Karthick Raj Rajendran1, Dr. Rakesh. K. Gupta2, and Rupsa Bhattacharjee3
1Philips Health Systems, Philips India Limited, Chennai, India, 2Deaprtment of Radiology, Fortis Memorial Research Institute, Gurugram, India, 3Philips Health Systems, Philips India Limited, Gurugram, India
Synopsis
Compressed SENSE i.e parallel imaging combined with Compressed Sensing,
plays a vital role in reducing the overall imaging time without compromising
the image quality. Objective of the study is to analyze the impact of using
Compressed SENSE to improve the patient comfort by reducing the scan time. This
is achieved by implementing Compressed SENSE in T1, T2, PD Weighted TSE, T1
& T2 FFE based sequences in 2D & 3D mode in routine MR Examinations.
Introduction
Since the introduction of MRI, one of the biggest challenge is the time
taken per study. Thus, key development in the field of MRI related to patient
comfort is the focus on acceleration techniques1, 2, 3. One of the
vastly use technique to accelerate MR acquisition is parallel imaging technique
like (SENSE). Due to limitations with the parallel imaging techniques, there
was always a need to more robust technique to further reduce the scan time. Recently
developed novel Compressed SENSE4 technique takes the advantage of
both compressed sensing & parallel imaging (SENSE) by adopting balanced
sampling method. In this study, we have applied compressed SENSE in 90% of the routine
sequences in different anatomical regions (except DWI, Multivane enabled
sequence which contributes to 10% of the complete sequences) to improve the
patient comfort by reducing the sequence time, i.e., total time for the patient
to be inside the magnet and also the breath hold duration for certain scans for
critically ill patients. Results from accelerated scanning protocol were
evaluated in terms of time reduction with respect to routine non-Compressed
SENSE protocols for each anatomy. Methodology
In this study, we have acquired MRI data of total 50 human subjects: 5
subjects each in 10 different anatomies such as: routine brain, pituitary,
orbit, neck, cervical spine, lumbar spine, 3-station-angio, breast, MRCP, Knee,
Ankle, Hip, Neck at 3.0 T Ingenia (Philips, The Netherlands). Coils used for
each of these anatomies were: 15 channel head coil for brain, pituitary, orbit,
Ankle, 44 channel spine coil for cervical spine, and lumbar spine, 32 channel
torso coil for MRCP, 8 channel coil for knee joint. Necessary sequences for a
comprehensive MR study for each anatomy were acquired using Compressed SENSE. Sequences
were optimized to keep the same geometrical parameters (slice thickness, gap,
field of view and resolution). Compressed SENSE factors used for each sequence
were designed and optimized to achieve possible scan duration reduction, while
keeping the visual image quality comparable with previously used sequences
(non-Compressed SENSE protocols). For an example, our comprehensive brain
examination protocol includes T1W IR TSE, T2W TSE, DWI, 3D Flair, SWI, Inflow
angio and T1 3D Black-blood sequence when contrast is used. Resulting images
from the compressed SENSE enabled fast scans were evaluated by senior
radiologist to ensure similar visual appearance without losing any critical clinical
information. Results
In this example, we have reported comprehensive exams for brain & MRCP.
Table 1 shows the sequences used for a comprehensive brain, MRCP and liver 3
phase study with the geometric parameters, scan time and breath hold time with
and without compressed SNESE. For our comprehensive brain exam 31% reduction in
scan duration is achieved and when contrast is given 34% reduction is achieved,
for 3D MRCP 65% reduction in scan duration and for Liver 3 Phase 53% reduction
in breath hold duration is achieved. Table
2 shows the list of studies and scan time reduction achieved using Compressed
Sense. Discussion
The results achieved from our study showed, patient comfort can be
improved with Compressed SENSE by reducing the scan duration more than 30% for
one MRI study without compromising the image quality. This is possible with
Compressed SENSE because this method uses variable density incoherently
under-sampling of k-space which has full freedom to optimize k-space sampling
for best possible signal and image sharpness.
Implementing Compressed SENSE in routine clinical practice can bring
down the number of repeat scans due to patient moment, sedation time and can
increase the patient throughput. It can also improve the patient comfort by
reducing the breath hold time by more than 50% for such sequences. In an alternate
way, using Compressed SENSE can add more value by increasing the resolution of
an existing sequence or by adding extra sequences without increasing the scan
time.
Conclusion
Compressed SENSE can
be a solution for the increasing demand in MR to perform imaging in a shorter time
without compromising image quality to improve patient comfort. Acknowledgements
No acknowledgement found.References
[1] Lustig M, Donoho D, Pauly JM. Sparse MRI: The
application of compressed sensing for rapid MR imaging. Magn Reson Med. 2007
Dec;58(6):1182-95.
[2] Donoho DL.
Compressed sensing. IEEE Transactions on Information Theory.
2006;52(4):1289-306.
[3] Jaspan ON, Fleysher R, Lipton ML. Compressed sensing
MRI: a review of the clinical literature. Br J Radiol. 2015;88 (1056):20150487
[4] Geerts-Ossevoort, L.,
de Weerdt, E., Duijndam, A., van IJperen, G., Peeters, H., Doneva, M.,
Nijenhuis, M. and Huang, A., Speed done right. Every time.