Attribution-Driven Explanation of the Deep Neural Network Model via Conditional Microstructure Image Synthesis.
Ontology highlight
ABSTRACT:
SUBMITTER: Liu S
PROVIDER: S-EPMC8793074 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Cancer research has seen explosive development exploring deep learning (DL) techniques for analysing magnetic resonance imaging (MRI) images for predicting brain tumours. We have observed a substantial gap in explanation, interpretability, and high accuracy for DL models. Consequently, we propose an explanation-driven DL model by utilising a convolutional neural network (CNN), local interpretable model-agnostic explanation (LIME), and Shapley additive explanation (SHAP) for the prediction of discrete subtypes of brain tumours (meningioma, glioma, and pituitary) using an MRI image dataset. Unlike previous models, our model used a dual-input CNN approach to prevail over the classification challenge with images of inferior quality in terms of noise and metal artifacts by adding Gaussian noise. Our CNN training results reveal 94.64% accuracy as compared to other state-of-the-art methods. We used SHAP to ensure consistency and local accuracy for interpretation as Shapley values examine all future predictions applying all possible combinations of inputs. In contrast, LIME constructs sparse linear models around each prediction to illustrate how the model operates in the immediate area. Our emphasis for this study is interpretability and high accuracy, which is critical for realising disparities in predictive performance, helpful in developing trust, and essential in integration into clinical practice. The proposed method has a vast clinical application that could potentially be used for mass screening in resource-constraint countries.
| S-EPMC8964286 | biostudies-literature
Project description:The etiological underpinnings of many CNS disorders are not well understood. This is likely due to the fact that individual diseases aggregate numerous pathological subtypes, each associated with a complex landscape of genetic risk factors. To overcome these challenges, researchers are integrating novel data types from numerous patients, including imaging studies capturing broadly applicable features from patient-derived materials. These datasets, when combined with machine learning, potentially hold the power to elucidate the subtle patterns that stratify patients by shared pathology. In this study, we interrogated whether high-content imaging of primary skin fibroblasts, using the Cell Painting method, could reveal disease-relevant information among patients. First, we showed that technical features such as batch/plate type, plate, and location within a plate lead to detectable nuisance signals, as revealed by a pre-trained deep neural network and analysis with deep image embeddings. Using a plate design and image acquisition strategy that accounts for these variables, we performed a pilot study with 12 healthy controls and 12 subjects affected by the severe genetic neurological disorder spinal muscular atrophy (SMA), and evaluated whether a convolutional neural network (CNN) generated using a subset of the cells could distinguish disease states on cells from the remaining unseen control-SMA pair. Our results indicate that these two populations could effectively be differentiated from one another and that model selectivity is insensitive to batch/plate type. One caveat is that the samples were also largely separated by source. These findings lay a foundation for how to conduct future studies exploring diseases with more complex genetic contributions and unknown subtypes.
| S-EPMC6710615 | biostudies-literature
Project description:The secreting function of pituitary adenomas (PAs) plays a critical role in making the treatment strategies. However, Magnetic Resonance Imaging (MRI) analysis for pituitary adenomas is labor intensive and highly variable among radiologists. In this work, by applying convolutional neural network (CNN), we built a segmentation and classification model to help distinguish functioning pituitary adenomas from non-functioning subtypes with 3D MRI images from 185 patients with PAs (two centers). Specifically, the classification model adopts the concept of transfer learning and uses the pre-trained segmentation model to extract deep features from conventional MRI images. As a result, both segmentation and classification models obtained high performance in two internal validation datasets and an external testing dataset (for segmentation model: Dice score = 0.8188, 0.8091 and 0.8093 respectively; for classification model: AUROC = 0.8063, 0.7881 and 0.8478, respectively). In addition, the classification model considers the attention mechanism for better model interpretation. Taken together, this work provides the first deep learning-based tumor region segmentation and classification models of PAs, which enables early diagnosis and subtyping PAs from MRI images.
| S-EPMC8178077 | biostudies-literature
Project description:Recent advances in label-free virtual histology promise a new era for real-time molecular diagnosis in the operating room and during biopsy procedures. To take full advantage of the rich, multidimensional information provided by these technologies, reproducible and reliable computational tools that could facilitate the diagnosis are in great demand. In this study, we developed a deep-learning-based framework to recognize cancer versus normal human breast tissue from real-time label-free virtual histology images, with a tile-level AUC (area under receiver operating curve) of 95% and slide-level AUC of 100% on unseen samples. Furthermore, models trained on a high-quality laboratory-generated dataset can generalize to independent datasets acquired from a portable intraoperative version of the imaging technology with a physics-based adapted design. Classification activation maps and final feature visualization revealed discriminative patterns, such as tumor cells and tumor-associated vesicles, that are highly associated with cancer status. These results demonstrate that through the combination of real-time virtual histopathology and a deep-learning framework, accurate real-time diagnosis could be achieved in point-of-procedure clinical applications.
| S-EPMC6917773 | biostudies-literature
Project description:PurposeThe existence of multicontrast magnetic resonance (MR) images increases the level of clinical information available for the diagnosis and treatment of brain cancer patients. However, acquiring the complete set of multicontrast MR images is not always practically feasible. In this study, we developed a state-of-the-art deep learning convolutional neural network (CNN) for image-to-image translation across three standards MRI contrasts for the brain.MethodsBRATS'2018 MRI dataset of 477 patients clinically diagnosed with glioma brain cancer was used in this study, with each patient having T1-weighted (T1), T2-weighted (T2), and FLAIR contrasts. It was randomly split into 64%, 16%, and 20% as training, validation, and test set, respectively. We developed a U-Net model to learn the nonlinear mapping of a source image contrast to a target image contrast across three MRI contrasts. The model was trained and validated with 2D paired MR images using a mean-squared error (MSE) cost function, Adam optimizer with 0.001 learning rate, and 120 epochs with a batch size of 32. The generated synthetic-MR images were evaluated against the ground-truth images by computing the MSE, mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM).ResultsThe generated synthetic-MR images with our model were nearly indistinguishable from the real images on the testing dataset for all translations, except synthetic FLAIR images had slightly lower quality and exhibited loss of details. The range of average PSNR, MSE, MAE, and SSIM values over the six translations were 29.44-33.25 dB, 0.0005-0.0012, 0.0086-0.0149, and 0.932-0.946, respectively. Our results were as good as the best-reported results by other deep learning models on BRATS datasets.ConclusionsOur U-Net model exhibited that it can accurately perform image-to-image translation across brain MRI contrasts. It could hold great promise for clinical use for improved clinical decision-making and better diagnosis of brain cancer patients due to the availability of multicontrast MRIs. This approach may be clinically relevant and setting a significant step to efficiently fill a gap of absent MR sequences without additional scanning.
| S-EPMC8992958 | biostudies-literature
Project description:An important goal in neuroscience is to elucidate the causal relationships between the brain's different regions. This can help reveal the brain's functional circuitry and diagnose lesions. Currently there are a lack of approaches to functional connectome estimation that leverage the state-of-the-art in deep learning architectures and training methodologies. Therefore, we propose a new framework based on a vector auto-regressive deep neural network (VARDNN) architecture. Our approach consists of a set of nodes, each with a deep neural network structure. These nodes can be mapped to any spatial sub-division based on the data to be analyzed, such as anatomical brain regions from which representative neural signals can be obtained. VARDNN learns to reproduce experimental time series data using modern deep learning training techniques. Based on this, we developed two novel directed functional connectivity (dFC) measures, namely VARDNN-DI and VARDNN-GC. We evaluated our measures against a number of existing functional connectome estimation measures, such as partial correlation and multivariate Granger causality combined with large dimensionality counter-measure techniques. Our measures outperformed them across various types of ground truth data, especially as the number of nodes increased. We applied VARDNN to fMRI data to compare the dFC between 41 healthy control vs. 32 Alzheimer's disease subjects. Our VARDNN-DI measure detected lesioned regions consistent with previous studies and separated the two groups well in a subject-wise evaluation framework. Summarily, the VARDNN framework has powerful capabilities for whole brain dFC estimation. We have implemented VARDNN as an open-source toolbox that can be freely downloaded for researchers who wish to carry out functional connectome analysis on their own data.
| S-EPMC8651499 | biostudies-literature
Project description:This study aimed to investigate deep convolutional neural network- (DCNN-) based artificial intelligence (AI) model using cephalometric images for the classification of sagittal skeletal relationships and compare the performance of the newly developed DCNN-based AI model with that of the automated-tracing AI software. A total of 1574 cephalometric images were included and classified based on the A-point-Nasion- (N-) point-B-point (ANB) angle (Class I being 0-4°, Class II > 4°, and Class III < 0°). The DCNN-based AI model was developed using training (1334 images) and validation (120 images) sets with a standard classification label for the individual images. A test set of 120 images was used to compare the AI models. The agreement of the DCNN-based AI model or the automated-tracing AI software with a standard classification label was measured using Cohen's kappa coefficient (0.913 for the DCNN-based AI model; 0.775 for the automated-tracing AI software). In terms of their performances, the micro-average values of the DCNN-based AI model (sensitivity, 0.94; specificity, 0.97; precision, 0.94; accuracy, 0.96) were higher than those of the automated-tracing AI software (sensitivity, 0.85; specificity, 0.93; precision, 0.85; accuracy, 0.90). With regard to the sagittal skeletal classification using cephalometric images, the DCNN-based AI model outperformed the automated-tracing AI software.
| S-EPMC9270345 | biostudies-literature
Project description:Most multicellular organisms require apoptosis, or programmed cell death, to function properly and survive. On the other hand, morphological and biochemical characteristics of apoptosis have remained remarkably consistent throughout evolution. Apoptosis is thought to have at least three functionally distinct phases: induction, effector, and execution. Recent studies have revealed that reactive oxygen species (ROS) and the oxidative stress could play an essential role in apoptosis. Advanced microscopic imaging techniques allow biologists to acquire an extensive amount of cell images within a matter of minutes which rule out the manual analysis of image data acquisition. The segmentation of cell images is often considered the cornerstone and central problem for image analysis. Currently, the issue of segmentation of mitochondrial cell images via deep learning receives increasing attention. The manual labeling of cell images is time-consuming and challenging to train a pro. As a courtesy method, mitochondrial cell imaging (MCI) is proposed to identify the normal, drug-treated, and diseased cells. Furthermore, cell movement (fission and fusion) is measured to evaluate disease risk. The newly proposed drug-treated, normal, and diseased image segmentation (DNDIS) algorithm can quickly segment mitochondrial cell images without supervision and further segment the highly drug-treated cells in the picture, i.e., normal, diseased, and drug-treated cells. The proposed method is based on the ResNet-50 deep learning algorithm. The dataset consists of 414 images mainly categorised into different sets (drug, diseased, and normal) used microscopically. The proposed automated segmentation method has outperformed and secured high precision (90%, 92%, and 94%); moreover, it also achieves proper training. This study will benefit medicines and diseased cell measurements in medical tests and clinical practices.
| S-EPMC9162846 | biostudies-literature
Project description:A variety of deep neural network (DNN)-based image denoising methods have been proposed for use with medical images. Traditional measures of image quality (IQ) have been employed to optimize and evaluate these methods. However, the objective evaluation of IQ for the DNN-based denoising methods remains largely lacking. In this work, we evaluate the performance of DNN-based denoising methods by use of task-based IQ measures. Specifically, binary signal detection tasks under signal-known-exactly (SKE) with background-known-statistically (BKS) conditions are considered. The performance of the ideal observer (IO) and common linear numerical observers are quantified and detection efficiencies are computed to assess the impact of the denoising operation on task performance. The numerical results indicate that, in the cases considered, the application of a denoising network can result in a loss of task-relevant information in the image. The impact of the depth of the denoising networks on task performance is also assessed. The presented results highlight the need for the objective evaluation of IQ for DNN-based denoising technologies and may suggest future avenues for improving their effectiveness in medical imaging applications.
| S-EPMC8673589 | biostudies-literature
Project description:Attribution studies on recent global warming by Global Climate Model (GCM) ensembles converge in showing the fundamental role of anthropogenic forcings as primary drivers of temperature in the last half century. However, despite their differences, all these models pertain to the same dynamical approach and come from a common ancestor, so that their very similar results in attribution studies are not surprising and cannot be considered as a clear proof of robustness of the results themselves. Thus, here we adopt a completely different, non-dynamical, data-driven and fully nonlinear approach to the attribution problem. By means of neural network (NN) modelling, and analysing the last 160 years, we perform attribution experiments and find that the strong increase in global temperature of the last half century may be attributed basically to anthropogenic forcings (with details on their specific contributions), while the Sun considerably influences the period 1910-1975. Furthermore, the role of sulphate aerosols and Atlantic Multidecadal Oscillation for better catching interannual to decadal temperature variability is clarified. Sensitivity analyses to forcing changes are also performed. The NN outcomes both corroborate our previous knowledge from GCMs and give new insight into the relative contributions of external forcings and internal variability to climate.
| S-EPMC5732275 | biostudies-literature