Facial paralysis grading diagnosis method and device based on artificial intelligence

Abstract

The invention discloses a facial paralysis grading diagnosis method and device based on artificial intelligence. The method comprises the steps: obtaining a front image of the face of a patient, and inputting the front image of the face of the patient into a facial paralysis diagnosis model to obtain an HB grade and an FNGS2.0 partition score, wherein the facial paralysis diagnosis model is obtained by training a Softmax nonlinear classifier. Through the method and the device, facial paralysis can be accurately and rapidly subjected to grading diagnosis, the inconsistency of subjective evaluation is eliminated, and doctors are assisted in making decisions. The input data are convenient to obtain, technical sensitivity is low, the output result is the grading standard which is widely accepted clinically, and the method and the device are easy to popularize and use.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method, device, electronic equipment and storage medium for grading and diagnosing facial paralysis based on artificial intelligence. Background technique [0002] Bell's palsy, trauma, tumor and iatrogenic injury may all cause facial paralysis, and the graded diagnosis of facial paralysis will run through the entire treatment process to evaluate the patient's condition. Compared with techniques such as electromyography, imaging methods are the most intuitive, non-invasive and convenient. At present, clinically limited by consultation time, consultation room space and funds, most doctors still use traditional subjective evaluation methods, that is, doctors observe specific facial expressions of patients on the spot, or photos of these expressions, and give grades based on experience according to certain grading standards . [0003] Artificial intelligence (AI)...

AI Technical Summary

Problems solved by technology

There are two shortcomings in this method: first, physicians are highly sticky to the use of grading standards, and HB grading and FNGS2.0 grading have a deep foundation for use at the grassroots level, domestic top hospitals, and even internationally, and it is difficult for physicians to change their usual methods. If the self-created grading standard does not have obvious advantages over the predecessors, it will be difficult to promote, thereby reducing the clinical application value of the method

Second, the characteristics of AI enable it to discover the objective laws behind human subjective cognition. The advantage is that it expands the scope of data acquisition, but also irrelevant noise (such as background) will also be included in the model, causing bias

Images