Localizing 2D Ultrasound Probe from Ultrasound Image Sequences Using Deep Learning for Volume Reconstruction

2020 ◽  
pp. 97-105
Author(s):  
Kanta Miura ◽  
Koichi Ito ◽  
Takafumi Aoki ◽  
Jun Ohmiya ◽  
Satoshi Kondo
Keyword(s):  
Deep Learning ◽  
Ultrasound Image ◽  
Image Sequences ◽  
Ultrasound Probe ◽  
Volume Reconstruction ◽  
2D Ultrasound

Pose Estimation of 2D Ultrasound Probe from Ultrasound Image Sequences Using CNN and RNN

2021 ◽  
pp. 96-105
Author(s):  
Kanta Miura ◽  
Koichi Ito ◽  
Takafumi Aoki ◽  
Jun Ohmiya ◽  
Satoshi Kondo
Keyword(s):  
Pose Estimation ◽  
Ultrasound Image ◽  
Image Sequences ◽  
Ultrasound Probe ◽  
2D Ultrasound

Motion estimation for ultrasound image sequences using deep learning

2020 ◽  
Vol 148 (4) ◽  
pp. 2487-2487
Author(s):  
Skanda Bharadwaj ◽  
Mohamed Almekkawy
Keyword(s):  
Deep Learning ◽  
Motion Estimation ◽  
Ultrasound Image ◽  
Image Sequences

scholarly journals Sequential U-Net Architecture for Automatic Femoral Artery Segmentation in Ultrasound Images

2021 ◽  
Vol 7 (1) ◽  
pp. 158-161
Author(s):  
Ana Estrada Lugo ◽  
Niclas Bockelmann ◽  
Felix von Haxthausen
Keyword(s):  
Femoral Artery ◽  
Automatic Segmentation ◽  
Ultrasound Image ◽  
Ultrasound Images ◽  
Segmentation Result ◽  
Ultrasound Probe ◽  
Sequential Structure ◽  
Tracked Ultrasound ◽  
Vessel Volume ◽  
2D Ultrasound

Abstract This work compares three different approaches to automatically segment the femoral artery from 2D ultrasound images. Two of the architectures follow a sequential structure, where each ultrasound image is considered a slice of the whole vessel volume, and its previous segmentation result will be part of the input, thus leading to a spatial prior. The Dice score on test data show a better performance on the baseline U-Net (0.819) compared to the sequential U-Net approaches (0.633, 0.725) for the femoral artery segmentation. This could be attributed to the misalignment of the slices being used in those networks. A possible improvement could be assumed in the implementation of a spatially calibrated and tracked ultrasound probe. Overall, these results indicate promising approaches for an automatic segmentation of the femoral artery using 2D ultrasound data.

scholarly journals Versatile Low-Cost Volumetric 3D Ultrasound Imaging Using Gimbal-Assisted Distance Sensors and an Inertial Measurement Unit

Sensors ◽  
2020 ◽  
Vol 20 (22) ◽  
pp. 6613
Author(s):  
Taehyung Kim ◽  
Dong-Hyun Kang ◽  
Shinyong Shim ◽  
Maesoon Im ◽  
Bo Kyoung Seo ◽  
...  
Keyword(s):  
Inertial Measurement Unit ◽  
Low Cost ◽  
Ultrasound Image ◽  
3D Ultrasound ◽  
Measurement Unit ◽  
Ultrasound Images ◽  
Ultrasound Probe ◽  
Inertial Measurement ◽  
Position Data ◽  
2D Ultrasound

This study aims at creating low-cost, three-dimensional (3D), freehand ultrasound image reconstructions from commercial two-dimensional (2D) probes. The low-cost system that can be attached to a commercial 2D ultrasound probe consists of commercial ultrasonic distance sensors, a gimbal, and an inertial measurement unit (IMU). To calibrate irregular movements of the probe during scanning, relative position data were collected from the ultrasonic sensors that were attached to a gimbal. The directional information was provided from the IMU. All the data and 2D ultrasound images were combined using a personal computer to reconstruct 3D ultrasound image. The relative position error of the proposed system was less than 0.5%. The overall shape of the cystic mass in the breast phantom was similar to those from 2D and sections of 3D ultrasound images. Additionally, the pressure and deformations of lesions could be obtained and compensated by contacting the probe to the surface of the soft tissue using the acquired position data. The proposed method did not require any initial marks or receivers for the reconstruction of a 3D ultrasound image using a 2D ultrasound probe. Even though our system is less than $500, a valuable volumetric ultrasound image could be provided to the users.

scholarly journals Deep learning-based recognition of cell structures in fluorescence microscopy sequences with respect to their morphology on cells infected with Marburg virus

2020 ◽  
Vol 6 (3) ◽  
pp. 501-504
Author(s):  
Dennis Schmidt ◽  
Andreas Rausch ◽  
Thomas Schanze
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Infected Cell ◽  
Network Architecture ◽  
Image Sequences ◽  
Marburg Virus ◽  
Microscopic Image ◽  
Cell Structures ◽  
Fast Recognition ◽  
Subviral Particles

AbstractThe Institute of Virology at the Philipps-Universität Marburg is currently researching possible drugs to combat the Marburg virus. This involves classifying cell structures based on fluoroscopic microscopic image sequences. Conventionally, membranes of cells must be marked for better analysis, which is time consuming. In this work, an approach is presented to identify cell structures in images that are marked for subviral particles. It could be shown that there is a correlation between the distribution of subviral particles in an infected cell and the position of the cell’s structures. The segmentation is performed with a "Mask-R-CNN" algorithm, presented in this work. The model (a region-based convolutional neural network) is applied to enable a robust and fast recognition of cell structures. Furthermore, the network architecture is described. The proposed method is tested on data evaluated by experts. The results show a high potential and demonstrate that the method is suitable.

Sign in / Sign up

Export Citation Format

Share Document