computers can understand and act like the human brain through data. The goal is for the computer to gain insight from the data.
The function of machine learning as a computation platform for analyzing data is undoubtedly beneficial. There are many machine learning algorithms that can be used for data processing and analysis, especially when compared to traditional statistical methods.
For example, in applying the method to traditional statistics, there are two assumptions that must be fulfilled. However, with machine learning being, it is more flexible, the use of any algorithm is just a matter of adjusting to existing data patterns. Furthermore, from various algorithms, the most optimal in accuracy, sensitivity, error value, and other model goodness criteria is selected.
“Machine learning helps medical personnel get insights from data to make it faster and more accurate in diagnosing or identifying diseases,” explained Ratih.
Offers many processing algorithms
Machine learning offers many algorithms that can be used to process data. These algorithms are divided into supervised, unsupervised, and reinforcement learning where each type is further divided into several types of algorithms. For example, for supervised learning known decision tree, support vector machine, neural networks, naive Bayes, and so forth.
“A health scientist before applying algorithms to the data processing analysis also needs to understand the case at hand. Furthermore, having knowledge related to problems to be resolved is very necessary in addition to knowledge related to machine learning,” she added.
The application of machine learning in the medical sector is like the detection of a disease, for example there is a lump in the skin, so machine learning can detect whether the lump is a benign tumor or a malignant (cancer) tumor. In addition, in the field of radiologists, machine learning can help read visualizations of X-Ray results more quickly and accurately.
“Hopefully in the future machine learning will be used more in processing data, so that it is more effective, efficient, maximal, and accurate. Moreover, it offers various methods and tools that are easy to use in processing data, ” she concluded. (*)
Author: Wildan Suyuti
