Intelligent decision support method and system based on electronic hand-drawn spiral test

Abstract

The invention provides an intelligent decision support method and system based on an electronic hand-drawn spiral test, and the method comprises the steps: obtaining training data comprising a plurality of hand-drawn spiral trajectories, the hand-drawn spiral trajectories having category labels indicating whether the hand-drawn spiral trajectories belong to Parkinson; extracting features of each hand-drawn spiral trajectory in the training data to train a random forest model in combination with the category label, and taking the trained random forest model as an intelligent decision support model; and inputting the features of the to-be-classified hand-drawn spiral trajectory into the intelligent decision support model to obtain the category to which the features belong, and taking the category as a decision support result of the to-be-classified hand-drawn spiral trajectory. According to the electronic spiral line test, aided decision support can be rapidly completed, and the problems that traditional Parkinson aided decision is low in feature dimension, single in test form and the like are solved.

Description

technical field [0001] The present invention relates to the technical fields of machine learning and human-computer interaction. Specifically, the present invention proposes an intelligent decision support method and system based on electronic hand-drawn spiral testing. Background technique [0002] Parkinson's disease is currently the second largest neurodegenerative disease in the world, and the number of patients worldwide has reached 7 million to more than 10 million. About 190 out of every 10,000 people over the age of 80 have Parkinson's disease, and the prevalence of Parkinson's disease will gradually increase with age. At present, the main symptoms of Parkinson's disease are mainly observed and evaluated clinically by specialists, such as tremor, slow movement and rigidity. However, since Parkinson's disease does not have obvious manifestations in the early stage, most Parkinson's disease patients are difficult to realize that they have Parkinson's disease in the ea...

AI Technical Summary

Problems solved by technology

However, the traditional helix is ​​drawn using paper and pen, so that many important time characteristics (such as the average speed, maximum speed, variance of speed, average acceleration, maximum acceleration and acceleration variance, etc.) cannot be accurately recorded when drawing the spiral

Secondly, in the research report "Let's Draw: Detecting and Measuring Parkinson's Disease on Smartphones" published in 2018, it was mentioned that when subjects repeated a single test, the tremors in their hands tended to gradually decrease. Restrictions, the test category is relatively single

In addition, the traditional method requires patients to be completed under the supervision of specific personnel, the diagnosis is constrained by time and space conditions, and the data recorded in paper materials is not conducive to dissemination and preservation

Images