Deep Resolve XA31

SIEMENS Healthineers Deep Resolve Clinical Application Job Aid Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Table of Contents Deep Resolve Introduction 3 Deep Resolve Gain Overview 4 Deep Resolve Sharp Overview 5 Deep Resolve Protocol Transformation 7 Deep Resolve Gain (DR Gain) 8 Deep Resolve Gain – Application Strategy 9 Deep Resolve Gain and Sharp – Application Strategy 12 Deep Resolve – Retrospective Reconstruction 14 ttttt 16 2 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Protocol Overview HR 1 -spine_12_tse_sag AR 2020 08 07-09:39:29-00021 MAGNETOM Sola "1/1/1900, F. 120Y MR XA30 CURRENT OLPH Depending on the image quality that you want to achieve, you can select between the following combinations of Deep Resolve (DR): tem 1 1. Deep Resolve Gain: suppression of locally varying noise LL TP F397 combined by intelligent denoising. SP RI2 8 FoV 260"250 307'3845 Sag>Trat-1) A1/SATI Zoom 3.15 • Images acquired with Deep Resolve Gain contain the SP2-4 1053 tseR241m17 / 150 Not for dagnasbcuso abbreviation: DRG. HR 2 pine 12 tse sag 2020.08 07-09:39.29-0002 MAGNETOM Sola -1/1/1900. F. 120Y 2 MR XA30 1. Deep Resolve Gain and Deep Resolve Sharp: suppression CURRENT 8/7/2020 of locally varying noise combined with deep-learning- based increase of resolution and sharpness. • Images acquired with Deep Resolve Gain contain the 1 cm abbreviation: DRS. L TP F397 SP R128 oV 260-250 DRS/FILDIS2D 614-768¢ Sag>Traft A1/SATI Zoom 1:58 SP2-4 1048 tseR241m17 / 150 Not for dagngsbeluno EOS 3 3 Restricted © Siemens Healthineers, 2020 Unrestricted © Siemens Healthineers, 2022 Deep Resolve Gain SIEMENS Healthineers What Is It and How Does It Work? Intelligent, localized denoising Deep Resolve Gain uses the acquired MRI data to Deep Resolve Gain is an intelligent reconstruction method to generate a noise map which reflects spatial noise increase the SNR (signal-to-noise ratio) of images. variations. The MR image and the corresponding noise map are used as input for an iterative process. With this, either shorter scan times or higher image quality with higher resolution can be achieved. Noise Map The result Higher image quality and improved SNR. Deep Resolve Gain uses the acquired MRI data to generate a noise map which reflects spatial noise variations. The MR image and the corresponding noise map are 4 used as input for an iterative process. Restricted © Siemens Healthineers, 2020 Unrestricted © Siemens Healthineers, 2022 Deep Resolve Sharp SIEMENS Healthineers What Is It? AI & Deep Learning Reconstruction Deep Resolve Sharp uses a deep neural network to improve the MR image quality by increasing the sharpness. The Neural Network Anticipates where to expect a sharp edge in an image. 48% Generates high-resolution output from low- faster resolution input. Original: Deep Resolve Gain & Sharp MAGNETOM Vida, 3T MAGNETOM Vida PD TSE fs, 2 Av, TA 3:33 PD TSE fs, 1 Av, TA 1:50 Matrix size: 307 x 512 Matrix size: 608 x 608 The Benefits Higher image sharpness while also speeding up scan time 5 Restricted © Siemens Healthineers, 2020 Unrestricted © Siemens Healthineers, 2022 Deep Resolve Sharp SIEMENS Healthineers How Does it Work? During the acquisition, phase Deep neural network Robust results encoding lines can be reduced, thus Now AI comes into play: the To ensure consistency with the reducing the amount of data and algorithm in Deep Resolve Sharp has acquired data, the high-resolution speeding up scan time. From the been trained on a large number of estimation is collated with the acquired raw data, a low-resolution image is pairs of low-resolution and high- data. The result is an images with reconstructed in a first step. resolution MR data. It reconstructs a sharp edges and high resolution Deep sharp image from low resolution data Resolve Sharp offers up to a factor of resulting in a high-resolution two in-plane resolution. – estimation. Raw data input Low-res reconstruction High-res reconstruction O O o o O 0 o O o o O O O O o 6 2 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve – Protocol Transformation How to transform a protocol into a Deep Resolve protocol. Routine Contrast Resolution Geometry System Physio Inline Sequence 1 Under the Resolution parameter card, click on the Common Acceleration Filter 1. Acceleration mode GRAPPA 2 Advanced Reconstruction Deep Resolve Acceleration card. Phase Partial Fourier Off 2. In the Advanced Reconstruction drop-down menu, select the option Deep Resolve. Reference Scans Self-cali ... 3. Deep Resolve Gain and Sharp can only be combined Acceleration Factor PE with some specific image filters. Other filters might get Reference Lines PE 32 deactivated. If, therefore, additional changes are Routine Contrast Resolution Geometry System Physio Inline Sequence Common Acceleration Filter required, the Scan Assistant window opens and shows Acceleration mode GRAPPA Advanced Reconstruction Deep Resolve ... the required filter changes. 3 If the Deep Resolve Sharp license is available, it is Scan Assistant ? 4. x automatically selected when Deep Resolve Advanced Reference Scan Acceleration Factor P ? Your last change Reconstruction is activated. Reference Lines P Advanced Reconstruction Off - Deep Resolve • With the DR Sharp license, the user has the option to will adapt the following parameters: use DR Sharp along with DR Gain or use only DR Elliptical Filter On Off Interpolation Off On Gain. OK Undo 7 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain (DR Gain) 1 Application of Deep Resolve Gain without Deep Resolve Deep Resolve ? x Sharp. Deep Resolve Gain 2 1. After Deep Resolve has been activated, you can adjust Denoising Strength Edge Enhancement_M the denoising strength and the edge enhancement to Enhancement Level fine-tune the effects of Deep Resolve Gain. 3 Deep Resolve Sharp 2. Click on the dots next to the drop-down menu and adjust the following fields in the Deep Resolve window: Close Denoising Strength: low (=1) to high (=8) Routine Contrast Resolution Geometry System Physio Inline Sequence • Common Acceleration • Edge Enhancement: low (=1) to high (=5) FOV Read 180 mm Trajectory Cartesian 3. If the Deep Resolve Sharp license is available, in order FOV Phase 100.0 ¢ % Slice Thickness 1.5 - mm to use DR Gain without DR Sharp, the checkmark must Base Resolution 320 be removed from the Deep Resolve Sharp checkbox. Phase Resolution 75 % % • To deactivate interpolation, select under the Resolution parameters card the Common card and Interpolation v remove the checkmark from the Interpolation checkbox. 8 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain – Application Strategy 1. How to increase image quality. Routine Contrast Resolution Geometry System Physio Inline Sequence Common Acceleration Filter Acceleration mode GRAPPA 1 Advanced Reconstruction Deep Resolve ... Original protocol SNR level: low. Phase Partial Fourier Off By turning a protocol with low SNR into a DR Gain protocol, you can increase image quality with less noise Reference Scans Self-cali ... and more SNR. Acceleration Factor PE Reference Lines PE 32 1. Select Deep Resolve in the drop-down menu. 2. Click on the dots next to the drop-down menu to open Deep Resolve ? x the Deep Resolve window and modify the default denoising value in the Deep Resolve window. Deep Resolve Gain M 2 Denoising Strength or • A higher Denoising Strength results in higher SNR. Edge Enhancement M Enhancement Level Please note: In some cases, using DR Gain may lead to a decrease in the perceived image sharpness. Applying Deep Resolve Sharp increased edge enhancement and increasing the image resolution can be helpful here. Close 9 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain – Application Strategy 2. How to reduce scan time. Routine Contrast Resolution Geometry System Physio nline Sequence Slice Group FoV Read 140 - mm Slices 30 + FOV Phase 100.0 4 % Original protocol SNR level: normal. Slice Thickness 3.0 - mm Distance Factor 10 6 By turning a protocol with normal SNR into a DR Gain TR 3760.0 ms Position R77.8 A15.3 F3.4 ... TE 36.00 ms protocol, you can reduce the scan time with comparable T > C8.3 > 51.2 1 Orientation Averages 1 4 image quality. Phase Encoding Dir. R >>>> L Concatenations 12 Phase Oversampling 100 = % AutoAlign Knee > Standard ... Coil Elements BC The options to reduce the scan time are as follows: 36.00 1. If you have an average value of more than one, TE 12.00. . 95.00 ms decrease the number of the Averages by clicking on the arrow in the Routine task card. Routine Contrast Resolution Geometry System Physio Inline Sequence 2. Increase the PAT factor by selecting a higher value for Commo Acceleration Filter the parameter Acceleration Factor PE under the Acceleration mode GRAPPA Advanced Reconstruction Deep Resolve ... Phase Partial Fourier Off Acceleration card. Please note: These adjustments can cause some artifacts Reference Scans Self-cali ... to become more severe, e.g., due to the lack of averaging 2 Acceleration Factor PE of regular movements such as the blood flow. Reference Lines PE 32 10 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain – Application Strategy 3. How to reduce image artifacts. Routine Contrast Resolution Geometry System Physio Inline Sequence Part 1 Part 2 Assistant Sequence Name tseR 1 Protocols which are affected by either susceptibility Bandwidth 223 › Hz/PX artifacts or by image distortions around metal implants. Dimension 2D Echo Spacing 11.9 ms In order to improve the image quality, you can increase the bandwidth. RF Pulse Type Normal Define Turbo Factor 1. In the Sequence task card, adjust the value of the Turbo Factor 9 Bandwidth by clicking on the arrow. Gradient Mode Fast Echo Trains per Slice 42 2. Increasing the bandwidth renders the protocol more Flow Compensation None robust against artifacts - however a higher bandwidth also introduces additional noise. By using DR Gain, SNR can be recovered, and the additional noise can be removed. 11 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain and Sharp – Application Strategy 1. How to improve image quality and resolution Routine Contrast Resolution Geometry System Physio Inline Sequence without changing the scan time. Common Acceleration Filter 1 Acceleration mode GRAPPA Advanced Reconstruction Deep Resolve ... Phase Partial Fourier Off Original protocol SNR level: normal 1. Select Deep Resolve in the drop-down menu. Reference Scans Self-cali ... 2. Click on the dots next to the drop-down menu and Acceleration Factor PE 2 adjust the default values in the Deep Resolve Reference Lines PE 32 + window. 2 Deep Resolve ? X Please note: Deep Resolve Sharp may enhance existing Deep Resolve Gain M but hidden artefacts and bring back a little bit of noise. Denoising Strength 6 Edge Enhancement M Enhancement Level Deep Resolve Sharp M Close 12 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain and Sharp – Application Strategy 2. How to reduce scan time by a smaller matrix size. Routine Contrast Resolution Geometry System Physio Inline Sequence Common Acceleration Filter Select Deep Resolve in the drop-down menu. Acceleration mode GRAPPA 1 Advanced Reconstruction Deep Resolve 2 1. Phase Partial Fourier Off 2. Click on the dots next to the drop-down menu and Deep Resolve adjust the default values in the Deep Resolve window. Reference Scans Integrat ... 3. Open the Common card. Change the value in the Base Acceleration Factor PE Resolution field to reduce the matrix size and reduce Reference Lines PE 38 - the scan time. Routine Contrast Resolution Geometry System Physio Inline Sequence Additional adjustments could include decreasing the Common Acceleration Filter FoV Read 180 mm Trajectory Cartesian number of the Averages or increasing the Acceleration FoV Phase 100.0 6 Factor PE. Slice Thickness 3.1 - mm Please note: These adjustments can cause some artifacts 3 Base Resolution 320 + to become more severe. Phase Resolution 80 96 Interpolation 13 Unrestricted © Siemens Healthineers, 2022 Deep Resolve SIEMENS Healthineers Deep Resolve Gain and Sharp – Retrospective Reconstruction Deep Resolve Gain settings can be modified after the 3. The protocol list opens. Select the required protocol scan through Retrospective Reconstruction to trial and and click on the Edit button to edit a protocol for fine-tune Deep Resolve settings. RetroRecon. 1. After the images have been acquired, go to the upper right corner and select Job View. 2. Select the MR Reconstructions tab Y Progress Raidfile Status Action SIEMENS Stich, M Healthineers * 13.04 CH Job View Edit Protocol for RetroReco Short Term Storage Maintenance MeasUid Proto Data Publisher Reference Scans Self-cali ... 20.01.2021 11:49:29 done Acceleration Factor PE Add parameter to editor 20.01.2021 11:49:29 done Reference Lines PE 15 Unrestricted © Siemens Healthineers, 2022 Deep Resolve Sharp Gain and Sharp SIEMENS Healthineers Retrospective Reconstruction continued DR Sharp Interpolation without DR Sharp DR Sharp Denoising: 4 Denoising: 8 Edge Enhancement: 2 Edge Enhancement: 5 Study AHR Northwell Health In ELL HEALTH_SYOSSET_VIDA_BRAIN ?, M, 1D S 8-5 55184677 'M RM 13/34 5.0 ORG/NORM/DRS/DIS2D Tra> Not for diagnostic use _17/150 )RS/DIS2D 16 1 Restricted © Siemens Healthineers, 2020 Unrestricted © Siemens Healthineers, 2022 6 SIEMENS Healthineers Please note that the learning material is for training purposes only! For the proper use of the software or hardware, please always use the Operator Manual or Instructions for Use (hereinafter collectively “Operator Manual”) issued by Siemens Healthineers. This material is to be used as training material only and shall by no means substitute the Operator Manual. Any material used in this training will not be updated on a regular basis and does not necessarily reflect the latest version of the software and hardware available at the time of the training. The Operator's Manual shall be used as your main reference, in particular for relevant safety information like warnings and cautions. Note: Some functions shown in this material are optional and might not be part of your system. Certain products, product related claims or functionalities (hereinafter collectively “Functionality”) may not (yet) be commercially available in your country. Due to regulatory requirements, the future availability of said Functionalities in any specific country is not guaranteed. Please contact your local Siemens Healthineers sales representative for the most current information. The reproduction, transmission or distribution of this training or its contents is not permitted without express written authority. Offenders will be liable for damages. All names and data of patients, parameters and configuration dependent designations are fictional and examples only. All rights, including rights created by patent grant or registration of a utility model or design, are reserved. Copyright © Siemens Healthcare GmbH 2021 Siemens Healthineers Headquarters Local Contact Information Legal Manufacturer Siemens Healthcare GmbH Siemens Medical Solutions USA, Inc. Imaging & Therapy Systems Henkestr. 127 40 Liberty Blvd. Magnetic Resonance 91052 Erlangen, Germany Malvern, PA, USA 19355-9998 Henkestr. 127 Phone: +49 9131 84-0 Telephone: +1-888-826-9702 DE-91052 Erlangen, Germany siemens-healthineers.com siemens.com/healthcare Telephone: +49 9191 18-0 siemens.com/healthcare 17 Job Aid Effective Date: 03/13/2022 I HOOD05162003263175 Unrestricted © Siemens Healthineers, 2022