Robust automated reading of the skin prick test via 3D imaging and parametric surface fitting

Abstract

The conventional reading of the skin prick test (SPT) for diagnosing allergies is prone to inter- and intra-observer variations. Drawing the contours of the skin wheals from the SPT and scanning them for computer processing is cumbersome. However, 3D scanning technology promises the best results in terms of accuracy, fast acquisition, and processing. In this work, we present a wide-field 3D imaging system for the 3D reconstruction of the SPT, and we propose an automated method for the measurement of the skin wheals. The automated measurement is based on pyramidal decomposition and parametric 3D surface fitting for estimating the sizes of the wheals directly. We proposed two parametric models for the diameter estimation. Model 1 is based on an inverted Elliptical Paraboloid function, and model 2 on a super-Gaussian function. The accuracy of the 3D imaging system was evaluated with validation objects obtaining transversal and depth accuracies within ± 0.1 mm and ± 0.01 mm, respectively. We tested the method on 80 SPTs conducted in volunteer subjects, which resulted in 61 detected wheals. We analyzed the accuracy of the models against manual reference measurements from a physician and obtained that the parametric model 2 on average yields diameters closer to the reference measurements (model 1: -0.398 mm vs. model 2: -0.339 mm) with narrower 95% limits of agreement (model 1: [-1.58, 0.78] mm vs. model 2: [-1.39, 0.71] mm) in a Bland-Altman analysis. In one subject, we tested the reproducibility of the method by registering the forearm under five different poses obtaining a maximum coefficient of variation of 5.24% in the estimated wheal diameters. The proposed method delivers accurate and reproducible measurements of the SPT.

Skin Prick Automated Test device offers more reliable allergy test results compared to a manual skin prick test

“…It was concluded that the agreement between the device and the manual procedure was moderate (25) . Other methods supporting the automated read out of the SPT results are based on 3D imaging (26) or are using a combination of visible-spectrum and thermal images (27) .…”

Skin prick test wheal detection in 3D images via convolutional neural networks

“…The method proposed by Pineda et al [9] consisted of multiple stages. First, the wheals are detected via multi-scale filtering with Laplacian of Gaussians, second decomposing the image into patches, and third determining their centers of mass using PCA.…”

“…The images used to train and validate the neural network are the same from Pineda et al [9], in which the fringe projection system shown in Fig. 3 is used, and the 3D images of the forearm are like the one shown in Fig.…”

“…However, despite sophisticated image processing software, the results are not entirely reliable [7] or have problems with skin tone [8]. Alternatively, Pineda et al [9] recently proposed a robust method using a 3D imaging system and measuring the wheal size automatically. This method is based on pyramidal surface decomposition and Principal Component Analysis (PCA) to detect the wheals and perform parametric fitting to calculate their diameter.…”

Deep-learning enabled single-shot 3D feature detection in fringe projection profilometry

“…Base surface removal of each 3D image is performed using the pyramid decomposition method. 11 This approach, in Fig. 2, for the Base Surface Removal stage relies on a standard Gaussian-Laplacian pyramid 12 to obtain S′ from the 3D image, S. A Gaussian pyramid decomposes S into subsets of progressively lower resolution image versions G l , which are then down-sampled and iterated n times for each level until G n has only a few pixels.…”