Objective The microscopic review of hematoxylin‐eosin–stained images of focal cortical dysplasia type IIb and cortical tuber of tuberous sclerosis complex remains challenging. Both entities are distinct subtypes of human malformations of cortical development that share histopathological features consisting of neuronal dyslamination with dysmorphic neurons and balloon cells. We trained a convolutional neural network (CNN) to classify both entities and visualize the results. Additionally, we propose a new Web‐based deep learning application as proof of concept of how deep learning could enter the pathologic routine. Methods A digital processing pipeline was developed for a series of 56 cases of focal cortical dysplasia type IIb and cortical tuber of tuberous sclerosis complex to obtain 4000 regions of interest and 200 000 subsamples with different zoom and rotation angles to train a neural network. Guided gradient‐weighted class activation maps (Guided Grad‐CAMs) were generated to visualize morphological features used by the CNN to distinguish both entities. Results Our best‐performing network achieved 91% accuracy and 0.88 area under the receiver operating characteristic curve at the tile level for an unseen test set. Novel histopathologic patterns were found through the visualized Guided Grad‐CAMs. These patterns were assembled into a classification score to augment decision‐making in routine histopathology workup. This score was successfully validated by 11 expert neuropathologists and 12 nonexperts, boosting nonexperts to expert level performance. Significance Our newly developed Web application combines the visualization of whole slide images with the possibility of deep learning–aided classification between focal cortical dysplasia IIb and tuberous sclerosis complex. This approach will help to introduce deep learning applications and visualization for the histopathologic diagnosis of rare and difficult‐to‐classify brain lesions.
A light microscopy-based histopathology diagnosis of human brain specimens obtained from epilepsy surgery remains the gold standard to confirm the underlying cause of a patient’s focal epilepsy and further inform postsurgical patient management. The differential diagnosis of neocortical specimens in the realm of epilepsy surgery remains, however, challenging. Herein, we developed an open access, deep learning-based classifier to histopathologically assess whole slide microscopy images (WSI) and to automatically recognize various subtypes of Focal Cortical Dysplasia (FCD), according to the ILAE consensus classification update of 2022. We trained a convolutional neuronal network (CNN) with fully digitalized WSI of hematoxylin–eosin stainings obtained from 125 patients covering the spectrum of mild malformation of cortical development (mMCD), mMCD with oligodendroglial hyperplasia in epilepsy (MOGHE), FCD ILAE Type 1a, 2a and 2b using 414 formalin-fixed and paraffin-embedded archival tissue blocks. An additional series of 198 postmortem tissue blocks from 59 patients without neurological disorders served as control to train the CNN for homotypic frontal, temporal and occipital areas and heterotypic Brodmann areas 4 and 17, entorhinal cortex and dentate gyrus. Special stains and immunohistochemical reactions were used to comprehensively annotate the region of interest. We then programmed a novel tile extraction pipeline and graphical dashboard to visualize all areas on the WSI recognized by the CNN. Our deep learning-based classifier is able to compute 1000 × 1000 µm large tiles and recognizes 25 anatomical regions and FCD categories with an accuracy of 98.8% (F1 score = 0.82). Microscopic review of regions predicted by the network confirmed these results. This deep learning-based classifier will be made available as online web application to support the differential histopathology diagnosis in neocortical human brain specimens obtained from epilepsy surgery. It will also serve as blueprint to build a digital histopathology slide suite addressing all major brain diseases encountered in patients with surgically amenable focal epilepsy.
Background: Processing whole-slide images (WSI) to train neural networks can be intricate and laborious. We developed an open-source library covering recurrent tasks in processing of WSI and in evaluating the performance of the trained networks for classification tasks. Methods: Two histopathology use-cases were selected. First we aimed to train a CNN to distinguish H&E-stained slides obtained from neuropathologically classified low-grade epilepsy-associated dysembryoplastic neuroepithelial tumor (DNET) and ganglioglioma (GG). The second project we trained a convolutional neural network (CNN) to predict the hormone expression of pituitary adenoms only from hematoxylin and eosin (H&E) stained slides. In the same approach, we addressed the issue to also predict clinically silent corticotroph adenoma. We included four clinico-pathological disease conditions in a multilabel approach. Results: Our best performing CNN achieved an area under the curve (AUC) of 0.97 for the receiver operating characteristic (ROC) for corticotroph adenoma, 0.86 for silent corticotroph adenoma and 0.98 for gonadotroph adenoma. Our DNET-GG classifier achieved an AUC of 1.00 for the ROC curve. All scores were calculated with the help of our library on predictions on a case basis. Conclusions: Our comprehensive library is most helpful to standardize the work-flow and minimize the work-burden in training CNN. It is also compatible with fastai. Indeed, our new CNNs reliably extracted neuropathologically relevant information from the H&E staining only. This approach will supplement the clinico-pathological diagnosis of brain tumors, which is currently based on cost-intense microscopic examination and variable panels of immunohistochemical stainings.
Background: Processing whole-slide images (WSI) to train neural networks can be intricate and labor intensive. We developed an open-source library dealing with recurrent tasks in the processing of WSI and helping with the training and evaluation of neuronal networks for classification tasks. Methods: Two histopathology use-cases were selected and only hematoxylin and eosin (H&E) stained slides were used. The first use case was a two-class classification problem. We trained a convolutional neuronal network (CNN) to distinguish between dysembryoplastic neuroepithelial tumor (DNET) and ganglioglioma (GG), two neuropathological low-grade epilepsy-associated tumor entities. Within the second use case, we included four clinicopathological disease conditions in a multilabel approach. Here we trained a CNN to predict the hormone expression profile of pituitary adenomas. In the same approach, we also predicted clinically silent corticotroph adenoma. Results: Our DNET-GG classifier achieved an AUC of 1.00 for the ROC curve. For the second use case, the best performing CNN achieved an area under the curve (AUC) of 0.97 for the receiver operating characteristic (ROC) for corticotroph adenoma, 0.86 for silent corticotroph adenoma, and 0.98 for gonadotroph adenoma. All scores were calculated with the help of our library on predictions on a case basis. Conclusions: Our comprehensive and fastai-compatible library is helpful to standardize the workflow and minimize the burden of training a CNN. Indeed, our trained CNNs extracted neuropathologically relevant information from the WSI. This approach will supplement the clinicopathological diagnosis of brain tumors, which is currently based on cost-intensive microscopic examination and variable panels of immunohistochemical stainings.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.