Optimization Image Quality of Knee MRI Sagital Plane T2 Weighted TSE Sequences with Variations of Echo Train Length (ETL) on Cartesian and Blade Technique

The use of Cartesian and Blade techniques also affects image quality. The Cartesian technique is more commonly used in Knee MRI examinations. Cartesian technique is vulnerable to movement, giving rise to motion artifacts. This motion artifact is often caused by too long scanning time, so ETL settings are important. An alternative to reducing this risk is to use the Blade technique and by setting the appropriate ETL. The research objective was to determine the differences image quality and anatomical information and to determine the technique that produced quality images and MRI anatomical information on the sagittal knee section between the Cartesian ETL and Blade T2 Weighted TSE combination. This is experiment research. Samples of 10 volunteers, carried out scanning with Cartesian and Blade techniques with variations ETL 14, 16, 18. Focus of assessment on ACL, PCL, meniscus, fluid, and fat. Image quality includes SNR and CNR. The assessment of the anatomical information by a radiologist. Data analyzed with Anova, Friedman, and Wilcoxon test. The results study showed that overall, there are significant differences in image quality between Cartesian and Blade techniques with p-value 0.005. There are differences in anatomical information with p-value <0.001. The best technique from image quality (SNR and CNR) is Cartesian-ETL 14 and based on anatomical information, the best technique is using Blade-ETL 14. Image quality in the two techniques with ETL 14 there is no significant difference with p-value of 0.414, so that Cartesian and Blade-ETL 14 settings produce the same image quality.


Introduction
setting between ETL and Blade there are other parameters that can be combined with ETL as an option to reduce artifacts due to movement, the technique is Cartesian [22]. Cartesian technique is the process of entering data from the acquisition into K-space following the Cartesian axis in the direction of the X axis along the Y axis, but the technique is very susceptible to movement and this is the reality in the field that the cartesian technique is used as a routine TSE sequence in knee MRI examination [23,24]. Therefore, the writer is interested in wanting to further analyze the use of the Cartesian T2TSE and T2TSE Blade techniques combined with the selection of the right ETL parameters so as to produce a good image and fast time [21,24,25].

Method
This type of research is quantitative research with an experimental study approach. Experiments were carried out to determine differences in image quality (SNR and CNR) of Knee MRI on cartesian and blade techniques with variations in ETL values and anatomical information on the Sagital T2 weighted TSE sequence, in order to obtain a short scanning time with good image quality. Data was collected at Radiology Installation at Haji Hospital Surabaya. The research sample was 10 voulunteers, each of which was carried out two treatments, namely in the T2 TSE Cartesian sequence with T2 TSE Blade. The procedure of the study was by scanning the MRI scan of the sagittal T2 TSE sequence, by adjusting the ETL parameters contained in the tool with variations of 14, 16, and 18 by combining the Cartesian T2 and T2 TSE Blade sequences. The assessment of knee MRI image results by measuring the SNR and CNR values in the area to be evaluated by adjusting the ROI in the ACL, PCL, meniscus, fluid, and fat regions. CNR measurement is done by calculating the difference in the SNR value of two adjacent tissue. CNR in this study is the difference between SNR ACL with fluid, meniscus with fluid, and PCL with fat. An assessment of anatomic information is carried out by a Radiology Specialist. Controlled variables are made constant so that they are not affected by external factors not examined in the study include TR 5450,0 ms, TE 64,0 ms, FOV 170 cm, Receive Bandwith 150,0 Hz/px, Slice Thicknes 3mm, Flip Angle 150 0 , Matriks 320x320, Interslice gap/ Distance Factor 10%.
Data analysis was performed with Anova for SNR and CNR followed by Post Hoc test with LSD. Friedman test was conducted to find out the optimal technique to produce anatomical information clarity, then continued with the Wilcoxon test to determine the differences between the two Cartesian techniques and the Blade technique with variations in ETL with alpha 5%.

Results and Discussion
Study was conducted on 10 male and female voulenter patients, ranging in age from 21 years to 50 years who underwent normal knee MRI examinations with TES Cartesian T2 sequences and TAG Blade sagittal pieces with 3 variations of ETL, namely 14, 16, and 18. Picture of knee MRI images on the use of a combination of cartesian techniques and techniques with ETL values as shown in Figure 1.      Table 5.      The difference to the SNR value can be seen in graph 1. From the graph it is known that the relationship between the use of Cartesian and BLADE techniques with the provision of ETL variations. In general, Cartesian is superior to BLADE in terms of displaying SNR. ETL 14 has the best image quality compared to 16 and 18. The long echo train on T2 FSE MRI weighting can cause blurring, affecting the MRI image quality, which allows reducing the SNR value and certainly will affect the CNR, this is one of the weaknesses of long ETL [22,24]. Excessive ETL applications cause the signal amplitude to decrease because ETL applications are large enough with Time Echo (TE) which increasingly leads to phase coding to fill K-Space faster. The result, can reduce the signal amplification occurs but the scanning time will be faster [21,25]. The greater the ETL, the shorter the scan time. Because the K-space coding process will be faster, causing scanning time to be faster too, but the consideration in getting diagnostic information is not only due to the speed of the scan time, but how it can produce high-quality image quality, and in this case ETL 14 which produces high SNR value [25]. The difference to the CNR value can be seen in graph 2, from the graph shows the relationship between the use of the Cartesian technique and the BLADE technique with the same ETL variations. Graph 2 shows the CNR values in both Cartesian and Blade techniques by giving the same ETL variations, that the Cartesian technique with ETL 14 gives optimal image quality results, compared to ETL 16 and 18. The difference in Cartesian and Blade techniques is due to differences in data acquisition to K-space which is a space where signals in the form of frequencies originating from patients are stored. In the Cartesian technique the process of entering data from the acquisition into the K-space follows the Cartesian axis in the direction of the X axis along the Y axis, so that the signal results are spread evenly throughout the K-space area. Whereas in the Blade technique the process of entering data from acquisition into K-space is processed using fourier transforms through the reconstruction of the algorithm fast fourier transformation with radial Blade meaning the central area signal (center) with the highest intensity is scattered in all directions so that the distortion effect of the movement is covered or reduced [20,21].
The advantage of the Cartesian technique compared to the Blade technique is in terms of time, because when encoding data into K-space the Cartesian technique is faster than the Blade technique. Variation in ETL values can affect the image quality of the Cartesian and Blade techniques but there is no significant difference because in each echo range it has a different phase enchode for each TR but will fill the lines on the K-spacenya similarly in image formation [15,20] Long ETL in weighting T2 MRI Knee TSE can cause blurring that affects SNR and CNR signals which is a weakness of TSE. So that the Cartesian T2 TSE and T2 TSE Blade techniques have significant differences [16,24]. b. Differences in Cartesian and Blade techniques with ETL variations on anatomic information The difference in image quality with anatomic information in Cartesian and Blade techniques with ETL variations is the Blade ETL 14 technique because it is indicated by a large mean rank value. due to the influence of the use of BLADE trajectory which aims to reduce artifacts running well in suppressing artifacts. According to Elliott artifacts that appear can interfere with anatomical imagery and can cause inaccuracies in making a diagnosis [25]. The emergence of artifacts in the Cartesian sequence is what reduces the value of anatomic information, resulting in anatomical information differences between Cartesian and Blade. Differences in anatomical information between Cartesian and BLADE on knee MRI examination can also be caused by the influence of different sequential parameters used respectively. The use of Cartesian On the other hand only wants a high SNR even though the resolution is low. Patients also affect the onset of noise. Every time a proton changes (down) from high energy to low energy, it emits photons (radio waves) with a random phase and will contribute to background noise. This can also be possible if there is a movement in the patient, where the activity changes in energy will impact on the contribution of noise to the MRI image [17,25].
Knee MRI image quality can be seen from both the Cartesian and Blade techniques with variations ETL on the CNR value that the best technique is the Cartesian technique with ETL 14, because the contrast will be clearly seen. Then if seen from the variable ETL the higher the ETL value used, the image contrast will be worse. A good CNR can show the difference between normal and abnormal areas both anotomically and pathologically. This difference is influenced by MRI parameters. Difference in signal differences between the two tissues against noise will affect the results of the contrast value by showing a clear contrast difference on MRI imaging. The greater the CNR value, image contrast quality will be better [21,25] b. Consideration of aspects of Anatomic Information Knee MRI image quality on both Cartesian and Blade techniques with ETL variations on anatomic information that the best technique is the Blade technique with ETL 14, this is because the Blade technique is able to reduce motion artifacts so that the distortion effects of movement can be covered or reduced. In making decisions based on three considerations, SNR, CNR, and anatomical information, researchers choose the Blade technique with ETL 14 as the best technique for obtaining MRI Knee image quality and establish diagnosis.
From the Wilcoxon test results on anatomic information, Cartesian comparison with ETL 14 and Blade with ETL 14 produces a p-value of 0.414 (> 0.05), so there is no significant difference between cartesian with ETL14 and Blade with ETL 14. From the Wilcoxon test results for anatomic information, the comparison of Cartesian with ETL 14 and Blade with ETL 14 produces a p-value of 0.414 (> 0.05), so there is no significant difference between cartesian with ETL14 and Blade with ETL 14.

Conclusion and Suggestions
Based on research that has been done, it can be concluded that: a. There is a difference between SNR and CNR Knee MRI using Cartesian