Edinburgh Napier University

  Aim: Morphological changes in the epidermis layer are critical for the diagnosis and assessment of various skin diseases. Due to its non-invasiveness, optical coherence tomography (OCT) is a good candidate for observing microstructural changes of skin. Convolutional neural network (CNN) has been successfully used for automated segmentation of the skin layers of OCT images to provide the objective evaluation of skin disorders. Such method is reliable provided that a large amount of labelled data is available which are however very time-consuming and tedious. The scarcity of patient data also puts another layer of difficulty to make the model more generalizable.
Approach: In this paper, we report a semi-supervised representation learning method to provide data augmentations using rodent models to train neural networks for accurate segmentation on clinical data.
Result: The learning quality is maintained with only one OCT labelled image per volume that is acquired from patients. Data augmentation introduces a semantically meaningful variance, allowing for better generalization. Our experiments demonstrate the proposed method can achieve accurate segmentation and thickness measurement of the epidermis.
Conclusion: This is the first report of semi-supervised representative learning applied to OCT images from clinical data by making full use of the data acquired from rodent models.