3D Convolutional Neural Network Research Articles

A non-enhanced CT-based deep learning diagnostic system for COVID-19 infection at high risk among lung cancer patients.

Pneumonia and lung cancer have a mutually reinforcing relationship. Lung cancer patients are prone to contracting COVID-19, with poorer prognoses. Additionally, COVID-19 infection can impact anticancer treatments for lung cancer patients. Developing an early diagnostic system for COVID-19 pneumonia can help improve the prognosis of lung cancer patients with COVID-19 infection. This study proposes a neural network for COVID-19 diagnosis based on non-enhanced CT scans, consisting of two 3D convolutional neural networks (CNN) connected in series to form two diagnostic modules. The first diagnostic module classifies COVID-19 pneumonia patients from other pneumonia patients, while the second diagnostic module distinguishes severe COVID-19 patients from ordinary COVID-19 patients. We also analyzed the correlation between the deep learning features of the two diagnostic modules and various laboratory parameters, including KL-6. The first diagnostic module achieved an accuracy of 0.9669 on the training set and 0.8884 on the test set, while the second diagnostic module achieved an accuracy of 0.9722 on the training set and 0.9184 on the test set. Strong correlation was observed between the deep learning parameters of the second diagnostic module and KL-6. Our neural network can differentiate between COVID-19 pneumonia and other pneumonias on CT images, while also distinguishing between ordinary COVID-19 patients and those with white lung. Patients with white lung in COVID-19 have greater alveolar damage compared to ordinary COVID-19 patients, and our deep learning features can serve as an imaging biomarker.

Detection and Classification of Network Traffic in Bot Network Using Deep Learning

One of the most dangerous threats to computer networks is the use of botnets, which can seriously harm systems and steal private data. They are remote-controlled networks of compromised computers that an individual or group of individuals is using for malicious purposes. These infected computers are frequently called “bots” or “zombies”. A wide variety of malicious activities, including the distribution of malware and credential theft, can be carried out using botnets. The CTU-13 dataset is a collection of network traffic information that includes examples of various botnet types. Using this, our study compares the abilities of decision trees, random forests, 1D convolutional neural networks, and a proposed system based on long short-term memory and residual neural networks to detect botnets. According to our findings, the suggested system performs better than every other algorithm, achieving a higher accuracy rate. Our suggested system has the ability to precisely identify botnet traffic patterns, which can assist organisations in proactively preventing botnet attacks.

Image Reconstruction in Ultrasonic Speed-of-Sound Computed Tomography Using Time of Flight Estimated by a 2D Convolutional Neural Networks

In ultrasonic nondestructive testing (NDT), accurately estimating the time of flight (TOF) of ultrasonic waves is crucial. Traditionally, TOF estimation involves the signal processing of a single measured waveform. In recent years, deep learning has also been applied to estimate the TOF; however, these methods typically process only single waveforms. In contrast, this study acquired fan-beam ultrasonic waveform profile data from 64 paths using an ultrasonic-speed computed tomography (CT) simulation of a circular column and developed a TOF estimation model using two-dimensional convolutional neural networks (CNNs) based on these data. We compared the accuracy of the TOF estimation between the proposed method and two traditional signal processing methods. Additionally, we reconstructed ultrasonic-speed CT images using the estimated TOF and evaluated the generated CT images. The results showed that the proposed method could estimate the longitudinal TOF more accurately than traditional methods, and the evaluation scores for the reconstructed images were high.

Virtual Restoration of Ancient Mold-Damaged Painting Based on 3D Convolutional Neural Network for Hyperspectral Image

Painted cultural relics hold significant historical value and are crucial in transmitting human culture. However, mold is a common issue for paper or silk-based relics, which not only affects their preservation and longevity but also conceals the texture, patterns, and color information, hindering cultural value and heritage. Currently, the virtual restoration of painting relics primarily involves filling in the RGB based on neighborhood information, which might cause color distortion and other problems. Another approach considers mold as noise and employs maximum noise separation for its removal; however, eliminating the mold components and implementing the inverse transformation often leads to more loss of information. To effectively acquire virtual restoration for mold removal from ancient paintings, the spectral characteristics of mold were analyzed. Based on the spectral features of mold and the cultural relic restoration philosophy of maintaining originality, a 3D CNN artifact restoration network was proposed. This network is capable of learning features in the near-infrared spectrum (NIR) and spatial dimensions to reconstruct the reflectance of visible spectrum, achieving the virtual restoration for mold removal of calligraphic and art relics. Using an ancient painting from the Qing Dynasty as a test subject, the proposed method was compared with the Inpainting, Criminisi, and inverse MNF transformation methods across three regions. Visual analysis, quantitative evaluation (the root mean squared error (RMSE), mean absolute percentage error (MAPE), mean absolute error (MEA), and a classification application were used to assess the restoration accuracy. The visual results and quantitative analyses demonstrated that the proposed 3D CNN method effectively removes or mitigates mold while restoring the artwork to its authentic color in various backgrounds. Furthermore, the color classification results indicated that the images restored with 3D CNN had the highest classification accuracy, with overall accuracies of 89.51%, 92.24%, and 93.63%, and Kappa coefficients of 0.88, 0.91, and 0.93, respectively. This research provides technological support for the digitalization and restoration of cultural artifacts, thereby contributing to the preservation and transmission of cultural heritage.

Machine learning classification of permeable conducting spheres in air and seawater using electromagnetic pulses

Abstract This paper presents machine learning classification on simulated data of permeable conducting spheres in air and seawater irradiated by low frequency electromagnetic pulses. Classification accuracy greater than 90% was achieved. The simulated data were generated using an analytical model of a magnetic dipole in air and seawater placed 1.5–3.5 m above the center of the sphere in 50 cm increments. The spheres had radii of 40 cm and 50 cm and were of permeable materials, such as steel, and non-permeable materials, such as aluminum. A series RL circuit was analytically modeled as the transmitter coil, and an RLC circuit as the receiver coil. Additive white Gaussian noise was added to the simulated data to test the robustness of the machine learning algorithms to noise. Multiple machine learning algorithms were used for classification including a perceptron and multiclass logistic regression, which are linear models, and a neural network, 1D convolutional neural network (CNN), and 2D CNN, which are nonlinear models. Feature maps are plotted for the CNNs and provide explainability of the salient parts of the time signature and spectrogram data used for classification. The pulses investigated, which expand the literature, include a two-sided decaying exponential, Heaviside step-off, triangular, Gaussian, rectangular, modulated Gaussian, raised cosine, and rectangular down-chirp. Propagation effects, including dispersion and frequency dependent attenuation, are encapsulated by the analytical model, which was verified using finite element modeling. The results in this paper show that machine learning methods are a viable alternative to inversion of electromagnetic induction (EMI) data for metallic sphere classification, with the advantage of real-time classification without the use of a physics-based model. The nonlinear machine learning algorithms used in this work were able to accurately classify metallic spheres in seawater even with significant pulse distortion caused by dispersion and frequency dependent attenuation. This paper presents the first effort towards the use of machine learning to classify metallic objects in seawater based on EMI sensing.

A Novel Multi-modal Sentiment Analysis Based on Multiple Kernel Learning with Margin-Dimension Constraint

Most of the sentiment analysis studies focus on the sentimental classification of pictures in the video, ignoring the spatio-temporal information of the sequence of picture frames as well as text and audio information. The multiple kernel learning is a new hotspot in the field of nuclear machine learning, capable of handling multiple modalities. For multiple kernel learning, it is easy to ignore the basic features that are not discriminative, and cannot make full use of the base features of different modes. This paper puts forward a novel multi-modal fusion model for sentiment analysis, in which a multiple kernel learning algorithm based on convolution margin-dimension constraint is proposed for feature fusion. Moreover, the 3D convolutional neural network is used to extract the features of visual information, and the multiple kernel learning algorithm based on margin-dimension constraint is used to fuse visual, text and audio sentiment features. Experiments conducted on the MOUD and IEMOCAP sentiment databases show that the proposed model outperforms existing models in the field of multi-modal sentiment analysis research.

Crystal structure identification with 3D convolutional neural networks with application to high-pressure phase transitions in SiO2

Efficient, reliable and easy-to-use structure recognition of atomic environments is essential for the analysis of atomic scale computer simulations. In this work, we train two neuronal network (NN) architectures, namely PointNet and dynamic graph convolutional NN (DG-CNN) using different hyperparameters and training regimes to assess their performance in structure identification tasks of atomistic structure data. We show benchmarks on simple crystal structures, where we can compare against established methods. The approach is subsequently extended to structurally more complex SiO2 phases. By making use of this structure recognition tool, we are able to achieve a deeper understanding of the crystallization process in amorphous SiO2 under shock compression. Lastly, we show how the NN based structure identification workflows can be integrated into OVITO using its python interface.

Fully Automated Hippocampus Segmentation using T2-informed Deep Convolutional Neural Networks

Hippocampal atrophy (tissue loss) has become a fundamental outcome parameter in clinical trials on Alzheimer's disease. To accurately estimate hippocampus volume and track its volume loss, a robust and reliable segmentation is essential. Manual hippocampus segmentation is considered the gold standard but is extensive, time-consuming, and prone to rater bias. Therefore, it is often replaced by automated programs like FreeSurfer, one of the most commonly used tools in clinical research. Recently, deep learning-based methods have also been successfully applied to hippocampus segmentation. The basis of all approaches are clinically used T1-weighted whole-brain MR images with approximately 1 mm isotropic resolution. However, such T1 images show low contrast-to-noise ratios (CNRs), particularly for many hippocampal substructures, limiting delineation reliability. To overcome these limitations, high-resolution T2-weighted scans are suggested for better visualization and delineation, as they show higher CNRs and usually allow for higher resolutions. Unfortunately, such time-consuming T2-weighted sequences are not feasible in a clinical routine.We propose an automated hippocampus segmentation pipeline leveraging deep learning with T2-weighted MR images for enhanced hippocampus segmentation of clinical T1-weighted images based on a series of 3D convolutional neural networks and a specifically acquired multi-contrast dataset. This dataset consists of corresponding pairs of T1- and high-resolution T2-weighted images, with the T2 images only used to create more accurate manual ground truth annotations and to train the segmentation network. The T2-based ground truth labels were also used to evaluate all experiments by comparing the masks visually and by various quantitative measures.We compared our approach with four established state-of-the-art hippocampus segmentation algorithms (FreeSurfer, ASHS, HippoDeep, HippMapp3r) and demonstrated a superior segmentation performance. Moreover, we found that the automated segmentation of T1-weighted images benefits from the T2-based ground truth data.In conclusion, this work showed the beneficial use of high-resolution, T2-based ground truth data for training an automated, deep learning-based hippocampus segmentation and provides the basis for a reliable estimation of hippocampal atrophy in clinical studies.

Prediction of epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients on computed tomography (CT) images using 3-dimensional (3D) convolutional neural network.

Noninvasively detecting epidermal growth factor receptor (EGFR) mutation status in lung adenocarcinoma patients before targeted therapy remains a challenge. This study aimed to develop a 3-dimensional (3D) convolutional neural network (CNN)-based deep learning model to predict EGFR mutation status using computed tomography (CT) images. We retrospectively collected 660 patients from 2 large medical centers. The patients were divided into training (n=528) and external test (n=132) sets according to hospital source. The CNN model was trained in a supervised end-to-end manner, and its performance was evaluated using an external test set. To compare the performance of the CNN model, we constructed 1 clinical and 3 radiomics models. Furthermore, we constructed a comprehensive model combining the highest-performing radiomics and CNN models. The receiver operating characteristic (ROC) curves were used as primary measures of performance for each model. Delong test was used to compare performance differences between different models. Compared with the clinical [training set, area under the curve (AUC) =69.6%, 95% confidence interval (CI), 0.661-0.732; test set, AUC =68.4%, 95% CI, 0.609-0.752] and the highest-performing radiomics models (training set, AUC =84.3%, 95% CI, 0.812-0.873; test set, AUC =72.4%, 95% CI, 0.653-0.794) models, the CNN model (training set, AUC =94.3%, 95% CI, 0.920-0.961; test set, AUC =94.7%, 95% CI, 0.894-0.978) had significantly better predictive performance for predicting EGFR mutation status. In addition, compared with the comprehensive model (training set, AUC =95.7%, 95% CI, 0.942-0.971; test set, AUC =87.4%, 95% CI, 0.820-0.924), the CNN model had better stability. The CNN model has excellent performance in non-invasively predicting EGFR mutation status in patients with lung adenocarcinoma and is expected to become an auxiliary tool for clinicians.

Decoding micro-electrocorticographic signals by using explainable 3D convolutional neural network to predict finger movements

BackgroundElectroencephalography (EEG) and electrocorticography (ECoG) recordings have been used to decode finger movements by analyzing brain activity. Traditional methods focused on single bandpass power changes for movement decoding, utilizing machine learning models requiring manual feature extraction. New methodThis study introduces a 3D convolutional neural network (3D-CNN) model to decode finger movements using ECoG data. The model employs adaptive, explainable AI (xAI) techniques to interpret the physiological relevance of brain signals. ECoG signals from epilepsy patients during awake craniotomy were processed to extract power spectral density across multiple frequency bands. These data formed a 3D matrix used to train the 3D-CNN to predict finger trajectories. ResultsThe 3D-CNN model showed significant accuracy in predicting finger movements, with root-mean-square error (RMSE) values of 0.26–0.38 for single finger movements and 0.20–0.24 for combined movements. Explainable AI techniques, Grad-CAM and SHAP, identified the high gamma (HG) band as crucial for movement prediction, showing specific cortical regions involved in different finger movements. These findings highlighted the physiological significance of the HG band in motor control. Comparison with existing methodsThe 3D-CNN model outperformed traditional machine learning approaches by effectively capturing spatial and temporal patterns in ECoG data. The use of xAI techniques provided clearer insights into the model's decision-making process, unlike the "black box" nature of standard deep learning models. ConclusionsThe proposed 3D-CNN model, combined with xAI methods, enhances the decoding accuracy of finger movements from ECoG data. This approach offers a more efficient and interpretable solution for brain-computer interface (BCI) applications, emphasizing the HG band's role in motor control.

Structural integrity of aging steel bridges by 3D laser scanning and convolutional neural networks

For steel bridges, corrosion has historically led to bridge failures, resulting in fatalities and injuries. To enhance public safety and prevent such incidents, authorities mandate in-situ evaluation and reporting of corroded members. The current inspection and evaluation protocol is characterized by intense labor, traffic delays, and poor capacity predictions. Here we combine full-scale experimental testing of a decommissioned girder, 3D laser scanning, and convolutional neural networks (CNNs) to introduce a continuous inspection and evaluation framework. Classification and regression CNNs are trained on a databank of 1,421 naturally inspired corrosion scenarios, generated computationally based on point clouds of three corroded girders collected in lab conditions. Results indicate low errors of up to 2.0% and 3.3%, respectively. The methodology is validated on eight real corroded ends and implemented for the evaluation of an in-service bridge. This framework promises significant advancements in assessing aging bridge infrastructure with higher accuracy and efficiency compared to analytical or semi-analytical approaches.

Deep learning-based prediction of velocity and temperature distributions in metal foam with hierarchical pore structure

Constrained by the substantial computational time required for numerical simulation, a deep learning technique is applied to investigate fluid flow and heat transfer processes in metal foam with a hierarchical pore structure. This work adopted a 3D convolutional neural network (CNN) combining U-Net architecture to predict velocity and temperature distributions, alongside corresponding permeability and overall heat transfer coefficient. This approach demonstrates excellent capability in intricate image segmentation. The training sets were acquired by lattice Boltzmann method (LBM) simulations. The CNN model, trained on a substantial amount of data, demonstrates remarkable precision, exhibiting mean relative errors of 0.57% for permeability prediction and 2.27% for overall heat transfer coefficient prediction. Moreover, in CNN prediction, a broader range of structure parameters and boundary conditions beyond those in the training set was used to evaluate the practicability of the trained CNN model. In contrast to numerical simulation, the CNN model economizes approximately 95.41% and 99.57% of computational time for velocity and temperature distribution prediction, respectively, providing a novel approach for exploring transport processes in metal foam with hierarchical pore structure.

Deep learning for ultrasound surface echo detection

Ultrasound array imaging techniques are often applied using a coupling medium to transmit waves from the transducer to the scanned component. In this type of tests, the echoes generated by the wave reflection at the component surface contain information that can be used to estimate the surface shape and its position relative to the probe, which are needed to compute the imaging focal laws. The most interesting feature to extract from surfaces echoes is the Time of Flight (TOF) for each reception channel. TOF measurement in acoustic signals is a problem that frequently arises not only in Non Destructive Testing (NDT), but also in medical imaging, geophysics and other fields. The surface echo in a signal is usually identified as the first high amplitude peak, and is traditionally detected with standard methods as threshold crossing, peak search or matched filters. All these methods are easily affected by Signal to Noise Ratio, which might be high in some channels, for example if the ray arrives at the corresponding element at a low sensibility angle. Moreover, spurious echoes from bubbles, the emission signal tail or echoes from other parts of the component can be incorrectly labelled as surface echoes. This kind of outliers are difficult to filter by conventional means and, thus, there is a need for more robust surface estimation methods. In this work, we study the application of a deep 3D Convolutional Neural Network (CNN) to detect surface echoes in Full Matrix Capture (FMC) ultrasound data. The CNN is trained using signals acquired with a matrix array and a reference component, using a robotic arm set-up to measure the probe position and orientation relative to the component. Thus, with knowledge of the set-up geometry, a model is used to compute the theoretical surface echo TOFs, which are used to label the data for training. The model was tested with a validation set, presenting good agreement with the ground truth. The results were compared with the threshold crossing method observing a significant reduction of outliers in the TOF estimation.

Improved microvascular imaging with optical coherence tomography using 3D neural networks and a channel attention mechanism

Skin microvasculature is vital for human cardiovascular health and thermoregulation, but its imaging and analysis presents significant challenges. Statistical methods such as speckle decorrelation in optical coherence tomography angiography (OCTA) often require multiple co-located B-scans, leading to lengthy acquisitions prone to motion artefacts. Deep learning has shown promise in enhancing accuracy and reducing measurement time by leveraging local information. However, both statistical and deep learning methods typically focus solely on processing individual 2D B-scans, neglecting contextual information from neighbouring B-scans. This limitation compromises spatial context and disregards the 3D features within tissue, potentially affecting OCTA image accuracy. In this study, we propose a novel approach utilising 3D convolutional neural networks (CNNs) to address this limitation. By considering the 3D spatial context, these 3D CNNs mitigate information loss, preserving fine details and boundaries in OCTA images. Our method reduces the required number of B-scans while enhancing accuracy, thereby increasing clinical applicability. This advancement holds promise for improving clinical practices and understanding skin microvascular dynamics crucial for cardiovascular health and thermoregulation.

Design of a model for multistage classification of diabetic retinopathy and glaucoma

This study addresses the escalating prevalence of diabetic retinopathy (DR) and glaucoma, major global causes of vision impairment. We propose an innovative iterative Q-learning model that integrates with fuzzy C-means clustering to improve diagnostic accuracy and classification speed. Traditional diagnostic frameworks often struggle with accuracy and delay in disease stage classification, particularly in discerning complex features like exudates and veins. Our model overcomes these challenges by combining fuzzy C means with Q learning, enhancing precision in identifying key retinal components. The core of our approach is a custom-designed 45-layer 2D convolutional neural network (CNN) optimized for nuanced detection of DR and glaucoma stages. Compared to previous approaches, the performance on the IDRID and SMDG-19 datasets and associated samples shows a 10.9% rise in precision, an 8.5% improvement in overall accuracy, an 8.3% enhancement in recall, a 10.4% larger area under the curve (AUC), a 5.9% boost in specificity, and a 2.9% decrease in latency. This methodology has the potential to bring about significant changes in the field of DR and glaucoma diagnosis, leading to prompt medical interventions and possibly decreasing vision loss. The use of sophisticated machine learning techniques in medical imaging establishes a model for future investigations in ophthalmology and other clinical situations.

Feature attention graph neural network for estimating brain age and identifying important neural connections in mouse models of genetic risk for Alzheimer’s disease

Abstract Alzheimer’s disease (AD), a widely studied neurodegenerative disorder, poses significant research challenges due to its high prevalence and complex etiology. Age, a critical risk factor for AD, is typically assessed by comparing physiological and estimated brain ages. This study utilizes mouse models expressing human alleles of APOE and human nitric oxide synthase 2 (hNOS2), replicating genetic risks for AD alongside a human-like immune response. We developed a multivariate model that incorporates brain structural connectomes, APOE genotypes, demographic traits (age and sex), environmental factors such as diet, and behavioral data to estimate brain age. Our methodology employs a Feature Attention Graph Neural Network (FAGNN) to integrate these diverse datasets. Behavioral data are processed using a 2D convolutional neural network (CNN), demographic traits via a 1D CNN, and brain connectomes through a graph neural network equipped with a quadrant attention module that accentuates critical neural connections. The FAGNN model demonstrated a mean absolute error in age prediction of 31.85 days and a root mean squared error of 41.84 days, significantly outperforming simpler models. Our analysis further focused on the brain age delta, which assesses accelerated or delayed aging by comparing brain age, predicted by FAGNN, to the chronological age. A high-fat diet and the presence of the human NOS2 gene were identified as significant accelerators of brain aging in the old age group. Key neural connections identified by FAGNN, such as those between the cingulum, corpus callosum, striatum, hippocampus, thalamus, hypothalamus, cerebellum, and piriform cortex, were found to be significant in the aging process. Validation using diffusion MRI-based metrics, including fractional anisotropy and return-to-origin probability measures across these connections, revealed significant age-related differences. These findings suggest that white matter degradation in the connections highlighted by FAGNN plays a key role in aging. Our findings suggest that the complex interplay of APOE genotype with sex, immunity, and environmental factors modulates brain aging and enhance our understanding of AD risk in mouse models of aging.

Innovative healthcare solutions: robust hand gesture recognition of daily life routines using 1D CNN.

Hand gestures are an effective communication tool that may convey a wealth of information in a variety of sectors, including medical and education. E-learning has grown significantly in the last several years and is now an essential resource for many businesses. Still, there has not been much research conducted on the use of hand gestures in e-learning. Similar to this, gestures are frequently used by medical professionals to help with diagnosis and treatment. We aim to improve the way instructors, students, and medical professionals receive information by introducing a dynamic method for hand gesture monitoring and recognition. Six modules make up our approach: video-to-frame conversion, preprocessing for quality enhancement, hand skeleton mapping with single shot multibox detector (SSMD) tracking, hand detection using background modeling and convolutional neural network (CNN) bounding box technique, feature extraction using point-based and full-hand coverage techniques, and optimization using a population-based incremental learning algorithm. Next, a 1D CNN classifier is used to identify hand motions. After a lot of trial and error, we were able to obtain a hand tracking accuracy of 83.71% and 85.71% over the Indian Sign Language and WLASL datasets, respectively. Our findings show how well our method works to recognize hand motions. Teachers, students, and medical professionals can all efficiently transmit and comprehend information by utilizing our suggested system. The obtained accuracy rates highlight how our method might improve communication and make information exchange easier in various domains.

Application of Photoplethysmography Combined with Deep Learning in Postoperative Monitoring of Flaps

Photoelectric volumetric tracing (PPG) exhibits high sensitivity and specificity in flap monitoring. Deep learning (DL) is capable of automatically and robustly extracting features from raw data. In this study, we propose combining PPG with 1D convolutional neural networks (1D-CNN) to preliminarily explore the method's ability to distinguish the degree of embolism and to localize the embolic site in skin flap arteries. Data were collected under normal conditions and various embolic scenarios by creating vascular emboli in a dermatome artery model and a rabbit dermatome model. These datasets were then trained, validated, and tested using 1D-CNN. As the degree of arterial embolization increased, the PPG amplitude upstream of the embolization site progressively increased, while the downstream amplitude progressively decreased, and the gap between the upstream and downstream amplitudes at the embolization site progressively widened. 1D-CNN was evaluated in the skin flap arterial model and rabbit skin flap model, achieving average accuracies of 98.36% and 95.90%, respectively. The combined monitoring approach of DL and PPG can effectively identify the degree of embolism and locate the embolic site within the skin flap artery.

AutoCorNN: An Unsupervised Physics-Aware Deep Learning Model for Geometric Distortion Correction of Brain MRI Images Towards MR-Only Stereotactic Radiosurgery.

Geometric distortions in brain MRI images arising from susceptibility artifacts at air-tissue interfaces pose a significant challenge for high-precision radiation therapy modalities like stereotactic radiosurgery, necessitating sub-millimeter accuracy. To achieve this goal, we developed AutoCorNN, an unsupervised physics-aware deep-learning model for correcting geometric distortions. Two publicly available datasets, the MPI-Leipzig Mind-Brain-Body with 318 subjects, and the Vestibular Schwannoma-SEG dataset, encompassing 242 patients were utilized. AutoCorNN integrates two 2D convolutional encoder-decoder neural networks with the forward physical model of MRI signal generation to predict undistorted MR and field map images from distorted MR input. The network is trained in an unsupervised manner by minimizing the mean absolute error between the measured and estimated k-space data, without requiring ground truth images during training or deployment. The model was evaluated on vestibular schwannoma cases. AutoCorNN achieved a peak signal-to-noise ratio (PSNR) of 41.35 ± 0.02dB, a root mean square error (RMSE) of 0.02 ± 0.003, and a structural similarity index (SSIM) of 0.99 ± 0.02 outperforming uncorrected and B0-mapping correction methods. Geometric distortions of about 1.6mm were observed at the air-tissue interfaces at the air canal and nasal cavity borders. Geometrically, distortion correction increased the target volume from 3.12 ± 0.52cc to 3.84 ± 0.54cc. Dosimetrically, AutoCorNN improved target coverage (0.96 ± 0.01 to 0.97 ± 0.02), conformity index (0.92 ± 0.03 to 0.94 ± 0.03), and reduced dose gradients outside the target. AutoCorNN achieves accurate geometric distortion correction comparable to conventional iterative methods while offering substantial computational acceleration, enabling precise target delineation and conformal dose delivery for improved radiation therapy outcomes.

Sequence Deep Learning for Seismic Ground Response Modeling: 1D-CNN, LSTM, and Transformer Approach

Accurate seismic ground response analysis is crucial for the design and safety of civil infrastructure and establishing effective mitigation measures against seismic risks and hazards. This is a complex process due to the nonlinear soil properties and complicated underground geometries. As a simplified approach, the one-dimensional wave propagation model, which assumes that seismic waves travel vertically through a horizontally layered medium, is widely adopted for its reasonable performance in many practical applications. This study explores the potential of sequence deep learning models, specifically 1D convolutional neural networks (1D-CNNs), long short-term memory (LSTM) networks, and transformers, as an alternative for seismic ground response modeling. Utilizing ground motion data from the Kiban Kyoshin Network (KiK-net), we train these models to predict ground surface acceleration response spectra based on bedrock motions. The performance of the data-driven models is compared with the conventional equivalent-linear analysis model, SHAKE2000. The results demonstrate that the deep learning models outperform the physics-based model across various sites, with the transformer model exhibiting the smallest average prediction error due to its ability to capture long-range dependencies. The 1D-CNN model also shows a promising performance, albeit with occasional higher errors than the other models. All the data-driven models exhibit efficient computation times of less than 0.4 s for estimation. These findings highlight the potential of sequence deep learning approaches for seismic ground response modeling.