Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,341 papers · 148 categories

Trend · papers per month

1234 · Feb 202019922001200920182026
48 results for supermarket scanner

Improved inter-scanner MS lesion segmentation through adversarial training.

problem Variability in MRI scanner or protocol differences affect automated lesion segmentation accuracy.
method Trained a CNN base model and a discriminator model adversarially on multi-scanner longitudinal data.
result Adversarial training improves inter-scanner consistency of lesion segmentations.

New CNN method improves brain MR segmentation across scanners and protocols.

problem Degradation of CNN accuracy on images from different scanners and protocols.
method Lifelong multi-domain learning with shared filters and domain-specific batch normalization.
result Significantly closes the gap to benchmark performance.

Improved image segmentation across scanners using asymmetric weights.

problem Deteriorated performance of classifiers trained on one scanner for segmentation of images from other scanners.
method A weighted ensemble of classifiers trained on individual images, with weights determined by the similarity between training and test images.
result The bag similarity measure is the most robust and achieves excellent results on various brain and white matter lesion segmentation datasets.

Sales data in a commodity market (supermarket sales to consumers) has been analysed by studying the fluctuation spectrum and noise correlations. Three related products (ketchup, mayonnaise and curry sauce) have been analysed. Most noise in sales is caused by promotions, but here we focus on the fluctuations in baseline…

2004-12-07abs ↗pdf ↗

Adapts CNN for robust medical image segmentation across different scanners and protocols.

problem Performance degradation of CNNs in medical image segmentation due to mismatch between training and test images.
method Designs a segmentation CNN as a concatenation of a shallow normalization CNN and a deep CNN. At test time, adapts the normalization sub-network for each test image using a denoising autoencoder.
result Consistently improves performance on multi-center MRI datasets of brain, heart, and prostate.

MimickNet approximates clinical ultrasound post-processing without proprietary data.

problem Matching proprietary clinical-grade ultrasound post-processing techniques.
method Deep learning framework MimickNet that transforms raw DAS beams into post-processed images.
result MimickNet achieves high SSIM scores (0.930-0.967) on test sets.

Dynamic memory prevents forgetting in continuous learning of medical images.

problem Catastrophic forgetting in machine learning models over time due to domain shifts.
method Dynamic memory to store and replay diverse training data subsets.
result Dynamic memory mitigates forgetting without knowing when shifts occur.

A neural network learns MRI scan-invariant features for brain tissue classification.

problem Lack of generalization in voxelwise classification methods due to scanner differences.
method Siamese neural network (MRAI-net) to learn acquisition-invariant representations.
result Linear classifier outperforms CNNs on limited training data for tissue classification.

Study investigates deep learning model's reliability in clinical MRI data.

problem Tackles reliability of DL models in clinical out-of-distribution MRI data.
method Investigated performance of DL model trained on diverse datasets compared to clinical data.
result Model performs better in similar protocols but worse in clinical data with different tissue contrasts.

Novel method uses image descriptors to harmonize MRI brain volumes across centers.

problem Inconsistencies in MRI brain volume measurements across different centers and scanners.
method Trained a Relevance Vector Machine (RVM) model using image descriptors to harmonize brain volumes.
result Decreases scanner and center variability while preserving measurements for longitudinal studies.

Paper introduces MSP Network for harmonizing images from different scanners.

problem Improving predictive performance and learning efficiency in data harmonization.
method Multi-Stage Prediction (MSP) Network integrating neural networks of different architectures.
result MSP Network shows 20% improvement in patch-based mean-squared error over state-of-the-art methods.

DALES offers a large annotated aerial LiDAR dataset for 3D deep learning.

problem Lack of large-scale annotated aerial LiDAR datasets for deep learning.
method Collection and annotation of over half a billion hand-labeled points from an ALS scanner.
result DALES is the most extensive publicly available ALS data set with improved resolution and coverage.

Study geodesic X-ray transform and streaking artifacts on simple surfaces or spaces of constant curvature.

problem Streaking artifacts in CT images due to metal regions.
method Geodesic X-ray transform on nontrapping compact Riemannian manifolds with strictly convex boundaries.
result Streaking artifacts result from conormal singularities along common tangent geodesics.

Deep learning predicts brain age from raw MRI data with high accuracy and reliability.

problem Predicting brain age from neuroimaging data to assess individual differences in brain aging.
method Convolutional Neural Networks (CNN) applied to raw T1-weighted MRI data.
result Brain-predicted age is a reliable and heritable biomarker of brain aging.

Unsupervised framework captures acquisition variability in structural connectomes.

problem Acquisition differences across sites, scanners, and protocols complicate structural connectome analysis.
method An unsupervised framework using architectural annealing to balance discrete and continuous latent variables.
result Architectural annealing produces stronger site learning than baseline models.

Study benchmarks methods for learning non-Cartesian k-space trajectories and reconstruction.

problem Benchmarking methods for learning non-Cartesian k-space trajectories and reconstruction.
method Comparing PILOT, BJORK, and HybLearn schemes to learn non-Cartesian k-space trajectories and reconstruction.
result HybLearn scheme outperforms other methods in learning and comparing non-Cartesian k-space trajectories and reconstruction.

This study assesses the reproducibility of 1H-MRS scans across different vendors and sessions.

problem Lack of harmonization in magnetic resonance spectroscopy protocols among vendors.
method Analysis of CV and ICC for within- and between-sessions, and correlation coefficients for across machines.
result Metabolite concentrations are highly reproducible across different vendors and sessions.

New estimators improve sparse semiparametric additive modeling.

problem Sparse semiparametric additive modeling with structured sparsity.
method Combines group subset selection with shrinkage for nonconvex optimization.
result New estimators outperform alternatives in synthetic and real-world data.

Self-supervised method enhances ultrasound images without needing clean targets.

problem Multiplicative speckle, acquisition blur, and scanner artifacts hamper ultrasound interpretation.
method Physics-guided degradation model trained on rotated/cropped patches with synthesized inputs.
result Achieves highest PSNR/SSIM across Gaussian and speckle noise levels, with significant improvements in heavy noise conditions.

Paper uses tensor regression to analyze point clouds for process optimization.

problem Challenges in modeling and analyzing high-dimensional point cloud data.
method Utilizes multilinear algebra and tensor regression techniques.
result Successfully models and links point cloud variational patterns to process variables.

Our analysis of financial data, in terms of super-exponential growth, suggests that the seed of the 2002/03 crisis of the Dutch supermarket giant AHOLD was planted in 1996. It became quite visible in 1999 when the post-bubble destabilization regime was well-developed and acted as the precursor of an inevitable collapse…

2004-03-22abs ↗pdf ↗

New algorithms for CLR with multiple observations and clusters.

problem Finding clusters of entities with minimized overall sum of squared errors.
method Exact mathematical programming, column generation, heuristic algorithms, genetic algorithms, modified Sp{ä}th algorithm.
result Performance comparison of algorithms on SKU clustering problem.

IDA adapts to non-iid data in federated learning for medical imaging.

problem Statistical heterogeneity in federated learning data, especially in medical imaging.
method IDA (Inverse Distance Aggregation) is a novel adaptive weighting approach for clients based on meta-information.
result IDA outperforms Federated Averaging in handling unbalanced and non-iid data in federated learning.

The study classifies Android malware using minhashing and Structural Equation Models.

problem Lack of consensus and consistency in malware signatures across different antivirus engines.
method Analyzed over 250k malware signatures from 61 engines on 82k Android apps.
result Identified 41 malware classes grouped into Adware, Harmful Threats, and Unknown categories.

Robots infer distances to invisible obstacles from 2D laser scans.

problem Mobile robots struggle with accurate distance estimation from 2D laser scanners.
method Trained a neural network to map raw 2D laser distances to actual obstacle distances.
result Trained network successfully infers distances from partial 2D laser readings in real-time.

Paper develops PGMM framework for debiased inference on nonparametric IV estimators.

problem Automatic debiased inference on nonparametric IV functionals.
method Penalized GMM (PGMM) framework for functionals of IV estimators.
result PGMM-based debiased estimator performs well, achieving near-nominal coverage.

KD-Net transfers knowledge from multi-modal to mono-modal segmentation networks.

problem Limited acquisition of multiple imaging modalities in clinical settings.
method Generalized distillation framework adapted for mono-modal networks.
result The student network outperforms baseline mono-modal networks in brain tumor segmentation.

PAD offers a principled approach to malware detection against evasion attacks.

problem Machine Learning techniques for malware detection are vulnerable to evasion attacks.
method PAD proposes a new adversarial training framework with convergence guarantees for robust optimization.
result PAD significantly outperforms state-of-the-art defenses and can harden ML-based malware detection against 27 evasion attacks.

A new house price prediction model using location data and multi-task learning.

problem Accurate house price prediction for various stakeholders.
method Location-centered data profiling and Multi-Task Learning (MTL) approach.
result MTL-based methods significantly outperform state-of-the-art approaches in house price prediction.