Deep Learning Method Applied to Autonomous Image Diagnosis for Prick Test.

BACKGROUND

The skin prick test (SPT) is used to diagnose sensitization to antigens. This study proposes a deep learning approach to infer wheal dimensions, aiming to reduce dependence on human interpretation.

METHODS

A dataset of SPT images (n = 5844) was used to infer a convolutional neural network for wheal segmentation (). Three methods for inferring wheal dimensions were evaluated: the ; the standard protocol (); and approximation of the area as an ellipse using diameters measured by an allergist (). The results were compared with assisted image segmentation (), the most accurate method. Bland-Altman analysis, distribution analyses, and correlation tests were applied to compare the methods. This study also compared the percentage deviation among these methods in determining the area of wheals with regular geometric shapes (n = 150) and with irregular shapes (n = 150).

RESULTS

The Bland-Altman analysis showed that the difference between methods was not correlated with the absolute area. The achieved a segmentation accuracy of 85.88% and a strong correlation with the method (ρ = 0.88), outperforming all other methods. Additionally, showed significant error (13.44 ± 13.95%) for pseudopods.

CONCLUSIONS

The protocol can potentially automate the reading of SPT, offering greater accuracy than the standard protocol.