Machine learning models for stroke detection by observing the eye-movement features under five-color visual stimuli in traditional Chinese medicine

Objective:To develop a novel diagnostic modality to identify and diagnose stroke in daily life scenarios for improving the therapeutic effects and prognoses of patients.Methods:In this study,16 stroke patients and 24 age-matched healthy participants as controls were recruited for comparative analysis.Leveraging a portable eye-tracking device and integrating traditional Chinese medicine theory with modern color psychology principles,we recorded the eye movement signals and calculated eye movement features.Meanwhile,the stroke recognition models based on eye movement features were further trained by using random forest(RF),k-nearest neighbors(KNN),decision tree(DT),gradient boosting classifier(GBC),XGBoost,and CatBoost.Results:The stroke group and the healthy group showed significant differences in some eye movement features(P<.05).The models trained based on eye movement characteristics had good performances in recognizing stroke individuals,with accuracies ranging from 77.40%to 88.45%.Under the red stimulus,the eye movement model trained by RF became the best machine learning model with a recall of 84.65%,a precision of 86.48%,a F1 score of 85.47%.Among the six algorithms,RF and CatBoost performed better in classification.Conclusion:This study pioneers the application of traditional Chinese medicine's five-color stimuli to visual observation tasks.On the basis of the combined design,the eye-movement models can accurately identify stroke,and the developed high-performance models may be used in daily life scenarios.

Integrating eye-movement analysis and the semantic differential method to analyze the visual effect of a traditional commercial block in Hefei, China

Eye-movement analysis was adopted to evaluate the visual perception of Chinese traditional commercial blocks,and the Chenghuangmiao block in Hefei,China,was selected as a typical case.Eye-movement data from 40 respondents viewing 14 sample pictures were recorded.The spatial elements of the sample pictures,including landmarks and commercial brands,were further extracted to analyze the visual characteristics of spatial elements and the factors that affect the respondents’perceptions of those elements.Then,the semantic differential method was used to analyze the relationship between visual preferences and psychological perceptions of spatial elements.Seventeen pairs of opposing adjectives were selected to score the intrinsic properties and visitors,feelings of spaces.The software SPSS 22.0 was used to analyze these data.Results showed that distinctive spatial elements,such as street corridors,Ma Tau Walls,and various landmarks,were the most visually attractive.In addition,the location of a given element within a picture was an important factor affecting eye movements.On this basis,strategies for improving spatial-visual effects are proposed.The strategies include emphasizing the visual characteristics of different spatial elements,considering the overall layout of spatial elements,and creating diversified spaces based on different spatial categories.