Image-driven classification of functioning and nonfunctioning pituitary adenoma by deep convolutional neural networks.
Ontology highlight
ABSTRACT:
SUBMITTER: Li H
PROVIDER: S-EPMC8178077 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Wheat blast is a threat to global wheat production, and limited blast-resistant cultivars are available. The current estimations of wheat spike blast severity rely on human assessments, but this technique could have limitations. Reliable visual disease estimations paired with Red Green Blue (RGB) images of wheat spike blast can be used to train deep convolutional neural networks (CNN) for disease severity (DS) classification. Inter-rater agreement analysis was used to measure the reliability of who collected and classified data obtained under controlled conditions. We then trained CNN models to classify wheat spike blast severity. Inter-rater agreement analysis showed high accuracy and low bias before model training. Results showed that the CNN models trained provide a promising approach to classify images in the three wheat blast severity categories. However, the models trained on non-matured and matured spikes images showing the highest precision, recall, and F1 score when classifying the images. The high classification accuracy could serve as a basis to facilitate wheat spike blast phenotyping in the future.
| S-EPMC8248543 | biostudies-literature
Project description:Celiac disease (CD) is a gluten-sensitive immune-mediated enteropathy. This proof-of-concept study used a convolutional neural network (CNN) to classify hematoxylin and eosin (H&E) CD histological images, normal small intestine control, and non-specified duodenal inflammation (7294, 11,642, and 5966 images, respectively). The trained network classified CD with high performance (accuracy 99.7%, precision 99.6%, recall 99.3%, F1-score 99.5%, and specificity 99.8%). Interestingly, when the same network (already trained for the 3 class images), analyzed duodenal adenocarcinoma (3723 images), the new images were classified as duodenal inflammation in 63.65%, small intestine control in 34.73%, and CD in 1.61% of the cases; and when the network was retrained using the 4 histological subtypes, the performance was above 99% for CD and 97% for adenocarcinoma. Finally, the model added 13,043 images of Crohn's disease to include other inflammatory bowel diseases; a comparison between different CNN architectures was performed, and the gradient-weighted class activation mapping (Grad-CAM) technique was used to understand why the deep learning network made its classification decisions. In conclusion, the CNN-based deep neural system classified 5 diagnoses with high performance. Narrow artificial intelligence (AI) is designed to perform tasks that typically require human intelligence, but it operates within limited constraints and is task-specific.
| S-EPMC11355344 | biostudies-literature
Project description:Spiking neural networks (SNNs) can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-valued deep Convolutional Neural Networks (CNNs) can be converted into accurate spiking equivalents. These networks did not include certain common operations such as max-pooling, softmax, batch-normalization and Inception-modules. This paper presents spiking equivalents of these operations therefore allowing conversion of nearly arbitrary CNN architectures. We show conversion of popular CNN architectures, including VGG-16 and Inception-v3, into SNNs that produce the best results reported to date on MNIST, CIFAR-10 and the challenging ImageNet dataset. SNNs can trade off classification error rate against the number of available operations whereas deep continuous-valued neural networks require a fixed number of operations to achieve their classification error rate. From the examples of LeNet for MNIST and BinaryNet for CIFAR-10, we show that with an increase in error rate of a few percentage points, the SNNs can achieve more than 2x reductions in operations compared to the original CNNs. This highlights the potential of SNNs in particular when deployed on power-efficient neuromorphic spiking neuron chips, for use in embedded applications.
| S-EPMC5770641 | biostudies-literature
Project description:BackgroundDecreased heart rate variability (HRV) has been reported to be associated with cardiac autonomic dysfunction. Hypopituitarism in nonfunctioning pituitary adenoma (NFPA) is often linked to increased cardiovascular mortality. We therefore hypothesized that postoperative NFPA patients with hormone deficiency have an elevated risk of HRV alterations indicating cardiac autonomic dysfunction.MethodsA total of 22 patients with NFPA were enrolled in the study. Between 3 and 6 months after surgery, a combined pituitary function test (CPFT) was performed, and HRV was measured. The period of sleep before the CPFT was deemed the most stable period, and the hypoglycemic period that occurred during the CPFT was defined as the most unstable period. Changes in HRV parameters in stable and unstable periods were observed and compared depending on the status of hormone deficiencies.ResultsIn patients with adrenocorticotropic hormone (ACTH) deficiency with other pituitary hormone deficiencies, the low frequency to high frequency ratio, which represents overall autonomic function and is increased in the disease state, was higher (P=0.005). Additionally, the standard deviation of the normal-to-normal interval, which decreases in the autonomic dysfunction state, was lower (P=0.030) during the hypoglycemic period. In panhypopituitarism, the low frequency to high frequency ratio during the hypoglycemic period was increased (P=0.007).ConclusionHRV analysis during CPFT enables estimation of cardiac autonomic dysfunction in patients with NFPA who develop ACTH deficiency with other pituitary hormone deficiencies or panhypopituitarism after surgery. These patients may require a preemptive assessment of cardiovascular risk.
| S-EPMC8258319 | biostudies-literature
Project description:The Machine Recognition of Crystallization Outcomes (MARCO) initiative has assembled roughly half a million annotated images of macromolecular crystallization experiments from various sources and setups. Here, state-of-the-art machine learning algorithms are trained and tested on different parts of this data set. We find that more than 94% of the test images can be correctly labeled, irrespective of their experimental origin. Because crystal recognition is key to high-density screening and the systematic analysis of crystallization experiments, this approach opens the door to both industrial and fundamental research applications.
| S-EPMC6010233 | biostudies-other
Project description:The goal of this study is to evaluate the efficacy of deep convolutional neural networks (DCNNs) in differentiating subtle, intermediate, and more obvious image differences in radiography. Three different datasets were created, which included presence/absence of the endotracheal (ET) tube (n = 300), low/normal position of the ET tube (n = 300), and chest/abdominal radiographs (n = 120). The datasets were split into training, validation, and test. Both untrained and pre-trained deep neural networks were employed, including AlexNet and GoogLeNet classifiers, using the Caffe framework. Data augmentation was performed for the presence/absence and low/normal ET tube datasets. Receiver operating characteristic (ROC), area under the curves (AUC), and 95% confidence intervals were calculated. Statistical differences of the AUCs were determined using a non-parametric approach. The pre-trained AlexNet and GoogLeNet classifiers had perfect accuracy (AUC 1.00) in differentiating chest vs. abdominal radiographs, using only 45 training cases. For more difficult datasets, including the presence/absence and low/normal position endotracheal tubes, more training cases, pre-trained networks, and data-augmentation approaches were helpful to increase accuracy. The best-performing network for classifying presence vs. absence of an ET tube was still very accurate with an AUC of 0.99. However, for the most difficult dataset, such as low vs. normal position of the endotracheal tube, DCNNs did not perform as well, but achieved a reasonable AUC of 0.81.
| S-EPMC5537094 | biostudies-other
Project description:The tumor-stroma ratio (TSR) reflected on hematoxylin and eosin (H&E)-stained histological images is a potential prognostic factor for survival. Automatic image processing techniques that allow for high-throughput and precise discrimination of tumor epithelium and stroma are required to elevate the prognostic significance of the TSR. As a variant of deep learning techniques, transfer learning leverages nature-images features learned by deep convolutional neural networks (CNNs) to relieve the requirement of deep CNNs for immense sample size when handling biomedical classification problems. Herein we studied different transfer learning strategies for accurately distinguishing epithelial and stromal regions of H&E-stained histological images acquired from either breast or ovarian cancer tissue. We compared the performance of important deep CNNs as either a feature extractor or as an architecture for fine-tuning with target images. Moreover, we addressed the current contradictory issue about whether the higher-level features would generalize worse than lower-level ones because they are more specific to the source-image domain. Under our experimental setting, the transfer learning approach achieved an accuracy of 90.2 (vs. 91.1 for fine tuning) with GoogLeNet, suggesting the feasibility of using it in assisting pathology-based binary classification problems. Our results also show that the superiority of the lower-level or the higher-level features over the other ones was determined by the architecture of deep CNNs.
| S-EPMC6286645 | biostudies-literature
Project description:Three-dimensional (3D) medical image classification is useful in applications such as disease diagnosis and content-based medical image retrieval. It is a challenging task due to several reasons. First, image intensity values are vastly different depending on the image modality. Second, intensity values within the same image modality may vary depending on the imaging machine and artifacts may also be introduced in the imaging process. Third, processing 3D data requires high computational power. In recent years, significant research has been conducted in the field of 3D medical image classification. However, most of these make assumptions about patient orientation and imaging direction to simplify the problem and/or work with the full 3D images. As such, they perform poorly when these assumptions are not met. In this paper, we propose a method of classification for 3D organ images that is rotation and translation invariant. To this end, we extract a representative two-dimensional (2D) slice along the plane of best symmetry from the 3D image. We then use this slice to represent the 3D image and use a 20-layer deep convolutional neural network (DCNN) to perform the classification task. We show experimentally, using multi-modal data, that our method is comparable to existing methods when the assumptions of patient orientation and viewing direction are met. Notably, it shows similarly high accuracy even when these assumptions are violated, where other methods fail. We also explore how this method can be used with other DCNN models as well as conventional classification approaches.
| S-EPMC7924426 | biostudies-literature
Project description:BackgroundNonfunctioning pituitary adenoma (NFPA) and growth hormone pituitary adenoma (GHPA) are major subtypes of pituitary adenomas (PAs). The primary treatment is surgical resection. However, radical excision remains challenging, and few effective medical therapies are available. It is urgent to find novel targets for the treatment. Bromodomain-containing protein 4 (BRD4) is an epigenetic regulator that leads to aberrant transcriptional activation of oncogenes. Herein, we investigated the pathological role of BRD4 and evaluated the effectiveness of BRD4 inhibitors in the treatment of NFPA and GHPA.MethodsThe expression of BRD4 was detected in NFPA, GHPA, and normal pituitary tissues. The efficacies of BRD4 inhibitors were evaluated in GH3 and MMQ cell lines, patient-derived tumor cells, and in vivo mouse xenograft models of PA. Standard western blots, real-time PCR, and flow cytometry experiments were performed to investigate the effect of BRD4 inhibitors on cell cycle progression, apoptosis, and the expression patterns of downstream genes.ResultsImmunohistochemistry studies demonstrated the overexpression of BRD4 in NFPA and GHPA. In vitro and in vivo studies showed that treatment with the BRD4 inhibitor ZBC-260 significantly inhibited the proliferation of PA cells. Further mechanistic studies revealed that ZBC-260 could downregulate the expression of c-Myc, B-cell lymphoma 2 (Bcl2), and related genes, which are vital factors in pituitary tumorigenesis.ConclusionIn this study, we determined the overexpression of BRD4 in NFPA and GHPA and assessed the effects of BRD4 inhibitors on PA cells in vitro and in vivo. Our findings suggest that BRD4 is a promising therapeutic target for NFPA and GHPA.
| S-EPMC7594556 | biostudies-literature
Project description:Introduction:Deep learning neural networks are especially potent at dealing with structured data, such as images and volumes. Both modified LiviaNET and HyperDense-Net performed well at a prior competition segmenting 6-month-old infant magnetic resonance images, but neonatal cerebral tissue type identification is challenging given its uniquely inverted tissue contrasts. The current study aims to evaluate the two architectures to segment neonatal brain tissue types at term equivalent age. Methods:Both networks were retrained over 24 pairs of neonatal T1 and T2 data from the Developing Human Connectome Project public data set and validated on another eight pairs against ground truth. We then reported the best-performing model from training and its performance by computing the Dice similarity coefficient (DSC) for each tissue type against eight test subjects. Results:During the testing phase, among the segmentation approaches tested, the dual-modality HyperDense-Net achieved the best statistically significantly test mean DSC values, obtaining 0.94/0.95/0.92 for the tissue types and took 80 h to train and 10 min to segment, including preprocessing. The single-modality LiviaNET was better at processing T2-weighted images than processing T1-weighted images across all tissue types, achieving mean DSC values of 0.90/0.90/0.88 for gray matter, white matter, and cerebrospinal fluid, respectively, while requiring 30 h to train and 8 min to segment each brain, including preprocessing. Discussion:Our evaluation demonstrates that both neural networks can segment neonatal brains, achieving previously reported performance. Both networks will be continuously retrained over an increasingly larger repertoire of neonatal brain data and be made available through the Canadian Neonatal Brain Platform to better serve the neonatal brain imaging research community.
| S-EPMC7114297 | biostudies-literature