PUBLICATIONS
RESOURCES - MRI
Novel Deep Learning Super Resolution Algorithm Allows Substantial Improvement of Brain MRI Image Quality and Scan Time in a Multi Site Study
Presented at RSNA 2022 The purpose of the study was to evaluate image quality, image resolution, and overall diagnostic sufficiency of low resolution brain MRI images processed by a novel deep learning (DL) resolution enhancement algorithm, compared to images acquired with higher resolution from multiple sites and MRI scanners.
Large-scale implementation of an Artificial Intelligence Solution to improve patient experience and enhance productivity in MRI departments. Alliar Brazil
Presented by Alliar, Brazil at RSNA 2022 The purpose of this study was to evaluate if the adoption of a proprietary Artificial Intelligence Software that reduces scan time without compromising image quality is correlated with a decrease in the average scan time considering all imaging protocols (with and without the software), an increase in the numbe r o f possible scheduling slots for MRI scans
Machine Learning-based Iterative Image Reconstruction Algorithm Allows Significant Reduction in Brain MRI Scan Times
Collaborative work with Lawrence N. Tanenbaum MD FACR, RadNet PURPOSE This study aims to evaluate the scan time shortening potential of iQMR, a novel 3D image enhancement algorithm, in brain, lumbar spine, and cervical spine MRI. iQMR provides image enhancement and increased Signal-to-Noise Ratio (SNR), thus permitting substantial scan time reduction, without changing the imaging contrast, resolution, or scanner hardware or software. This study reports
THE EFFECT OF A POST-SCAN PROCESSING DENOISING SYSTEM ON IMAGE QUALITY AND MORPHOMETRIC ANALYSIS
MR image quality and subsequent brain morphometric analysis are inevitably affected by noise. The purpose of this study was to evaluate the effectiveness of an artificial intelligence (AI)-based post-scan processing denoising system, intelligent Quick Magnetic Resonance (iQMR), on MR image quality and brain morphometric analysis. Fig. 1. MR images (A) without denoising, (B) processed with intelligent Quick Magnetic Resonance, and (C) with the median filter.
Substantial Shortening of Neuro MR Scan Times Using Compressed-Sense and Machine Learning-assisted Image Enhancement Technology – Approved abstract ASNR 2021
To evaluate the ability to substantially reduce neuro MRI scans time by applying Compressed Sense (CS) with Machine Learning (ML) assisted image enhancement technology. Figure 1 Axial T1w 3D images from a Philips Ingenia 1.5T scanner. (A) Accelerated routine protocol (with CS-3), AT=2:15min (B) Fast protocol (accelerated withCS-4), AT=1:42 and (C) Fast ML processed protocol (accelerated with CS-4), AT=1:42
Machine Learning-assisted Resolution Enhancement Algorithm Allows for Substantial Reduction of Brain MRI Scan Time – Approved abstract ASNR 2021
Authors: R Shreter , J Brace , M Goldfeld , I Page , G Stenoien , L Mori , T Aharoni , L Lucato PURPOSE To assess scan time reduction in brain MRI scans by acquiring reduced-resolution faster protocols, post-processed by novel machine-learning (ML) assisted technology. Assessment was conducted by comparing quality, contrast, resolution, noise and diagnostic sufficiency of the images.
Implementation of AI-assisted Technology to Enhance Service Quality and Productivity in Outpatient Diagnostic Centers
This report was conducted with Houston Medical Imaging’s (HMI) outpatient diagnostic centers. An increase in demand for MRI scans at HMI prompted a Productivity and Service Quality improvement project. The project aimed to optimize the performance of 7 MRI scanners across 6 sites in order to facilitate more scans and ultimately improve patient service. 49% SCAN-TIME REDUCTION | HITACHI OASIS 1.2T | Brain, AX T2 RADAR
COMPARING ACCELERATION OF MRI BRAIN SCANS: COMPRESSED SENSING AND MACHINE LEARNING IMAGE PROCESSING TECHNOLOGIES
The report compared complete brain exams that were acquired by the site’s routine scan (11 minutes 50 seconds) and by accelerated acquisition (5 minutes 50 seconds, 50% scan time reduction). Acquired images were processed and reviewed blindly by acknowledged neuroradiologists. The results demonstrate definite preference towards IIR-processed fast MR scans, for all evaluation characteristics including scan time reduction, and image quality. (A) Standard scan (Benchmark), AT=4:22 min. (B) CS
Substantial Shortening of Neuro MR Scan Times Using Compressed-Senseand Machine Learning-assisted Image Enhancement Technology
To evaluate the ability to substantially reduce neuro MRI scans time by applying Compressed Sense (CS) with Machine Learning (ML) assisted imageenhancement technology. Figure 1 Axial T1w 3D images from a Philips Ingenia 1.5T scanner. (A) Accelerated routine protocol (with CS-3), AT=2:15min (B) Fast protocol (accelerated withCS-4), AT=1:42 and (C) Fast ML processed protocol (accelerated with CS-4), AT=1:42
Substantial shortening of brain and lumbar spine mr scan time using iqmr™ machine learning-based iterative image reconstruction algorithm: multi-site study
This study aims to evaluate the scan time shortening potential of iQMR, a novel 3D image enhancement algorithm, in brain and lumbar spine MRI. iQMR provides image enhancement and increased Signal – to Noise Ratio (SNR), thus permitting substantial scan time reduction, without changing the imaging contrast, resolution, scanner hardware, or software. This study reports findings of 56 patients (37 brain, 19 lumbar spine) from three sites. Image quality, diagnostic quality, brain grey-white (GW) matter differentiation and artifacts’ appearance of
Application of Machine Learning-based Iterative Image Reconstruction Algorithms for MRI Scan Time Reduction in Brain and Spine – Approved Abstract ASNR 2018
Collaborative work with Lawrence N. Tanenbaum MD FACR, RadNet PURPOSE To investigate whether accelerated brain and spine MR protocols enhanced with a machine-learning based algorithm (iQMR by MedicVision) can serve as a possible alternative to routine protocols.
Machine Learning-based Iterative Image Reconstruction Algorithm Allows Significant Reduction in Brain MRI Scan Times – Approved abstract ASNR 2018
To evaluate image quality, grey-white (GW) matter differentiation and overall diagnostic sufficiency of brain MRI images acquired with reduced scan time protocols, processed with a novel 3D image enhancement algorithm (“iQMR” by Medic Vision Ltd.) compared with images from sites’ routine protocols. Figure 1. Axial Conventional and processed – short protocols 1.5T imaging of normal brain. Conventional (upper row) T1-weighted imaging (A) and T2-weighted imaging (B)
MRI CASE STUDIES AND REPORTS
Large-Scale Deployment of iQMR, an AI-Assisted Technology, Enabling Short Protocols on MRI Scanners – Case Report
Alliar, the third-largest diagnostic imaging provider in Brazil has more than 100 outpatient imaging centers across the country. provides imaging services to 10,000 people a month and performs more than 800,000 MRI exams per year with more than 100 scanners of various makers and types. As the demand
Medic Vision and Nuclear Light Industry Ltd. Deliver AdvancedMRI Services in South Korea
This case study describes the implementation of fast MRI protocols at the Wiltse Memorial Hospital radiology department in the Suwon city of South Korea.Medic Vision’s iQMR system allowed the hospital to accommodate more patients per day while maintaining or even increasing image quality for diagnostics.
RESOURCES - CT
COMPARISON OF RECONSTRUCTION APPROACHES FOR IMPROVING LOW DOSE CT IMAGES
C.R. Deible*, K. Tung, J.M. Lacomis, C.R. Fuhrman, R. Jarosz, D.Gur. *University of Pittsburgh Medical Center, Pittsburgh, PA. RSNA Annual Meeting, Chicago, IL, November 2012.
PROSPECTIVE CLINICAL STUDY TO ASSESS IMAGE BASED ITERATIVE RECONSTRUCTION FOR ABDOMINAL CT ACQUIRED AT THREE RADIATION DOSE LEVELS
S. Pourjabbar, MD* , S. Singh, MD, R. Perez Johnston. MD, A.S. Shenoy-Bhangle, MD, S. Do, PhD, Shima Aran, MD, Michael Blake, MD, Anders Persson, MD, Mannudeep Kalra, MD. * Massachusetts General Hospital, Boston, MA. RSNA Annual Meeting, Chicago, IL, November 2012.
ROLE OF IMAGE-BASED ITERATIVE RECONSTRUCTION TECHNIQUE FOR LOW RADIATION DOSE CHEST CT
S. Pourjabbar, MD*; R.D.A. Khawaja, MD; S. Singh, MD; A. Padole, MD; D. Lira, MD; M. Kalra, MD. *Dept. of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA. Radiological Case Review, Applied Radiology, July 2013.
CO-REGISTERED IMAGE QUALITY COMPARISON IN HYBRID ITERATIVE RECONSTRUCTION TECHNIQUES: SAFIRE AND SAFECT
Seungwan Lee*, Aran Shima, Sarabjeet Singh, Mannudeep K. Kalra, Hee-Joung Kim, Synho Do, *Department of Radiology, Massachusetts General Hospital, Boston, MA. Medical Imaging 2013:Physics of Medical Imaging, edited by Robert M. Nishikawa, Bruce R. Whiting, and Christoph Hoeschen, Proc. of SPIE, Vol. 8668, 86683G-1.
SUBMILLISIEVERT CHEST CT WITH FILTERED BACK PROJECTION AND ITERATIVE RECONSTRUCTION TECHNIQUES
Atul Padole1, Sarabjeet Singh, Jeanne B. Ackman, Carol Wu, Synho Do, Sarvenaz Pourjabbar, Ranish Deedar Ali Khawaja, Alexi Otrakji, Subba Digumarthy, Jo-Anne Shepard, Mannudeep Kalra, AJR:203, October 2014.
CT imaging: radiation risk reduction– real-life experience in a metropolitan outpatient imaging network
John O. Johnson, MD* , Jon M. Robins, MD. * Imaging Healthcare Specialists, Radiology Medical Group, San Diego, CA. Journal of American College of Radiology 2012, No.9, pp. 808-813.
Evaluation of Noise Reduction Techniques in Chest CT
H.Moriya MD, K.Hashimoto, S. Muramatsu, M.Suzuki, H.Chiba, Y. Nakajou, Y. Ohashi, S. Tsukagoshi, N. Kameda, T. Tanaka Ohara General Hospital, Fukushima, Japan Ziosoft Inc, Canon Medical Systems Coporation, NAGASE & Co, Ltd