Statistical analysis process of nystagmus displacement vector

Abstract

The invention relates to a statistical analysis method of eye movement displacement vector based on energy spectrum analysis of displacement statistical chart, which comprises following steps: locating eyeballs in a video image; making a displacement statistical chart; carrying out Gabor transformation of the displacement statistical chart; analyzing components; constructing characteristic vectors; obtaining weak classifiers by SVM training and grouping; constructing strong classifiers; and constructing analysis model and reporting the result. Based on videonystagmography, the method provided by the invention can execute energy spectrum analysis for the eye movement displacement statistical chart to extract the dominant and recessive characteristics to obtain the mediate analysis result, which is displayed on a computer screen in an image interface manner, thus providing helpful reference information related to the video image characteristics of nystagmus for medical science.

Description

technical field [0001] The invention relates to a statistical analysis method of graphic information, in particular to a nystagmus medical feature extraction and analysis method which uses a nystagmus displacement statistical graph as an object, performs principal component analysis on its energy spectrum, and obtains an intermediate information form. Background technique [0002] Nystagmus, referred to as nystagmus, is an involuntary, rhythmic, back-and-forth swinging eye movement, often caused by diseases of the visual system, extraocular muscles, inner ear labyrinth, and central nervous system. Nystagmus symptoms are closely related to Alzheimer's disease, Parkinson's syndrome, epilepsy, inner ear and central nervous system diseases, vertigo, balance dysfunction and other symptoms. Clinically, due to the lack of scientific reference for neurological diseases, they are often positioned as intractable diseases, which mainly depend on the subjective judgment of doctors, whic...

AI Technical Summary

Problems solved by technology

However, the ENG test has its inherent defects: first, it cannot detect rotational nystagmus; second, it is easily affected by many external factors, such as the drugs taken by the patient recently, the awakening state during the test, other biological signal interference, and the experience of the operator. It can affect the accuracy of the test results; thirdly, the performance and stability of the electrodes affect the signal acquisition. Some scholars have concluded that 95% of the faults in the electronystagmogram are electrode problems; in addition, the inspection content of the electronystagmogram is cumbersome and expensive , currently only used in large hospitals in China

However, the research on the VNG system is still very weak. In addition to applying the existing achievements of the ENG system, it mainly focuses on extracting the intuitive temporal and spatial information of the video nystagmus for analysis. There are major problems: (1) The time- Null characteristics are not enough to fully reflect the pathological characteristics and causes of various diseases. Focusing on the analysis of intuitive parameters of nystagmus will ignore important microscopic information, and these information are likely to contain important pathological characteristics; (2) Existing research The method does not analyze the energy distribution characteristics of nystagmus, which limits the extraction of medical features, thus losing some important reference information; (3) does not make full use of the existing advanced target classification and pattern recognition technology, and the pathological analysis and discrimination methods for nystagmus In-depth and systematic research, resulting in poor system robustness, inconvenient operation, and excessive reliance on the subjective judgment of doctors

Images