Semantic Segmentation of Histopathological Slides for the Classification of Cutaneous Lymphoma and Eczema

Abstract

Mycosis fungoides (MF) is a rare, potentially life threatening skin disease, which in early stages clinically and histologically strongly resembles Eczema, a very common and benign skin condition. In order to increase the survival rate, one needs to provide the appropriate treatment early on. To this end, one crucial step for specialists is the evaluation of histopathological slides (glass slides), or Whole Slide Images (WSI), of the patients' skin tissue. We introduce a deep learning aided diagnostics tool that brings a twofold value to the decision process of pathologists. First, our algorithm accurately segments WSI into regions that are relevant for an accurate diagnosis, achieving a Mean-IoU of 69% and a Matthews Correlation score of 83% on a novel dataset. Additionally, we also show that our model is competitive with the state of the art on a reference dataset. Second, using the segmentation map and the original image, we are able to predict if a patient has MF or Eczema. We created two models that can be applied in different stages of the diagnostic pipeline, potentially eliminating life-threatening mistakes. The classification outcome is considerably more interpretable than using only the WSI as the input, since it is also based on the segmentation map. Our segmentation model, which we call EU-Net, extends a classical U-Net with an EfficientNet-B7 encoder which was pre-trained on the Imagenet dataset.

Domain Adaptation-Based Deep Learning for Automated Tumor Cell (TC) Scoring and Survival Analysis on PD-L1 Stained Tissue Images

“…For instance, Liu et al [19] used Inception V3 network [20] to perform a patch based classification on whole slide images to produce course segmentation maps to detect cancer metastasis. To obtain dense segmentation results, Scheurer et al [21] used a UNet network [18] with an efficient net B7 backbone [22] for classification of cutaneous lymphoma and eczema. To incorporate more contextual information from large histopathological images into the networks, Graham et al [23] used multi-scale network to perform instance-based nuclei segmentation.…”

Objective hand eczema severity assessment with automated lesion anatomical stratification

“…One study's approach consisted in the detection (as opposed to segmentation) of atopic eczema lesions based on 1393 patients' pictures followed by the severity classification of seven clinical signs 30 . Segmentation and classification of eczema lesions was also performed on histopathological slides 31 …”

Current state and prospects of artificial intelligence in allergy

“…AI guided image analysis has been performed to diagnose eczema. [123][124][125][126] One study developed a classifier of atopic dermatitis in multiphoton tomography images, reaching over 97% accuracy 127 through transfer learning. Highlighted areas of interest in the images could support clinicians in faster diagnosis.…”

“…AI guided image analysis has been performed to diagnose eczema 123–126 . One study developed a classifier of atopic dermatitis in multiphoton tomography images, reaching over 97% accuracy 127 through transfer learning.…”