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.

169,181 papers · 148 categories

Trend · papers per month

count % of corpus

4 results for “DCE-MRI”

Automatic Renal Segmentation in DCE-MRI using Convolutional Neural Networkscs.CV

Paper proposes efficient method for automatic renal segmentation in DCE-MRI.

problem Automatic segmentation of renal parenchyma in DCE-MRI images.

method Cascaded application of two 3D CNNs for localization and segmentation.

result Achieved high segmentation accuracy with mean dice coefficients of 91.4 and 83.6 for normal and abnormal kidneys, respectively.

Renal Segmentation DCE-MRI Convolutional Neural Networks Kidney Diseases Children's Health

2017-12-19abs ↗pdf ↗

Deep learning-based prediction of response to HER2-targeted neoadjuvant chemotherapy from pre-treatment dynamic breast MRI: A multi-institutional validation studyq-bio.QM

Deep learning predicts response to HER2-targeted breast cancer therapy.

problem Predicting response to HER2-targeted neoadjuvant chemotherapy.

method Developed and validated a deep learning approach using pre-treatment dynamic breast MRI.

result Deep learning model achieved strong performance in predicting pathological complete response.

deep learning breast cancer neoadjuvant chemotherapy MRI pathological complete response

2020-01-22abs ↗pdf ↗

k-Space Deep Learning for Parallel MRI: Application to Time-Resolved MR Angiographycs.CV

Deep learning improves k-space data for faster MRI scans.

problem Limited temporal resolution and fixed k-space patterns in TWIST.

method k-space deep learning approach for parallel MRI.

result Accurate k-space interpolations for multiple coils.

MRI Deep learning Parallel MRI TWIST k-space

2018-06-03abs ↗pdf ↗

Unsupervised Deep Learning for MR Angiography with Flexible Temporal Resolutioneess.IV

New unsupervised deep learning method improves temporal resolution in tMRA.

problem Limited temporal resolution in tMRA due to fixed view-sharing scheme.

method Optimal transport driven cycle-consistent generative adversarial network (cycleGAN) without fully sampled k-space reference data.

result Can generate high quality reconstructions at various temporal resolutions.

MR angiography unsupervised learning time-resolved cycleGAN optimal transport

2020-03-29abs ↗pdf ↗