Utilizing feedback in adaptive SAR ATR systems

Phase unwrapping is an essential process in synthetic aperture radar interferometry that restores phase signals constrained within the range of (-π, π) to their true phase values. Traditional algorithm-based methods can introduce significant errors due to rapid and steep phase gradient or noise, which negatively impact terrain elevation and surface displacement analyses. To overcome these limitations, deep learningbased phase unwrapping techniques have been proposed; however, there has been insufficient previous studies due to the lack of accurate training and test data. This paper aims to share the training data generated using the phase unwrapping simulation method with locally-different phase noise. The data were generated by the simulation of topographic phases and phase noise, atmospheric and orbital distortions. Additionally, data augmentation for phase variation and noise levels was applied to address data imbalance issues. The shared data consists of two types: one with a constant phase noise level for each patch, and another that simulates locally different phase noise based on augmented coherence data. This data is primarily effective for unwrapping topographic phase components and holds significance as the first phase unwrapping training data of synthetic aperture radar interferograms shared in Korea. We expect this resource to serve as foundational data for future phase unwrapping technology research, including applications for upcoming satellites like KOMPSAT-6 and water resource satellites.

A critical part in preoperative planning for revision arthroplasty surgery involves the identification of the failed implant. Using a predictive artificial neural network (ANN) model, the objectives of this study were: (1) to develop a machine-learning algorithm using operative big data to identify an implant from a radiograph; and (2) to compare algorithms that optimise accuracy in a timely fashion. Using 2116 postoperative anteroposterior (AP) hip radiographs of total hip arthroplasties from 2002 to 2019, 10 artificial neural networks were modeled and trained to classify the radiograph according to the femoral stem implanted. Stem brand and model was confirmed with 1594 operative reports. Model performance was determined by classification accuracy toward a random 706 AP hip radiographs, and again on a consecutive series of 324 radiographs prospectively collected over 2019. The Dense-Net 201 architecture outperformed all others with 100.00% accuracy in training data, 95.15% accuracy on validation data, and 91.16% accuracy in the unique prospective series of patients. This outperformed all other models on the validation (p < 0.0001) and novel series (p < 0.0001). The convolutional neural network also displayed the probability (confidence) of the femoral stem classification for any input radiograph. This neural network averaged a runtime of 0.96 (SD 0.02) seconds for an iPhone 6 to calculate from a given radiograph when converted to an application. Neural networks offer a useful adjunct to the surgeon in preoperative identification of the prior implant.

<title>Abstract</title> Machine-Learning Interatomic Potentials (MLIPs) have surged in popularity due to their promise of expanding the spatiotemporal scales possible for simulating molecules with high fidelity. The accuracy of any MLIP is dependent on the data used for its training; thus, for large molecules, like polymers, where accurate training data is prohibitively difficult to obtain, it becomes necessary to pursue non-traditional methods to construct MLIPs, many of which are based on constructing MLIPs using smaller, analogous chemical systems. However, we have yet to understand the limits to which smaller molecules can be used as a proxy for extrapolating macromolecular energetics. Here, we provide a ''control study'' for such experiments, exploring the ability of MLIP approaches to extrapolate between n=1-8 n-polyalkanes at identical conditions. Through Principal Covariates Classification, we quantitatively demonstrate how convergence in chemical environments between training and testing datasets coincides with an MLIP's transferability. Additionally, we show how careful attention to the construction of an MLIP's neighbor list can promote greater transferability when considering various levels of the energetic hierarchy. Our results establish a roadmap for how one can create transferable MLIPs for macromolecular systems without the prohibitive cost of constructing system-specific training data.

A feature selection method for overlapping peaks in vibrational spectroscopy using nonnegatively constrained classical least squares

Fibre tract delineation from diffusion magnetic resonance imaging (MRI) is a valuable clinical tool for neurosurgical planning and navigation, as well as in research neuroimaging pipelines. Several popular methods are used for this task, each with different strengths and weaknesses making them more or less suited to different contexts. For neurosurgical imaging, priorities include ease of use, computational efficiency, robustness to pathology and ability to generalise to new tracts of interest. Many existing methods use streamline tractography, which may require expert neuroimaging operators for setting parameters and delineating anatomical regions of interest, or suffer from as a lack of generalisability to clinical scans involving deforming tumours and other pathologies. More recently, data-driven approaches including deep-learning segmentation models and streamline clustering methods have improved reproducibility and automation, although they can require large amounts of training data and/or computationally intensive image processing at the point of application. We describe an atlas-based direct tract mapping technique called 'tractfinder', utilising tract-specific location and orientation priors. Our aim was to develop a clinically practical method avoiding streamline tractography at the point of application while utilising prior anatomical knowledge derived from only 10-20 training samples. Requiring few training samples allows emphasis to be placed on producing high quality, neuro-anatomically accurate training data, and enables rapid adaptation to new tracts of interest. Avoiding streamline tractography at the point of application reduces computational time, false positives and vulnerabilities to pathology such as tumour deformations or oedema. Carefully filtered training streamlines and track orientation distribution mapping are used to construct tract specific orientation and spatial probability atlases in standard space. Atlases are then transformed to target subject space using affine registration and compared with the subject's voxel-wise fibre orientation distribution data using a mathematical measure of distribution overlap, resulting in a map of the tract's likely spatial distribution. This work includes extensive performance evaluation and comparison with benchmark techniques, including streamline tractography and the deep-learning method TractSeg, in two publicly available healthy diffusion MRI datasets (from TractoInferno and the Human Connectome Project) in addition to a clinical dataset comprising paediatric and adult brain tumour scans. Tract segmentation results display high agreement with established techniques while requiring less than 3 min on average when applied to a new subject. Results also display higher robustness than compared methods when faced with clinical scans featuring brain tumours and resections. As well as describing and evaluating a novel proposed tract delineation technique, this work continues the discussion on the challenges surrounding the white matter segmentation task, including issues of anatomical definitions and the use of quantitative segmentation comparison metrics.

We aim to detect the class and orientation of a vehicle by training a model with synthetic data. However, the distribution of the classes in the training data is imbalanced, and the model trained on the synthetic image is difficult to predict in real-world images. We propose a two-stage detection model with photo-realistic image generation to tackle this issue. Our model mainly takes four steps to detect the class and orientation of the vehicle. (1) It builds a table containing the image, class, and location information of objects in the image, (2) transforms the synthetic images into real-world images style, and merges them into the meta table. (3) Classify vehicle class and orientation using images from the meta-table. (4) Finally, the vehicle class and orientation are detected by combining the pre-extracted location information and the predicted classes. We achieved 4 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> place in IEEE BigData Challenge 2022 Vehicle class and Orientation Detection (VOD) with our approach. Our code and project material will be available at https://github.com/inu-RAISE/VOD_Challenge

Sparse regression methods such as l1-regularized linear regression, or Lasso [18], are commonly used for analysis of high-dimensional, small-sample datasets, due to their good generalization and feature-selection properties. However, predictive accuracy of sparse regression can be further improved by incorporating more realistic data-modeling assumptions (e.g., nonlinearity) and by fully exploiting all available data, as suggested by the transductive approach [6], which makes instance-specific predictions based on both labeled (training) data and unlabeled (test) data, instead of learning a single fixed model from training data. Based on these ideas, we develop a novel method, called Transductive HSIC Lasso, that incorporates transduction into a nonlinear sparse regression approach known as HSIC Lasso [19]. Unlike the existing transductive Lasso algorithm of [1], our approach does not rely on imputation, i.e., on estimation of unknown labels using a predictor built on training data; the latter may sometimes result in poor overall performance due to unreliable label estimates. Instead, our method exploits the structure of the HSIC Lasso, which maximizes the relevance between the selected features and the label, while minimizing the redundancy between the selected features; transduction is achieved by including unlabeled samples into the redundancy computation. Our experiments demonstrate advantages of the proposed method over the state-of-the-art approaches, both on simulated and real-life data, such as prediction of phenotypic traits from genomic data, and prediction of subject’s pain level from his/her functional MRI data.

In recent years, synthetic aperture radar automatic target recognition (SAR ATR) has been widely researched for its ability to achieve high-performance target classification through supervised training, facilitating battlefield reconnaissance, and intelligence generation. When dealing with unknown class data for ATR model training, significant time and effort are typically required for manual interpretation and labeling. However, when unknown class data shares the same domain as the known labeled data in the library, leveraging their shared deep semantic knowledge can enable automatic classification and labeling of the unknown class data. In this article, we investigate the novel category discovery (NCD) problem in SAR images, using labeled data to guide the clustering process of new class data. Specifically, we utilize the Unified Objective function training framework to address the training imbalance between labeled and unlabeled data in the NCD process, incorporating various improvements on this foundation. Through a binary segmentation-based strategy, we effectively mitigate the interference of “noise pairs” with significant semantic differences on the model's self-supervised pretraining. In addition, we introduce multicrop consistency loss and equal distance loss to impose constraints on training by leveraging intraclass and interclass relationships in the latent space, thereby obtaining representations with higher interclass separability. Our method achieves state-of-the-art clustering performance in multiple scenarios. Extensive experimental results on the MSTAR benchmark dataset demonstrate the effectiveness of the proposed methods.

Supervised and unsupervised classification procedures are developed and applied to synthetic aperture radar (SAR) polarimetric images in order to identify their various Earth terrain components. For supervised classification processing, the Bayes technique is used to classify fully polarimetric and normalized polarimetric SAR data. Simpler polarimetric discriminates, such as the absolute and normalized magnitude response of the individual receiver channel returns, in addition to the phase difference between the receiver channels, are also considered. Another processing algorithm, based on comparing general properties of the Stokes parameters of the scattered wave to that of simple scattering models, is also discussed. This algorithm, which is an unsupervised technique, classifies terrain elements based on the relationship between the orientation angle and handedness of the transmitting and receiving polarization states. These classification procedures have been applied to San Francisco Bay and Traverse City SAR images, supplied by the Jet Propulsion Laboratory. It is shown that supervised classification yields the best overall performance when accurate classifier training data are used, whereas unsupervised classification is applicable when training data are not available.

Summary In this research, the application of adaptive neuro-fuzzy inference system (ANFIS) was investigated in determining the optimum exploratory boreholes points, using exploratory data. The study area is Vertaveh iron deposit that is located in the south of Kashan and 8 kilometers away from south-east of Ghamsar city. The input layers to adaptive neuro-fuzzy inference system consisted geomagnetic data and iron grade that obtained from powder drilling (RC). After the preparation of input layers, 30 percent of data selected as test data and remaining 70 percent as training data. The training data was done for three function of adaptive neuro-fuzzy inference system, then these three functions were compared by two error indices and the best function with minimum error indices was selected. Finally, the best regions were identified in terms of grade and thickness and also the optimum exploratory boreholes points were determined.

Improving autocoding performance of rare categories in injury classification: Is more training data or filtering the solution?

Vehicle-activated signs (VAS) are speed-warning signs activated by radar when the driver speed exceeds a pre-set threshold, i.e. the trigger speed. The trigger speed is often set relative to the speed limit and is displayed for all types of vehicles. It is our opinion that having a static setting for the trigger speed may be inappropriate, given that traffic and road conditions are dynamic in nature. Further, different vehicle classes (mainly cars and trucks) behave differently, so a uniform trigger speed of such signs may be inappropriate to warn different types of vehicles. The current study aims to investigate an automatic VAS, i.e. one that could warn vehicle users with an appropriate trigger speed by taking into account vehicle types and road conditions. We therefore investigated different vehicle classes, their speeds, and the time of day to be able to conclude whether different trigger speeds of VAS are essential or not. The current study is entirely data driven; data are initially presented to a self-organising map (SOM) to be able to partition the data into different clusters, i.e. vehicle classes. Speed, time of day, and length of vehicle were supplied as inputs to the SOM. Further, the 85th percentile speed for the next hour is predicted using appropriate prediction models. Adaptive neuro-fuzzy inference systems and random forest (RF) were chosen for speed prediction; the mean speed, traffic flow, and standard deviation of vehicle speeds were supplied as inputs for the prediction models. The results achieved in this work show that RF is a reliable model in terms of accuracy and efficiency, and can be used in finding appropriate trigger speeds for an automatic VAS.

Wind is a natural phenomenon that plays an essential role in various aspects of human life, including the spread of pests in plants. This variable is right for regions often hit by strong winds. The development of machine learning technology now makes predictions of wind speed to anticipate future impacts. This study proposes wind speed predictions using Recurrent Neural Network (RNN) with Long Short Term Memory (LSTM). The data used was obtained from the Nganjuk Meteorology and Geophysics Agency (BMKG), East Java from 2008 to 2017. The results showed that the use of the Adam model could provide 92.7% accuracy for training data and 91.6% for new data.

COVID-19 (Coronavirus Disease 2019) is a severe global pandemic infection induced by a new coronavirus also known as SARS Coronavirus (SARS-CoV). It is a dynamic infection that affects different people in different ways; thus, a delay in the precise detection and treatment leads to the spread of the infection leading to more deaths. Automatic intelligent diagnostic models are thus needed to help detect such severe diseases at an early stage, thereby saving lives. For this, researchers developed a machine learning (ML) methodology for screening and COVID-19 classification using SARS-CoV2 CT scan images along with in-depth analysis of various features that is fascinating due to their better performance than conventional CT scan image classification for COVID-19. In addition, CT scan image feature extraction was measured using in-depth analysis with global and local feature descriptors. Principal component analysis (PCA) was employed to determine the optimal subset for COVID-19 recognition and classification. We compared and checked the best-in-class ML strategies on SARS-CoV2 CT scan images. The accuracy of training data was 98.78% and the accuracy of test data was 93.96% for a proposed model tested on the SARS-CoV2 CT scan images dataset. The findings of the experiment indicate a major improvement in COVID-19 classification, which aids in better COVID-19 investigation and disease prevention.

A major bottleneck for building statistical spoken dialogue systems for new domains and applications is the need for large amounts of training data. To address this problem, we adopt the multi-dimensional approach to dialogue management and evaluate its potential for transfer learning. Specifically, we exploit pre-trained task-independent policies to speed up training for an extended task-specific action set, in which the single summary action for requesting a slot is replaced by multiple slot-specific request actions. Policy optimisation and evaluation experiments using an agenda-based user simulator show that with limited training data, much better performance levels can be achieved when using the proposed multi-dimensional adaptation method. We confirm this improvement in a crowd-sourced human user evaluation of our spoken dialogue system, comparing partially trained policies. The multi-dimensional system (with adaptation on limited training data in the target scenario) outperforms the one-dimensional baseline (without adaptation on the same amount of training data) by 7% perceived success rate.