A Whole-Slide Image Managing Library Based on Fastai for Deep Learning in the Context of Histopathology: Two Use-Cases Explained

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.

A Whole Slide Image Managing Library Based on Fastai for Deep Learning in the Context of Histopathology: Two Use-Cases Explained

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.

A Whole Slide Image Managing Library Based on Fastai for Deep Learning in the Context of Histopathology: Two Use-Cases Explained

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.

UPLC-MS/MS-based Lipidomic Profiles Revealed Aberrant Lipids Associated with Invasiveness of Silent Corticotroph Adenoma

Context The accumulation of aberrant lipids and abnormal lipid metabolism in silent corticotroph adenomas (SCAs) could contribute to changes in clinical phenotypes, especially sphenoid sinus invasion. Objective To systematically investigate lipidomic and transcriptomic alterations associated with invasiveness and their potential molecular mechanisms in SCAs and to provide candidate biomarkers for predicting invasiveness and novel treatment options for invasive SCAs by targeting lipids. Methods Fifty-four SCAs (34 invasive/20 noninvasive) were subjected to lipidomic analysis based on ultraperformance liquid chromatography mass spectrometry, and 42 clinically nonfunctioning pituitary adenomas (23 invasive/19 noninvasive) were subjected to transcriptomic analysis. Differential analysis was performed to determine differential lipids and genes between invasive and noninvasive tumors. A functionally connected network was constructed with the molecular pathways as cores. Multiple machine learning methods were applied to identify the most critical lipids, which were further used to construct a lipidomic signature to predict invasive SCAs by multivariate logistic regression, and its performance was evaluated by receiver operating characteristic analysis. Results Twenty-eight differential lipids were identified, and a functionally connected network was constructed with 2 lipids, 17 genes, and 4 molecular pathways. Connectivity Map (CMap) analysis further revealed 32 potential drugs targeting 4 genes and related pathways. The 4 most critical lipids were identified as risk factors contributing to the invasive phenotype. A lipidomic signature was constructed and showed excellent performance in discriminating invasive and noninvasive SCAs. Conclusions The lipidomic signature could serve as a promising predictor for the invasive SCA phenotype and provide potential therapeutic targets for SCAs.

Postoperative expression of Cushing disease in a young male: metamorphosis of silent corticotroph adenoma?

Summary Silent corticotroph adenoma (SCA) is an unusual type of nonfunctioning pituitary adenoma (NFA) that is silent both clinically and biochemically and can only be recognized by positive immunostaining for ACTH. Under rare circumstances, it can transform into hormonally active disease presenting with severe Cushing syndrome. It might often produce diagnostic dilemma with difficult management issue if not thoroughly investigated and subtyped accordingly following surgery. Here, we present a 21-year-old male who initially underwent pituitary adenomectomy for presumed NFA with compressive symptoms. However, he developed recurrent and invasive macroadenoma with severe clinical as well as biochemical hypercortisolism during post-surgical follow-up. Repeat pituitary surgery was carried out urgently as there was significant optic chiasmal compression. Immunohistochemical analysis of the tumor tissue obtained on repeat surgery proved it to be an aggressive corticotroph adenoma. Though not cured, he showed marked clinical and biochemical improvement in the immediate postoperative period. Anticipating recurrence from the residual tumor, we referred him for cyber knife radio surgery. Learning points: Pituitary NFA commonly present with compressive symptoms such as headache and blurred vision. Post-surgical development of Cushing syndrome in such a case could be either drug induced or endogenous. In the presence of recurrent pituitary tumor, ACTH-dependent Cushing syndrome indicates CD. Rarely a SCA presenting initially as NFA can transform into an active corticotroph adenoma. Immunohistochemical marker for ACTH in the resected tumor confirms the diagnosis.

Pediatric silent corticotroph pituitary adenoma and role for proton therapy: case report

Tumors involving the sella are commonly craniopharyngiomas, optic pathway gliomas, or pituitary adenomas. Functioning adenomas are expected, with prolactinomas topping the differential. The authors present the case of a silent corticotroph adenoma, which has not been described in the pediatric population, and they detail the use of proton therapy, which is also novel.