RESUMEN

Objective: To achieve an accurate assessment of orthodontic and restorative treatments, tooth segmentation of dental panoramic X-ray images is a critical preliminary step, however, dental panoramic X-ray images suffer from poorly defined interdental boundaries and low root-to-alveolar bone contrast, which pose significant challenges to tooth segmentation. In this article, we propose a multi-feature coordinate position learning-based tooth image segmentation method for tooth segmentation. Methods: For better analysis, the input image is randomly flipped horizontally and vertically to enhance the data. Our method extracts multi-scale tooth features from the designed residual omni-dimensional dynamic convolution and the designed two-stream coordinate attention module can further complement the tooth boundary features, and finally the two features are fused to enhance the local details of the features and global contextual information, which achieves the enrichment and optimization of the feature information. Results: The publicly available adult dental datasets Archive and Dataset and Code were used in the study. The experimental results were 87.96% and 92.04% for IoU, 97.79% and 97.32% for ACC, and 86.42% and 95.64% for Dice. Conclusion: The experimental results show that the proposed network can be used to assist doctors in quickly viewing tooth positions, and we also validate the effectiveness of the proposed two modules in fusing features.

RESUMEN

Automatic seizure detection has significant value in epilepsy diagnosis and treatment. Although a variety of deep learning models have been proposed to automatically learn electroencephalography (EEG) features for seizure detection, the generalization performance and computational burden of such deep models remain the bottleneck of practical application. In this study, a novel lightweight model based on random convolutional kernel transform (ROCKET) is developed for EEG feature learning for seizure detection. Specifically, random convolutional kernels are embedded into the structure of a wavelet scattering network instead of original wavelet transform convolutions. Then the significant EEG features are selected from the scattering coefficients and convolutional outputs by analysis of variance (ANOVA) and minimum redundancy-maximum relevance (MRMR) methods. This model not only preserves the merits of the fast-training process from ROCKET, but also provides insight into seizure detection by retaining only the helpful channels. The extreme gradient boosting (XGboost) classifier was combined with this EEG feature learning model to build a comprehensive seizure detection system that achieved promising epoch-based results, with over 90% of both sensitivity and specificity on the scalp and intracranial EEG databases. The experimental comparisons showed that the proposed method outperformed other state-of-the-art methods for cross-patient and patient-specific seizure detection.

Asunto(s)

RESUMEN

BACKGROUND: Ensuring the efficient recognition and management of crop pests is crucial for maintaining the balance in global agricultural ecosystems and ecological harmony. Deep learning-based methods have shown promise in crop pest recognition. However, prevailing methods often fail to address a critical issue: biased pest training dataset distribution stemming from the tendency to collect images primarily in certain environmental contexts, such as paddy fields. This oversight hampers recognition accuracy when encountering pest images dissimilar to training samples, highlighting the need for a novel approach to overcome this limitation. RESULTS: We introduce the Decoupled Feature Learning (DFL) framework, leveraging causal inference techniques to handle training dataset bias. DFL manipulates the training data based on classification confidence to construct different training domains and employs center triplet loss for learning class-core features. The proposed DFL framework significantly boosts existing baseline models, attaining unprecedented recognition accuracies of 95.33%, 92.59%, and 74.86% on the Li, DFSPD, and IP102 datasets, respectively. CONCLUSION: Extensive testing on three pest datasets using standard baseline models demonstrates the superiority of DFL in pest recognition. The visualization results show that DFL encourages the baseline models to capture the class-core features. The proposed DFL marks a pivotal step in mitigating the issue of data distribution bias, enhancing the reliability of deep learning in agriculture. © 2024 Society of Chemical Industry.

RESUMEN

PURPOSE: Clinical needle insertion into tissue, commonly assisted by 2D ultrasound imaging for real-time navigation, faces the challenge of precise needle and probe alignment to reduce out-of-plane movement. Recent studies investigate 3D ultrasound imaging together with deep learning to overcome this problem, focusing on acquiring high-resolution images to create optimal conditions for needle tip detection. However, high-resolution also requires a lot of time for image acquisition and processing, which limits the real-time capability. Therefore, we aim to maximize the US volume rate with the trade-off of low image resolution. We propose a deep learning approach to directly extract the 3D needle tip position from sparsely sampled US volumes. METHODS: We design an experimental setup with a robot inserting a needle into water and chicken liver tissue. In contrast to manual annotation, we assess the needle tip position from the known robot pose. During insertion, we acquire a large data set of low-resolution volumes using a 16  ×  16 element matrix transducer with a volume rate of 4 Hz. We compare the performance of our deep learning approach with conventional needle segmentation. RESULTS: Our experiments in water and liver show that deep learning outperforms the conventional approach while achieving sub-millimeter accuracy. We achieve mean position errors of 0.54 mm in water and 1.54 mm in liver for deep learning. CONCLUSION: Our study underlines the strengths of deep learning to predict the 3D needle positions from low-resolution ultrasound volumes. This is an important milestone for real-time needle navigation, simplifying the alignment of needle and ultrasound probe and enabling a 3D motion analysis.

RESUMEN

Formulating the methodology of machine learning by bilevel optimization techniques provides a new perspective to understand and solve automated machine learning problems.

RESUMEN

Objective.Head and neck radiotherapy planning requires electron densities from different tissues for dose calculation. Dose calculation from imaging modalities such as MRI remains an unsolved problem since this imaging modality does not provide information about the density of electrons.Approach.We propose a generative adversarial network (GAN) approach that synthesizes CT (sCT) images from T1-weighted MRI acquisitions in head and neck cancer patients. Our contribution is to exploit new features that are relevant for improving multimodal image synthesis, and thus improving the quality of the generated CT images. More precisely, we propose a Dual branch generator based on the U-Net architecture and on an augmented multi-planar branch. The augmented branch learns specific 3D dynamic features, which describe the dynamic image shape variations and are extracted from different view-points of the volumetric input MRI. The architecture of the proposed model relies on an end-to-end convolutional U-Net embedding network.Results.The proposed model achieves a mean absolute error (MAE) of18.76±5.167in the target Hounsfield unit (HU) space on sagittal head and neck patients, with a mean structural similarity (MSSIM) of0.95±0.09and a Frechet inception distance (FID) of145.60±8.38. The model yields a MAE of26.83±8.27to generate specific primary tumor regions on axial patient acquisitions, with a Dice score of0.73±0.06and a FID distance equal to122.58±7.55. The improvement of our model over other state-of-the-art GAN approaches is of 3.8%, on a tumor test set. On both sagittal and axial acquisitions, the model yields the best peak signal-to-noise ratio of27.89±2.22and26.08±2.95to synthesize MRI from CT input.Significance.The proposed model synthesizes both sagittal and axial CT tumor images, used for radiotherapy treatment planning in head and neck cancer cases. The performance analysis across different imaging metrics and under different evaluation strategies demonstrates the effectiveness of our dual CT synthesis model to produce high quality sCT images compared to other state-of-the-art approaches. Our model could improve clinical tumor analysis, in which a further clinical validation remains to be explored.

Asunto(s)

RESUMEN

To address challenges related to the inadequate representation and inaccurate discrimination of pedestrian attributes, we propose a novel method for person re-identification, which leverages global feature learning and classification optimization. Specifically, this approach integrates a Normalization-based Channel Attention Module into the fundamental ResNet50 backbone, utilizing a scaling factor to prioritize and enhance key pedestrian feature information. Furthermore, dynamic activation functions are employed to adaptively modulate the parameters of ReLU based on the input convolutional feature maps, thereby bolstering the nonlinear expression capabilities of the network model. By incorporating Arcface loss into the cross-entropy loss, the supervised model is trained to learn pedestrian features that exhibit significant inter-class variance while maintaining tight intra-class coherence. The evaluation of the enhanced model on two popular datasets, Market1501 and DukeMTMC-ReID, reveals improvements in Rank-1 accuracy by 1.28% and 1.4%, respectively, along with corresponding gains in the mean average precision (mAP) of 1.93% and 1.84%. These findings indicate that the proposed model is capable of extracting more robust pedestrian features, enhancing feature discriminability, and ultimately achieving superior recognition accuracy.

RESUMEN

This work develops a generalizable neural network, SENSORNET, for sensor feature learning across various applications. The primary challenge addressed is the poor portability of pretrained neural networks to new applications with limited sensor data. To solve this challenge, we design SensorNet, which integrates the flexibility of self-attention with multi-scale feature locality of convolution. Moreover, we invent patch-wise self-attention with stacked multi-heads to enrich the sensor feature representation. SensorNet is generalizable to pervasive applications with any number of sensor inputs, and is much smaller than the state-of-the-art self-attention and convolution hybrid baseline (0.83 M vs. 3.87 M parameters) with similar performance. The experimental results show that SensorNet is able to achieve state-of-the-art performance compared with the top five models on a competition activity recognition dataset (SHL'18). Moreover, pretrained SensorNet in a large inertial measurement unit (IMU) dataset can be fine-tuned to achieve the best accuracy on a much smaller IMU dataset (up to 5% improvement in WISDM) and to achieve the state-of-the-art performance on an EEG dataset (SLEEP-EDF-20), showing the strong generalizability of our approach.

RESUMEN

Finding the most interesting areas of an image is the aim of saliency detection. Conventional methods based on low-level features rely on biological cues like texture and color. These methods, however, have trouble with processing complicated or low-contrast images. In this paper, we introduce a deep neural network-based saliency detection method. First, using semantic segmentation, we construct a pixel-level model that gives each pixel a saliency value depending on its semantic category. Next, we create a region feature model by combining both hand-crafted and deep features, which extracts and fuses the local and global information of each superpixel region. Third, we combine the results from the previous two steps, along with the over-segmented superpixel images and the original images, to construct a multi-level feature model. We feed the model into a deep convolutional network, which generates the final saliency map by learning to integrate the macro and micro information based on the pixels and superpixels. We assess our method on five benchmark datasets and contrast it against 14 state-of-the-art saliency detection algorithms. According to the experimental results, our method performs better than the other methods in terms of F-measure, precision, recall, and runtime. Additionally, we analyze the limitations of our method and propose potential future developments.

RESUMEN

Medical images play a vital role in medical analysis by providing crucial information about patients' pathological conditions. However, the quality of these images can be compromised by many factors, such as limited resolution of the instruments, artifacts caused by movements, and the complexity of the scanned areas. As a result, low-resolution (LR) images cannot provide sufficient information for diagnosis. To address this issue, researchers have attempted to apply image super-resolution (SR) techniques to restore the high-resolution (HR) images from their LR counterparts. However, these techniques are designed for generic images, and thus suffer from many challenges unique to medical images. An obvious one is the diversity of the scanned objects; for example, the organs, tissues, and vessels typically appear in different sizes and shapes, and are thus hard to restore with standard convolution neural networks (CNNs). In this paper, we develop a dynamic-local learning framework to capture the details of these diverse areas, consisting of deformable convolutions with adjustable kernel shapes. Moreover, the global information between the tissues and organs is vital for medical diagnosis. To preserve global information, we propose pixel-pixel and patch-patch global learning using a non-local mechanism and a vision transformer (ViT), respectively. The result is a novel CNN-ViT neural network with Local-to-Global feature learning for medical image SR, referred to as LGSR, which can accurately restore both local details and global information. We evaluate our method on six public datasets and one large-scale private dataset, which include five different types of medical images (i.e., Ultrasound, OCT, Endoscope, CT, and MRI images). Experiments show that the proposed method achieves superior PSNR/SSIM and visual performance than the state of the arts with competitive computational costs, measured in network parameters, runtime, and FLOPs. What is more, the experiment conducted on OCT image segmentation for the downstream task demonstrates a significantly positive performance effect of LGSR.

Asunto(s)

RESUMEN

Background. Pancreatic cancer is one of the most malignant tumours, demonstrating a poor prognosis and nearly identically high mortality and morbidity, mainly because of the difficulty of early diagnosis and timely treatment for localized stages.Objective. To develop a noncontrast CT (NCCT)-based pancreatic lesion detection model that could serve as an intelligent tool for diagnosing pancreatic cancer early, overcoming the challenges associated with low contrast intensities and complex anatomical structures present in NCCT images.Approach.We design a multiscale and multiperception (MSMP) feature learning network with ResNet50 coupled with a feature pyramid network as the backbone for strengthening feature expressions. We added multiscale atrous convolutions to expand different receptive fields, contextual attention to perceive contextual information, and channel and spatial attention to focus on important channels and spatial regions, respectively. The MSMP network then acts as a feature extractor for proposing an NCCT-based pancreatic lesion detection model with image patches covering the pancreas as its input; Faster R-CNN is employed as the detection method for accurately detecting pancreatic lesions.Main results. By using the new MSMP network as a feature extractor, our model outperforms the conventional object detection algorithms in terms of the recall (75.40% and 90.95%), precision (40.84% and 68.21%), F1 score (52.98% and 77.96%), F2 score (64.48% and 85.26%) and Ap50 metrics (53.53% and 70.14%) at the image and patient levels, respectively.Significance.The good performance of our new model implies that MSMP can mine NCCT imaging features for detecting pancreatic lesions from complex backgrounds well. The proposed detection model is expected to be further developed as an intelligent method for the early detection of pancreatic cancer.

Asunto(s)

RESUMEN

Radiomics has been widely recognized for its effectiveness in decoding tumor phenotypes through the extraction of quantitative imaging features. However, the robustness of radiomic methods to estimate clinically relevant biomarkers non-invasively remains largely untested. In this study, we propose Cascaded Data Processing Network (CDPNet), a radiomic feature learning method to predict tumor molecular status from medical images. We apply CDPNet to an epigenetic case, specifically targeting the estimation of O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation from Magnetic Resonance Imaging (MRI) scans of glioblastoma patients. CDPNet has three components: 1) Principal Component Analysis (PCA), 2) Fisher Linear Discriminant (FLD), and 3) a combination of hashing and blockwise histograms. The outlined architectural framework capitalizes on PCA to reconstruct input image patches, followed by FLD to extract discriminative filter banks, and finally using binary hashing and blockwise histogram module for indexing, pooling, and feature generation. To validate the effectiveness of CDPNet, we conducted an exhaustive evaluation on a comprehensive retrospective cohort comprising 484 IDH-wildtype glioblastoma patients with pre-operative multi-parametric MRI scans (T1, T1-Gd, T2, and T2-FLAIR). The prediction of MGMT promoter methylation status was cast as a binary classification problem. The developed model underwent rigorous training via 10-fold cross-validation on a discovery cohort of 446 patients. Subsequently, the model's performance was evaluated on a distinct and previously unseen replication cohort of 38 patients. Our method achieved an accuracy of 70.11% and an area under the curve of 0.71 (95% CI: 0.65 - 0.74).

RESUMEN

Introduction: A growing body of research indicates that microorganisms play a crucial role in human health. Imbalances in microbial communities are closely linked to human diseases, and identifying potential relationships between microbes and diseases can help elucidate the pathogenesis of diseases. However, traditional methods based on biological or clinical experiments are costly, so the use of computational models to predict potential microbe-disease associations is of great importance. Methods: In this paper, we present a novel computational model called MLFLHMDA, which is based on a Multi-View Latent Feature Learning approach to predict Human potential Microbe-Disease Associations. Specifically, we compute Gaussian interaction profile kernel similarity between diseases and microbes based on the known microbe-disease associations from the Human Microbe-Disease Association Database and perform a preprocessing step on the resulting microbe-disease association matrix, namely, weighting K nearest known neighbors (WKNKN) to reduce the sparsity of the microbe-disease association matrix. To obtain unobserved associations in the microbe and disease views, we extract different latent features based on the geometrical structure of microbes and diseases, and project multi-modal latent features into a common subspace. Next, we introduce graph regularization to preserve the local manifold structure of Gaussian interaction profile kernel similarity and add Lp,q-norms to the projection matrix to ensure the interpretability and sparsity of the model. Results: The AUC values for global leave-one-out cross-validation and 5-fold cross validation implemented by MLFLHMDA are 0.9165 and 0.8942+/-0.0041, respectively, which perform better than other existing methods. In addition, case studies of different diseases have demonstrated the superiority of the predictive power of MLFLHMDA. The source code of our model and the data are available on https://github.com/LiangzheZhang/MLFLHMDA_master.

RESUMEN

Objective.Cardiovascular magnetic resonance (CMR) can measure T1 and T2 relaxation times for myocardial tissue characterization. However, the CMR procedure for T1/T2 parametric mapping is time-consuming, making it challenging to scan heart patients routinely in clinical practice. This study aims to accelerate CMR parametric mapping with deep learning.Approach. A deep-learning model, SwinUNet, was developed to accelerate T1/T2 mapping. SwinUNet used a convolutional UNet and a Swin transformer to form a hierarchical 3D computation structure, allowing for analyzing CMR images spatially and temporally with multiscale feature learning. A comparative study was conducted between SwinUNet and an existing deep-learning model, MyoMapNet, which only used temporal analysis for parametric mapping. The T1/T2 mapping performance was evaluated globally using mean absolute error (MAE) and structural similarity index measure (SSIM). The clinical T1/T2 indices for characterizing the left-ventricle myocardial walls were also calculated and evaluated using correlation and Bland-Altman analysis.Main results. We performed accelerated T1 mapping with ≤4 heartbeats and T2 mapping with 2 heartbeats in reference to the clinical standard, which required 11 heartbeats for T1 mapping and 3 heartbeats for T2 mapping. SwinUNet performed well in all the experiments (MAE < 50 ms, SSIM > 0.8, correlation > 0.75, and Bland-Altman agreement limits < 100 ms for T1 mapping; MAE < 1 ms, SSIM > 0.9, correlation > 0.95, and Bland-Altman agreement limits < 1.5 ms for T2 mapping). When the maximal acceleration was used (2 heartbeats), SwinUNet outperformed MyoMapNet and gave measurement accuracy similar to the clinical standard.Significance. SwinUNet offers an optimal solution to CMR parametric mapping for assessing myocardial diseases quantitatively in clinical cardiology.

Asunto(s)

RESUMEN

Kidney tumor segmentation is a difficult task because of the complex spatial and volumetric information present in medical images. Recent advances in deep convolutional neural networks (DCNNs) have improved tumor segmentation accuracy. However, the practical usability of current CNN-based networks is constrained by their high computational complexity. Additionally, these techniques often struggle to make adaptive modifications based on the structure of the tumors, which can lead to blurred edges in segmentation results. A lightweight architecture called the contextual deformable attention and edge-enhanced U-Net (CDA2E-Net) for high-accuracy pixel-level kidney tumor segmentation is proposed to address these challenges. Rather than using complex deep encoders, the approach includes a lightweight depthwise dilated ShuffleNetV2 (LDS-Net) encoder integrated into the CDA2E-Net framework. The proposed method also contains a multiscale attention feature pyramid pooling (MAF2P) module that improves the ability of multiscale features to adapt to various tumor shapes. Finally, an edge-enhanced loss function is introduced to guide the CDA2E-Net to concentrate on tumor edge information. The CDA2E-Net is evaluated on the KiTS19 and KiTS21 datasets, and the results demonstrate its superiority over existing approaches in terms of Hausdorff distance (HD), intersection over union (IoU), and dice coefficient (DSC) metrics.

RESUMEN

The goal of visual place recognition (VPR) is to determine the location of a query image by identifying its place in a collection of image databases. Visual sensor technologies are crucial for visual place recognition as they allow for precise identification and location of query images within a database. Global descriptor-based VPR methods face the challenge of accurately capturing the local specific regions within a scene; consequently, it leads to an increasing probability of confusion during localization in such scenarios. To tackle feature extraction and feature matching challenges in VPR, we propose a modified patch-NetVLAD strategy that includes two new modules: a context-aware patch descriptor and a context-aware patch matching mechanism. Firstly, we propose a context-driven patch feature descriptor to overcome the limitations of global and local descriptors in visual place recognition. This descriptor aggregates features from each patch's surrounding neighborhood. Secondly, we introduce a context-driven feature matching mechanism that utilizes cluster and saliency context-driven weighting rules to assign higher weights to patches that are less similar to densely populated or locally similar regions for improved localization performance. We further incorporate both of these modules into the patch-NetVLAD framework, resulting in a new approach called contextual patch-NetVLAD. Experimental results are provided to show that our proposed approach outperforms other state-of-the-art methods to achieve a Recall@10 score of 99.82 on Pittsburgh30k, 99.82 on FMDataset, and 97.68 on our benchmark dataset.

RESUMEN

BACKGROUND: Facial skin characteristics can provide valuable information about a patient's underlying health conditions. OBJECTIVE: In practice, there are often samples with divergent characteristics (commonly known as divergent samples) that can be attributed to environmental factors, living conditions, or genetic elements. These divergent samples significantly degrade the accuracy of diagnoses. METHODOLOGY: To tackle this problem, we propose a novel multi-feature learning method called Multi-Feature Learning with Centroid Matrix (MFLCM), which aims to mitigate the influence of divergent samples on the accurate classification of samples located on the boundary. In this approach, we introduce a novel discriminator that incorporates a centroid matrix strategy and simultaneously adapt it to a classifier in a unified model. We effectively apply the centroid matrix to the embedding feature spaces, which are transformed from the multi-feature observation space, by calculating a relaxed Hamming distance. The purpose of the centroid vectors for each category is to act as anchors, ensuring that samples from the same class are positioned close to their corresponding centroid vector while being pushed further away from the remaining centroids. RESULTS: Validation of the proposed method with clinical facial skin dataset showed that the proposed method achieved F1 scores of 92.59%, 83.35%, 82.84% and 85.46%, respectively for the detection the Healthy, Diabetes Mellitus (DM), Fatty Liver (FL) and Chronic Renal Failure (CRF). CONCLUSION: Experimental results demonstrate the superiority of the proposed method compared with typical classifiers single-view-based and state-of-the-art multi-feature approaches. To the best of our knowledge, this study represents the first to demonstrate concept of multi-feature learning using only facial skin images as an effective non-invasive approach for simultaneously identifying DM, FL and CRF in Han Chinese, the largest ethnic group in the world.

Asunto(s)

RESUMEN

The multi-layer network consists of the interactions between different layers, where each layer of the network is depicted as a graph, providing a comprehensive way to model the underlying complex systems. The layer-specific modules of multi-layer networks are critical to understanding the structure and function of the system. However, existing methods fail to characterize and balance the connectivity and specificity of layer-specific modules in networks because of the complicated inter- and intra-coupling of various layers. To address the above issues, a joint learning graph clustering algorithm (DRDF) for detecting layer-specific modules in multi-layer networks is proposed, which simultaneously learns the deep representation and discriminative features. Specifically, DRDF learns the deep representation with deep nonnegative matrix factorization, where the high-order topology of the multi-layer network is gradually and precisely characterized. Moreover, it addresses the specificity of modules with discriminative feature learning, where the intra-class compactness and inter-class separation of pseudo-labels of clusters are explored as self-supervised information, thereby providing a more accurate method to explicitly model the specificity of the multi-layer network. Finally, DRDF balances the connectivity and specificity of layer-specific modules with joint learning, where the overall objective of the graph clustering algorithm and optimization rules are derived. The experiments on ten multi-layer networks showed that DRDF not only outperforms eight baselines on graph clustering but also enhances the robustness of algorithms.

Asunto(s)

RESUMEN

Lysine glycation is one of the most significant protein post-translational modifications, which changes the properties of the proteins and causes them to be dysfunctional. Accurately identifying glycation sites helps to understand the biological function and potential mechanism of glycation in disease treatments. Nonetheless, the experimental methods are ordinarily inefficient and costly, so effective computational methods need to be developed. In this study, we proposed the new model called iGly-IDN based on the improved densely connected convolutional networks (DenseNet). First, one hot encoding was adopted to obtain the original feature maps. Afterward, the improved DenseNet was adopted to capture feature information with the importance degrees during the feature learning. According to the experimental results, Acc reaches 66%, and Mathews correlation coefficient reaches 0.33 on the independent testing data set, which indicates that the iGly-IDN can provide more effective glycation site identification than the current predictors.

Asunto(s)

RESUMEN

BACKGROUND: Structural magnetic resonance imaging (sMRI) is vital for early Alzheimer's disease (AD) diagnosis, though confirming specific biomarkers remains challenging. Our proposed Multi-Scale Self-Attention Network (MUSAN) enhances classification of cognitively normal (CN) and AD individuals, distinguishing stable (sMCI) from progressive mild cognitive impairment (pMCI). OBJECTIVE: This study leverages AD structural atrophy properties to achieve precise AD classification, combining different scales of brain region features. The ultimate goal is an interpretable algorithm for this method. METHODS: The MUSAN takes whole-brain sMRI as input, enabling automatic extraction of brain region features and modeling of correlations between different scales of brain regions, and achieves personalized disease interpretation of brain regions. Furthermore, we also employed an occlusion sensitivity algorithm to localize and visualize brain regions sensitive to disease. RESULTS: Our method is applied to ADNI-1, ADNI-2, and ADNI-3, and achieves high performance on the classification of CN from AD with accuracy (0.93), specificity (0.82), sensitivity (0.96), and area under curve (AUC) (0.95), as well as notable performance on the distinguish of sMCI from pMCI with accuracy (0.85), specificity (0.84), sensitivity (0.74), and AUC (0.86). Our sensitivity masking algorithm identified key regions in distinguishing CN from AD: hippocampus, amygdala, and vermis. Moreover, cingulum, pallidum, and inferior frontal gyrus are crucial for sMCI and pMCI discrimination. These discoveries align with existing literature, confirming the dependability of our model in AD research. CONCLUSION: Our method provides an effective AD diagnostic and conversion prediction method. The occlusion sensitivity algorithm enhances deep learning interpretability, bolstering AD research reliability.

Asunto(s)