Establishing an objective level of pain after a surgical operation process is not easy because the level of pain in a person is very subjective and what may be unbearable pain for one person may be mild pain for another.
Therefore, looking for patterns that can lead to establishing a model of medical understanding that allows an objective pain scale to be classified by means of certain patterns is very interesting.
In this demonstration project (carried out only for demonstrative purposes for the hackathon https://awshealthai.devpost.com/ , and without any medical value since it has not been developed to its full potential) the aim is to establish an objective level of pain in the postoperative process of a woman who has suffered cancer of the breast and has undergone surgery.
To carry out this project, two data models and two systems are combined:
In the first place, and with the established example code, the aim is to visualize the data that patients can report on related issues. The dataset corresponds to: Stanford University developed the Q-Pain (https://physionet.org/content/q-pain/1.0.0/)
Second, it is pursued by using Amazon Comprehend Medical and after extracting and cleaning the “procedure description” (postoperative) data from medical report samples (MTSamples (https://www.mtsamples.com/)) to achieve a mapping of the entities that may exist.
It is with the combination and expert knowledge of both models that a medical specialist could establish a target pain level for a given patient that meets specific requirements with specific patterns.
Built With
- amazon-comprehend-medical
- amazon-web-services
- sagemaker
Log in or sign up for Devpost to join the conversation.