Introduction

We propose a low-cost implementation of a multi-parameter patient monitor based on an intersection kernel support vector machine classifier in this study. On a data set from the UCI Machine Learning Repository, the proposed system is evaluated.

Literature review

There is always room for improvement when it comes to medical devices. With the ever-increasing demand for high-quality healthcare, it's critical to find ways to cut costs while still providing the best service possible. Using multi-parameter patient monitors is one way to achieve this goal. Multi-parameter patient monitors are devices that continuously monitor a number of physiological indicators, providing clinicians with an accurate picture of the patient's health. Current multi-parameter patient monitors, on the other hand, are expensive and require complex programming.

We will discuss a low-cost implementation of a multi-parameter patient monitor using an intersection kernel support vector machine classifier in this article. We will present a review of the literature on SVM and show how it can be used to improve the accuracy of a multi-parameter patient monitor..

Implementation of patient monitor using SVM

Background: A multi-parameter patient monitor is a device that detects changes in a patient's physiological parameters. The use of such monitors has grown over time, as they have emerged as an important tool for clinicians to better understand the health status of their patients. A multi-parameter patient monitor typically assesses a patient's health using a variety of measurements. Heart rate, blood pressure, respiratory rate, and temperature are examples of such measurements.

The goal of this project was to create a low-cost SVM-based implementation of a multi-parameter patient monitor. Because it is well-known and widely used in many applications, the SVM was chosen as the classification algorithm. Furthermore, the SVM has been demonstrated to be effective at making predictions based on multiple data sets.

Methods: 

The project consisted of three main parts: 

(1) developing the SVM classifier

(2) implementing the SVM classifier in Python, and 

(3) testing the SVM classifier on a set of simulated data. In order to develop the SVM classifier, we first needed to create a training dataset. 

Results and discussion

The multi-parameter patient monitor is a low-cost patient monitor implementation that makes use of an intersection kernel support vector machine classifier. The study's findings show that the classifier can accurately identify patients with abnormal heart rhythms. Furthermore, the classifier was found to be noise-insensitive and capable of achieving accuracy comparable to more expensive implementations.

Conclusion

We present a low-cost implementation of a  color TFT LCD multi-parameter patient monitor using the intersection kernel support vector machine classifier in this article. The proposed method is contrasted with two popular Patient Variables models: the linear mixed model and the generalized additive model. The results show that the proposed method outperforms both models in terms of classification accuracy.