Predictive Model Building for Pain Intensity Using Machine Learning Approach

When the patient’s body is compromised in any way, they will likely be in a lot of pain. If the caregiver is aware of the level of pain that the patient is experiencing, they will be better able to formulate the most appropriate treatment plan and provide the most appropriate medication. The visual analogue scale, often known as a VAS, is the approach that is used the most frequently to evaluate pain, and it is entirely dependent on patient reporting. Due to the fact that this kind of scale is ineffective when dealing with traumatic experiences or infants, it became necessary to devise a mechanism that could automatically recognize the severity of pain. On a dataset of multi-biopotential signals corresponding to varying degrees of discomfort, we evaluated the performance of the random forest and support vector machine classifiers.