Clinical study on the application of intelligent and rapid magnetic resonance technology in improving the quality of knee joint scan images

Clinical study on the application of intelligent and rapid magnetic resonance technology in improving the quality of knee joint scan images | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clinical study on the application of intelligent and rapid magnetic resonance technology in improving the quality of knee joint scan images hao wang, xinyi guo, naiqian liu, jinghong wang, yi cao, lin guo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8199495/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective: To explore the application value of Intelligent Quick Magnetic Resonance (IQMR) in improving the image quality of knee joint scans. Methods: A total of 100 patients suspected of having knee joint lesions underwent conventional and accelerated knee joint magnetic resonance examinations on a GE Discovery 750 3.0T MRI scanner. The sagittal PDWI-FS sequence of the knee joint was selected, and the fast PDWI-FS sequence (Fast-MR, NEX=1) and the original PDWI-FS sequence (Original-MR, NEX=2) were scanned respectively. The two groups of sequences were automatically transferred to the IQMR post-processing system to reconstruct Fast-IQMR and Original-IQMR images. Three radiologists independently gave subjective scores on a five-point scale for the lesion details, anatomical structure clarity, overall image artifacts, and overall image quality of the four groups of images in the PD-FS sequence. The signal values of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, cartilage, and fat, as well as the background noise intensity were measured. With the muscle as the background, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of each tissue were calculated, and the above indicators of the four groups of images were compared and analyzed. Results: The Fast-IQMR sequence reduced the average examination time by 49% compared with the Original-IQMR sequence. By Kendall's W test, in the four groups of images, the three subjective scores given by the three physicians showed significant consistency ( p <0.05). There were statistical differences in the three subjective scores among the four groups of images ( p <0.001). The SNR and CNR of the Fast-IQMR and Original-IQMR images in the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage were significantly higher than those of the Fast - MR and Original - MR images respectively (p < 0.001), and the SNR and CNR of the Fast - IQMR images were significantly higher than those of the Original - MR group images (p < 0.001). Conclusion: The images reconstructed by IQMR have certain improvements in both scanning speed and image quality compared with conventional images. The IQMR technology has significant advantages in improving the image quality of knee joint magnetic resonance scans, which can significantly improve the detail expression and clarity of the images, increase the SNR and CNR of the images, and more accurately show the scope of the lesions. Compared with the Original - MR images, the improvement effect of the image quality of the Fast - IQMR images is significant. That is, the Fast - IQMR sequence with NEX = 1 can not only ensure the image quality but also improve the scanning efficiency. Intelligent Quick Magnetic Resonance technology Number of excitations Knee joint Magnetic resonance scan Image quality Figures Figure 1 Figure 2 Figure 3 1. Introduction Magnetic Resonance Imaging (MRI) is a crucial tool in the field of modern medical diagnosis. Due to its advantages such as non-radiation, high resolution, and multi-parameter imaging, it has been widely used in the medical field [ 1 ] . Especially in the evaluation of musculoskeletal system diseases, such as ligament and meniscus injuries, cartilage injuries, fractures, degenerative changes, and tumors, MRI has demonstrated its unique advantages [ 2 ] . However, the traditional MRI scanning process is time-consuming and has high requirements for patients' cooperation. Particularly for patients with knee joint diseases, slight movements of the lower limbs and the need to maintain a static posture for a long time are likely to cause motion artifacts, which in turn affect the image quality and accurate diagnosis. With the development of magnetic resonance rapid imaging technology, MRI technology has made breakthroughs in image resolution, scanning speed, and clinical applications [ 3 – 4 ] . In terms of fast imaging algorithms for MRI, the earliest approach was to achieve fast magnetic resonance imaging by reducing the number of sampling points in the k-space. For example, the partial Fourier technique only acquires the central part and some high-frequency signals of the k-space, and then uses symmetry to complete the k-space data and generate a full image [ 5 ] . With the widespread application of phased-array coils, parallel acquisition techniques based on k-space undersampling have gradually been applied in clinical practice [ 6 – 7 ] . However, this hardware-based acceleration method has physical limitations. The higher the acceleration factor, the more the noise is amplified. Moreover, due to the requirement of incoherent sampling, the number of coils needs to be much larger than the acceleration factor, which undoubtedly greatly increases the hardware cost and limits the acceleration performance. Therefore, when hardware acceleration approaches its limit, the compressed sensing technology was subsequently proposed to further reduce the MR scanning time. It can reconstruct high-quality images with fewer acquired signals, achieving image reconstruction under high-multiple undersampling. While ensuring image quality, it greatly increases the acceleration factor and has been widely used [ 8 ] . However, at extremely high acceleration factors, aliasing artifacts are likely to occur, and the reconstruction technology required for compressed sensing has extremely high requirements for computing resources, which limits its clinical application [ 9 ] . In 2016, with the application of artificial intelligence-artificial neural networks in the field of magnetic resonance, the fast imaging technology and image reconstruction technology of magnetic resonance have witnessed new breakthroughs. Using artificial intelligence can not only restore the details and fine structures of MR images but also be compatible with online reconstruction algorithms to achieve more efficient image reconstruction [ 10 ] . In recent years, Intelligent Quick Magnetic Resonance (IQMR), as an emerging technology, combines deep learning algorithms, iterative reconstruction modules, and k-space correction modules. It is an image iterative reconstruction technology assisted by artificial intelligence (AI). The principle is to jointly estimate and separate noise and signals by calculating the features of each image information from the input raw data. The iteration process stops and the optimal value is output when the best image in the same MRI is obtained [ 11 ] . Previously, some studies have used IQMR to improve the image quality of shoulder joint, cervical spine, and cranial MRI [ 12 – 14 ] . Although good results and performance have been achieved, whether the IQMR technology can improve the MRI image quality for different sequences and different body parts needs further exploration. Moreover, how to make the IQMR technology better exert its advantages through the adjustment and coordination of different parameters is also a key point for further research. This study focuses on the application of IQMR technology in knee joint MRI scanning. The aim is to explore the clinical application potential of IQMR technology by comparing and analyzing the quality of IQMR sequence images with different numbers of excitations and conventional sequence images. By improving the image quality, the IQMR technology is expected to provide a more comfortable and efficient examination experience for patients with knee joint diseases, and at the same time improve the accuracy and efficiency of diagnosis. 1. Materials and Methods 1.1 General information A total of 100 patients who underwent MR examinations due to clinically suspected knee joint lesions from August 2024 to September 2024 at Tianjin Hospital in Tianjin were selected as the research subjects. Among them, there were 55 males and 45 females, aged from 23 to 85 years old, with an average age of (57.43 ± 14.85) years. Inclusion criteria: ① Patients presented with symptoms such as knee joint pain and limited knee joint mobility, and clinically there was suspicion of knee joint lesions; ② Patients needed to undergo knee joint MRI examination for definite diagnosis or condition assessment; ③ Patients or their legal guardians signed the informed consent form and agreed to use their imaging data in this study. Exclusion criteria: ① Patients received treatments or surgeries that might affect the diagnostic results of knee joint MRI during the study period; ② Patients had severe mental or cognitive disorders. 1.2 MRI Acquisition Scanning was performed using a GEDiscovery750 3.0T MRI scanner with an 8-channel dedicated knee coil. The patients were placed in a supine position, with their feet first into the scanner, and their hands naturally relaxed on both sides of the body. It was ensured that the mid-sagittal plane of the human body was accurately aligned with the midline of the examination bed. A fast sequence with NEX = 1 and an original sequence with NEX = 2 were used for PD-FS sagittal scanning. After the scanning was completed, the images were uploaded, and IQMR post-processing reconstruction was automatically carried out. All four sets of image data were transferred to the Picture Archiving and Communication System (PACS). All the image data were downloaded from this system in the standard DICOM format for subsequent analysis.Sequence scanning parameters(Table 1 ): Table 1 Sequence scanning parameters Parameters Fast-MR(NEX = 1) Fast-IQMR(NEX = 1) Original-MR(NEX = 2) Original-IQMR(NEX = 2) TR/ms 2200 2200 2200 2200 TE/ms 40 40 40 40 FOV/mm2 160×160 160×160 160×160 160×160 Matrix 256×224 256×224 256×224 256×224 Thickness/mm 3.5 3.5 3.5 3.5 NEX 1 1 2 2 Number of layers 18 18 18 18 Scanning time/s 1min17s 1min17s 2min30s 2min30s 2. Image analysis and indicators 2.1 Subjective evaluation Three radiologists independently conducted subjective evaluations of the images. Among them, Radiologist 1 has over 12 years of work experience and holds an associate senior professional technical title; Radiologist 2 has over 8 years of work experience and also holds an associate senior professional technical title; Radiologist 3 has 4 years of work experience and holds a junior professional technical title. The above three radiologists independently scored the following three key indicators of the four groups of images: lesion details, anatomical structure clarity, overall image artifacts, and overall image quality. The specific scoring criteria(Table 2 ). Table 2 Subjective Scoring Criteria Table Score points Evaluation content Lesion details, clarity of anatomical structures Overall image artifacts Overall image quality 1-point The details of the lesion and the anatomical structure are extremely unclear and cannot be used for diagnosis. Severe artifacts,making diagnosis impossible. Non-visualization 2 points The details of the lesions and anatomical structures are not clear, which has a significant impact on the diagnosis. Moderate artifacts,affect diagnosis. Bad 3 - point The details of the lesion and the clarity of the anatomical structure are acceptable, which has a certain impact on the diagnosis. Moderate artifacts, which do not affect the diagnosis Medium 4-point The details of the lesions and the clarity of the anatomical structures are relatively good, which basically has no impact on the diagnosis. Mild artifacts, which do not affect the diagnosis. Good 5 - point The details of the lesion and the anatomical structure are very clear, which has no impact on the diagnosis at all. Artifact-free Excellent 2.2 Objective evaluation After the image data is transmitted to the GE post-processing workstation, two radiologists are respectively tasked with delineating the Region of Interest (ROI) and measuring relevant parameters. In the sagittal plane of the knee joint PD-FS in four groups of images, the most clearly displayed and signal-uniform middle layer is selected. ROIs are delineated for the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, cartilage, and fat to measure the Signal Intensity (SI) values and the background noise intensity (Standard Deviation, SDnoise). The sizes of the ROIs are 60–70 mm², 20–30 mm², 20–30 mm², 20–30 mm², 10–20 mm², and 60–70 mm² respectively. Each measurement is repeated three times and the average value is taken to ensure the accuracy of the data. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage are calculated. The measurement method of background noise intensity is as follows: Select smaller regions at the four corners of the background area outside the image, measure the standard deviations of the signal intensities in the four regions, and take the average value as the noise (SDnoise). The calculation formulas are as follows [ 17 ] : SNR tibial plateau =SI tibial plateau /SD noise (1) SNR anterior cruciate ligament =SI anterior cruciate ligament /SD noise (2) SNR posterior cruciate ligament =SI posterior cruciate ligament /SD noise (3) SNR cartilage =SI cartilage /SD noise (4) CNR tibial plateau = (SI tibial plateau -SI fat )/SD noise (5) CNR anterior cruciate ligament =(SI anterior cruciate ligament -SI fat )/SD noise (6) CNR posterior cruciate ligament =(SI posterior cruciate ligament -SI fat )/SD noise (7) CNR cartilage =(SI cartilage -SI fat )/SD noise (8) SI is the measured signal intensity value, and SDnoise is the calculated average value of the background noise.The measurement(Figure 1 ): 3. Statistical analysis First, the Shapiro-Wilk test for normal distribution was performed on the measurement data (such as signal intensity, noise, signal-to-noise ratio, contrast-to-noise ratio,etc.). For data conforming to the normal distribution, the mean ± standard deviation was used for representation; for data not conforming to the normal distribution and rank data (such as subjective scores), the median (upper and lower quartiles) was used for representation.The data in this study conformed to the normal distribution.The ANOVA test was used to compare the statistical differences among the four groups of scanned images in terms of signal intensity (SI), average background noise (SD),signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR), and the Bonferroni method was used for multiple comparisons [ 18 ] .The Kendall’s W test was used to evaluate the consistency of the subjective scores of the images by three imaging diagnostic physicians. p < 0.05 significant consistency was considered to exist.The Friedman test was used to compare the overall subjective score differences of the four types of scanned images.This test is applicable to ordered categorical variables or continuous variables,and the Nemenyi test was used for multiple comparisons. All statistical analyses were performed using the R 4.4.2 software. p < 0.05 indicated that the results were statistically significant. 4. Results 4.1 Scanning time In this study, the scanning time of the 18-slice PDWI-FS Fast-MR was 1 minute and 17 seconds. By adjusting the sequence parameters and reducing the number of excitations (NEX = 1), the scanning time was reduced. The scanning time of the Fast - MR was 49% shorter than that of the Original-MR sequence (2 minutes and 30 seconds). The SNR and CNR of the Fast-IQMR sequence were significantly improved compared with those of the conventional Fast-MR and Original-MR groups ( p < 0.05). 4.2 Comparative analysis of subjective scores In the four groups of images, three subjective scores (lesion details, anatomical structure clarity, overall image artifacts, and overall image quality) performed by three physicians all showed significant consistency ( p < 0.05). There were statistical differences in the three subjective scores among the four groups of images ( p < 0.001) (Table 3 、4、5). Further post-hoc pairwise comparisons revealed that compared with the Original-MR images, the differences in the three subjective scores of the Fast-IQMR group images were all statistically significant ( p < 0.001). Compared with the Original-MR images, the differences in the three subjective scores of the Original-IQMR image group were all statistically significant ( p < 0.001). In the subjective evaluation, when comparing the four different reconstructed images of the bone marrow edema of the knee patella of the same patient (Fig. 2), the IQMR reconstructed images improved the display of the clarity and sharpness of the boundary of the bone marrow edema area compared with the conventional images. Note: The Nemenyi test was used for pairwise comparisons after the event, and all the differences were statistically significant, with all p values less than 0.05. Table 3 Comparison of uniformity scores of four groups of images Note: The Nemenyi test was used for pairwise comparisons after the event, and all the differences were statistically significant, with all p values less than 0.05. Uniformity 1 2 3 4 5 Q P N1 N = 97 1 0(0%) 20(21%) 77(79%) 0(0%) 0(0%) 158.53 < 2.2e-16 N1Q N = 97 1 0(0%) 0(0%) 0(0%) 75(77%) 22(23%) N2 N = 97 1 1(1.0%) 0(0%) 31(32%) 65(67%) 0(0%) N2Q N = 97 1 0(0%) 0(0%) 0(0%) 35(36%) 62(64%) Table 4 Comparison of artifact scores of four groups of images Artifact 1 2 3 4 5 Q P N1 N = 97 0(0%) 18(19%) 72(74%) 7(7.2%) 0(0%) 156.25 < 2.2e-16 N1Q N = 97 0(0%) 0(0%) 0(0%) 60(62%) 37(38%) N2 N = 97 0(0%) 0(0%) 23(24%) 72(74%) 2(2.1%) N2Q N = 97 0(0%) 0(0%) 0(0%) 16(16%) 81(84%) Table 5 Comparison of the overall image quality scores among the four groups Overall image quality 1 2 3 4 5 Q P N1 N = 97 0(0%) 11(11%) 83(86%) 3(3.1%) 0(0%) 155.25 < 2.2e-16 N1Q N = 97 0(0%) 0(0%) 0(0%) 45(46%) 52(54%) N2 N = 97 0(0%) 0(0%) 7(7.2%) 88(91%) 2(2.1%) N2Q N = 97 0(0%) 0(0%) 0(0%) 6(6.2%) 91(94%) 4.3 Comparative analysis of objective scoring In the four sets of images, significant differences were observed in the SNR and CNR data of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage ROI ( p < 0.001)(Table 6 ). Further pairwise comparisons revealed that the SNR and CNR values in the Fast-IQMR and Original-IQMR groups were significantly higher than those in the two conventional groups ( p < 0.001). Specifically, compared to the Fast-MR images, the SNR of the tibial plateau, SNR of the anterior cruciate ligament, SNR of the posterior cruciate ligament, SNR of cartilage, CNR of the tibial plateau, CNR of the anterior cruciate ligament, CNR of the posterior cruciate ligament, and CNR of cartilage in the Fast-IQMR images under the PD-FSSag sequence increased by 144.68%, 141.33%, 133.51%, 138.38%, 151.28%, 142.94%, 137.46%, and 124.36%, respectively.Compared with Original-MR images, the Original-IQMR group demonstrated improvements of 165.63%, 157.4%, 150.59%, 159.74%, 161.7%, 182.86%, 144.31%, and 158.2%, respectively (Fig. 3 ). Note: Pairwise comparisons after the event were conducted using the Nemenyi test, and all the differences were statistically significant, with all P values less than 0.05. Table 6 Comparison of the differences in SNR and CNR among the four groups of images Group SNR Tibial plateau SNR Anterior cruciate ligament SNR Posterior cruciate ligament SNR Cartilage CNR Tibial plateau CNR Anterior cruciate ligament CNR Posterior cruciate ligament CNR Cartilage NEX = 1 33.37(15.38, 70.11) 45.12(15.66, 100.62) 35.57(1.00, 79.52) 130.52(32.21, 262.90) 8.93(0.06, 60.28) 10.99(0.00, 76.94) 11.13(0.07, 68.22) 88.16(15.05, 193.79) NEX = 1 IQMR 81.65(20.21, 193.00) 108.89(36.68, 296.92) 83.06(1.00, 218.50) 311.14(85.52, 793.80) 22.44(0.23, 165.90) 26.70(0.00, 233.10) 26.43(0.03, 189.95) 197.80(44.11, 515.07) NEX = 2 46.42(20.35, 101.94) 63.13(17.38, 157.22) 46.90(7.07, 121.07) 176.96(67.04, 340.35) 11.96(0.06, 65.61) 13.89(0.16, 104.63) 16.88(0.41, 75.02) 118.35(10.07, 256.09) NEX = 2 IQMR 123.31(29.72, 278.29) 162.50(39.65, 547.50) 117.53(28.16, 395.33) 459.65(121.23, 1,091.43) 31.30(0.72, 230.63) 39.29(0.42, 369.25) 41.24(1.44, 192.6) 305.58(24.25, 814.86) 285.09 270.01 275.98 282.69 218.39 153.87 212.69 274.46 p <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 5. Discussion As a non-invasive and radiation-free examination method, magnetic resonance imaging (MRI) has been widely used in the precise diagnosis of the nervous system and musculoskeletal system. However, MRI examinations of the limb joints in the musculoskeletal system often face numerous challenges due to the relatively long acquisition time and the requirement for high patient cooperation. Currently, most studies on shortening the scanning time adopt methods such as reducing the number of excitations, half-scan technology, or parallel acquisition technology, but all of these come at the cost of reducing image quality [19] . Therefore, in clinical diagnosis and treatment, the focus is on balancing image quality, meeting clinical diagnostic requirements, and ensuring an appropriate scanning time. The emergence of IQMR technology provides new ideas and methods for solving this problem. It has significant advantages in improving the image quality of knee joint magnetic resonance scans and can significantly enhance the detail expression and clarity of the images. By leveraging the statistical prior knowledge of noise distribution and signal-to-noise ratio enhancement in MRI images, the low-quality images generated by short-time scanning sequences are processed through a deep learning module to determine the iterative image reconstruction parameters, and K-space correction is used to reduce noise and specific artifacts. [14] Ultimately, high-quality images can be obtained while significantly shortening the scanning time, which is of great significance for improving patient comfort, reducing motion artifacts, and enhancing diagnostic accuracy [20–21] . Currently, the widely studied and popular denoising method based on convolutional neural network (CNN) has been applied in MRI image reconstruction [22–25] . However, Fan et al. [26] believe that there are certain limitations because most of the current training uses simulated noise data, and a large number of real images have not been used for denoising training by sequences and body parts. In this study, the scores of image clarity, artifacts, and overall image quality of the images optimized by IQMR are better than those of the fast sequence and the conventional sequence. The reason is mainly related to the core algorithm of IQMR. The convolutional neural network formed by AI learning and training in the deep learning module of IQMR, which includes a collection of high-quality image feature parameters and optimal parameters for image post-processing, is the basis for the iterative reconstruction module to reconstruct and restore the input low-resolution images to high-resolution images [27–28] . Secondly, the multiple iterative reconstruction steps of IQMR technology have a good effect on noise filtering and contrast enhancement [29–30] . This study aims to explore the clinical application value of intelligent quick magnetic resonance (IQMR) technology in improving the image quality of knee joint magnetic resonance scans. Through comparative analysis of four groups of PD-FS sag sequence images, we found that the IQMR technology showed significant advantages in improving image quality, enhancing the expressiveness of image details, reducing artifacts, and increasing the signal-to - noise ratio (SNR) and contrast - to - noise ratio (CNR). First of all, from the perspective of image quality scores, the results of this study showed that the scores of the IQMR group in lesion details, anatomical structure clarity, overall image artifacts, and overall image quality were significantly higher than those of the conventional group. This result is consistent with the findings of previous studies by Kanemaru et al. [14] , which pointed out that the IQMR technology can significantly improve the clarity and contrast of images by optimizing scanning parameters and image processing algorithms, thereby enhancing the readability and diagnostic value of images. In this study, the IQMR technology also demonstrated its remarkable effect in improving the quality of knee joint magnetic resonance images, providing doctors with more accurate and reliable imaging reference. Secondly, in terms of SNR and CNR, the results of this study showed that the SNR of the tibial plateau, SNR of the anterior cruciate ligament, SNR of the posterior cruciate ligament, SNR of the cartilage, CNR of the tibial plateau, CNR of the anterior cruciate ligament, CNR of the posterior cruciate ligament, and CNR of the cartilage in each sequence of the IQMR group images were higher than those of the conventional group. This finding echoes the results of a study by Xu et al [13] , which also reported the significant effect of the IQMR technology in improving SNR and CNR. The improvement of SNR and CNR helps doctors to more accurately identify lesions and distinguish normal tissues from abnormal tissues, thereby improving the accuracy and reliability of diagnosis. In addition, it is worth noting that although the detection of lesions in conventional images is consistent with that in IQMR images, the IQMR reconstructed images show the scope of lesions more accurately. This finding may be closely related to the ability of IQMR technology to effectively reduce image noise and artifacts. Noise and artifacts have always been key factors affecting image quality [31–32] , and they can interfere with doctors' diagnostic judgments. Compared with traditional MRI denoising techniques [33] , IQMR technology can more effectively reduce the interference of noise and artifacts by optimizing scanning parameters and image processing algorithms, thereby improving the clarity and contrast of images [34] , enabling doctors to observe the fine structures of lumbar lesions more clearly. The advantage of this study lies in verifying that while IQMR improves image quality, it also determines how to use IQMR more efficiently by adjusting the key parameter NEX, which is directly related to the scanning time and image quality.Multiple studies have proven [25, 35–36] that the MR deep-learning image reconstruction technology can improve the quality of the input low-quality MR images by using the MR image feature parameters learned through machine learning. Therefore, the overall scanning time of the sequence can be reduced by minimizing image data acquisition and sequence design. Wang et al.'s research showed [37] that by reducing the data acquisition in the K-space and then using the deep-learning reconstruction technology, high-quality prostate images can be obtained with a deep-learning sequence. Moreover, the deep-learning sequence shortens the scanning time by 32.1% compared to the conventional sequence. However, the deep-learning sequence adopted by their team requires complex installation and is only compatible with the MR model and specific sequence they used. In contrast, the IQMR system used in this study, as a third-party image post-processing system, only applies deep-learning technology at the image reconstruction end. It can achieve fully automated image processing in the background and image transmission, with simpler installation and wider applicability to different MR models. In this study, the scanning time of the 18-layer PDWI-FS accelerated sequence is 1 minute and 17 seconds. By adjusting the sequence parameters and reducing the number of excitations (NEX = 1), the scanning time is decreased. The scanning time of the fast sequence is 49% shorter than that of the conventional sequence (2minutes and 30seconds). Less scanning time can reduce the discomfort of patients with knee joint injuries, effectively reduce motion artifacts, and significantly improve the efficiency of magnetic resonance examinations. However, the reduction in the number of excitations also leads to a decrease in the data filling the K-space, resulting in a decline in important image quality indicators such as the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and resolution. Therefore, it is necessary to introduce the IQMR system to perform post-processing on the images of the accelerated scanning sequence to improve the image quality. In this study, the SNR and CNR of the Fast-IQMR sequence are significantly higher than those of the Fast-MR group and the Original-MR group ( p < 0.05). However, this study also has certain deficiencies and limitations. First, the sample size of this study is limited, and it may not comprehensively reflect the application effects of the IQMR technology in all patients with knee joint diseases. Future studies need to expand the sample size to further verify the general applicability and accuracy of the IQMR technology. Second, this study only conducted a preliminary exploration of the application of the IQMR technology in knee joint magnetic resonance scanning under a single acceleration factor (changing NEX), and did not involve the evaluation and analysis of the effects of different acceleration parameters on image quality. Relevant research is currently underway. Finally, this study did not conduct an in-depth analysis of the cost - effectiveness of the IQMR technology. Although the IQMR technology improves the image quality, its cost may be relatively high. Therefore, future studies need to comprehensively consider the cost-effectiveness ratio of the IQMR technology to evaluate its feasibility in clinical practice. In conclusion, the IQMR technique has significant advantages in improving the image quality of knee joint magnetic resonance scans. It can not only significantly improve key indicators such as image quality scores, SNR, and CNR, but also more accurately display the scope of lesions and reduce the interference of noise and artifacts. With these remarkable advantages, the IQMR technique shows broad application potential and important clinical practical value in the accurate diagnosis of knee joint diseases. In the future, we will continue to explore the application effects of the IQMR technique in images of other body parts and other sequences to expand its clinical application scope and provide better medical services for patients. Declarations Authors' declaration of conflicts of interest: All authors declare no conflicts of interest. In accordance with the Declaration of Helsinki, this research has been approved by the Ethics Committee of Tianjin Hospital. The ethics approval number is: 2025 MedEthics Review 118. All the participants in the study agreed to have their results published. This research did not receive any financial support from any institution. Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Author Contribution Hao Wang:Writing-Original-Draft,Writing-Review.&·Editing.XinYi Guo:Investigation,Validation,Methodology.NaiQian Liu:Investigation,Methodology,Conceptualization.JingHong Wang:Investigation,Methodology.Yi Cao:Preparation, creation and/or presentation of the published workLin Guo:ProjectAdministration,Supervision,Review&Editing. Data Availability The data are, however, available upon request and with the permission of Norwegian Social Research (NOVA) References Sukerkar PA, Doyle Z. Imaging of Osteoarthritis of the Knee, Radiologic Clinics of North America 60 (2022) 605–16. https://doi.org/10.1016/j.rcl.2022.03.004 Eck BL, Yang M, Elias JJ, et al. Quantitative MRI for Evaluation of Musculoskeletal Disease: Cartilage and Muscle Composition, Joint Inflammation, and Biomechanics in Osteoarthritis. Invest Radiol. 2023;58:60–75. https://doi.org/10.1097/RLI.0000000000000909 . Yoon MA, Gold GE, Chaudhari AS. Accelerated Musculoskeletal Magnetic Resonance Imaging. J Magn Reson Imaging. 2024;60:1806–22. https://doi.org/10.1002/jmri.29205 . Safari M, Eidex Z, Chang C-W et al. Fast MRI Reconstruction Using Deep Learning-based Compressed Sensing: A Systematic Review, ArXiv (2024) arXiv:2405.00241v1. Feinberg DA, Hale JD, Watts JC, et al. Halving MR imaging time by conjugation: demonstration at 3.5 kG. Radiology. 1986;161:527–31. https://doi.org/10.1148/radiology.161.2.3763926 . Chang Y, Pham HA, Li Z. A dual-interpolator method for improving parallel MRI Reconstruction. Magn Reson Imaging. 2022;92:108–19. https://doi.org/10.1016/j.mri.2022.06.009 . Lee Y, Yoon S, Paek M, et al. Advanced MRI techniques in abdominal imaging. Abdom Radiol (NY). 2024;49:3615–36. https://doi.org/10.1007/s00261-024-04369-7 . Dieckmeyer M, Roy AG, Senapati J, et al. Effect of MRI acquisition acceleration via compressed sensing and parallel imaging on brain volumetry. MAGMA. 2021;34:487–97. https://doi.org/10.1007/s10334-020-00906-9 . Koori N, Yamamoto S, Kamekawa H, et al. Comparison of image quality evaluation methods for magnetic resonance imaging using compressed sensing-sensitivity encoding (CS-SENSE). Radiol Phys Technol. 2025;18:597–605. https://doi.org/10.1007/s12194-025-00911-4 . Wang S, Su Z, Ying L, ACCELERATING MAGNETIC RESONANCE IMAGING VIA DEEP LEARNING. Proc IEEE Int Symp Biomed Imaging,, 2016 (2016) 514–517. https://doi.org/10.1109/ISBI.2016.7493320 Do W-J, Seo S, Han Y, et al. Reconstruction of multicontrast MR images through deep learning. Med Phys. 2020;47:983–97. https://doi.org/10.1002/mp.14006 . Xie X, Wang J, Yan X. The Application Value of Intelligent Rapid MRI Technology in the Treatment of Ischial Tendon Injuries. Chin J Magn Reson Imaging. 2024.;15(10):148–52. 10.12015/issn.1674-8034.2024.10.025 . XU M, WU Y, LIU J. Application value of intelligent quick magnetic resonance technique in magnetic resonance scanning of cervical vertebra[J]. Chin J Magn Reson Imaging. 2023;14(10):111–5. 10.12015/issn.1674-8034.2023.10.019 . Kanemaru N, Takao H, Amemiya S. The effect of a post-scan processing denoising system on image quality and morphometric analysis. J Neuroradiol. 2022;49:205–12. https://doi.org/10.1016/j.neurad.2021.11.007 . Bratke G, Rau R, Weiss K, et al. Accelerated MRI of the Lumbar Spine Using Compressed Sensing: Quality and Efficiency. J Magn Reson Imaging. 2019;49:e164–75. https://doi.org/10.1002/jmri.26526 . Nölte I, Gerigk L, Brockmann MA, et al. MRI of degenerative lumbar spine disease: comparison of non-accelerated and parallel imaging. Neuroradiology. 2008;50:403–9. https://doi.org/10.1007/s00234-008-0363-0 . Maeder Y, Dunet V, Richard R, et al. Bone Marrow Metastases: T2-weighted Dixon Spin-Echo Fat Images Can Replace T1-weighted Spin-Echo Images. Radiology. 2018;286:948–59. https://doi.org/10.1148/radiol.2017170325 . Ueda T, Ohno Y, Yamamoto K, et al. Deep Learning Reconstruction of Diffusion-weighted MRI Improves Image Quality for Prostatic Imaging. Radiology. 2022;303:373–81. https://doi.org/10.1148/radiol.204097 . Cho SJ, Choi YJ, Chung SR, et al. High-resolution MRI using compressed sensing-sensitivity encoding (CS-SENSE) for patients with suspected neurovascular compression syndrome: comparison with the conventional SENSE parallel acquisition technique. Clin Radiol. 2019;74:e8179–81714. https://doi.org/10.1016/j.crad.2019.06.023 . Padole A, Digumarthy S, Flores E, et al. Assessment of chest CT at CTDIvol less than 1 mGy with iterative reconstruction techniques. Br J Radiol. 2017;90:20160625. https://doi.org/10.1259/bjr.20160625 . Padole A, Singh S, Ackman JB, et al. Submillisievert chest CT with filtered back projection and iterative reconstruction techniques. AJR Am J Roentgenol. 2014;203:772–81. https://doi.org/10.2214/AJR.13.12312 . Lu X, Liu WV, Yan Y, et al. Evaluation of deep learning-based reconstruction late gadolinium enhancement images for identifying patients with clinically unrecognized myocardial infarction. BMC Med Imaging. 2024;24:127. https://doi.org/10.1186/s12880-024-01308-2 . Ren J, Eriksen JG, Nijkamp J, et al. Comparing different CT, PET and MRI multi-modality image combinations for deep learning-based head and neck tumor segmentation. Acta Oncol. 2021;60:1399–406. https://doi.org/10.1080/0284186X.2021.1949034 . Eo T, Jun Y, Kim T, et al. KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images. Magn Reson Med. 2018;80:2188–201. https://doi.org/10.1002/mrm.27201 . Yoo H, Yoo R-E, Choi SH, et al. Deep learning-based reconstruction for acceleration of lumbar spine MRI: a prospective comparison with standard MRI. Eur Radiol. 2023;33:8656–68. https://doi.org/10.1007/s00330-023-09918-0 . Fan L, Zhang F, Fan H, et al. Brief review of image denoising techniques. Vis Comput Ind Biomed Art. 2019;2:7. https://doi.org/10.1186/s42492-019-0016-7 . Chen Q, Fang S, Yuchen Y, et al. Clinical feasibility of deep learning reconstruction in liver diffusion-weighted imaging: Improvement of image quality and impact on apparent diffusion coefficient value. Eur J Radiol. 2023;168:111149. https://doi.org/10.1016/j.ejrad.2023.111149 . Yang R, Zou Y, Liu WV, et al. High-Resolution Single-Shot Fast Spin-Echo MR Imaging with Deep Learning Reconstruction Algorithm Can Improve Repeatability and Reproducibility of Follicle Counting. J Clin Med. 2023;12:3234. https://doi.org/10.3390/jcm12093234 . Zhang X, Wang Y, Xu X, et al. Bladder MRI with deep learning-based reconstruction: a prospective evaluation of muscle invasiveness using VI-RADS. Abdom Radiol (NY). 2024;49:1615–25. https://doi.org/10.1007/s00261-024-04280-1 . Gao Y, Liu WV, Li L, et al. Usefulness of T2-Weighted Images with Deep-Learning-Based Reconstruction in Nasal Cartilage. Diagnostics (Basel). 2023;13:3044. https://doi.org/10.3390/diagnostics13193044 . Serai SD, Ho M-L, Artunduaga M, et al. Components of a magnetic resonance imaging system and their relationship to safety and image quality. Pediatr Radiol. 2021;51:716–23. https://doi.org/10.1007/s00247-020-04894-9 . Feuerriegel GC, Sutter R. Managing hardware-related metal artifacts in MRI: current and evolving techniques. Skeletal Radiol. 2024;53:1737–50. https://doi.org/10.1007/s00256-024-04624-4 . Khan SU, Ullah N, Ahmed I, Imaging MRI, et al. Comparison of MRI with other Modalities, Noise in MRI Images and Machine Learning Techniques for Noise Removal: A Review. Curr Med Imaging Rev. 2019;15:243–54. https://doi.org/10.2174/1573405614666180726124952 . Noordman CR, Yakar D, Bosma J, et al. Complexities of deep learning-based undersampled MR image reconstruction. Eur Radiol Exp. 2023;7:58. https://doi.org/10.1186/s41747-023-00372-7 . Liu Z, Wen B, Wang Z, et al. Deep learning-based reconstruction enhances image quality and improves diagnosis in magnetic resonance imaging of the shoulder joint. Quant Imaging Med Surg. 2024;14:2840–56. https://doi.org/10.21037/qims-23-1412 . Xie Y, Tao H, Li X, et al. Prospective Comparison of Standard and Deep Learning-reconstructed Turbo Spin-Echo MRI of the Shoulder. Radiology. 2024;310:e231405. https://doi.org/10.1148/radiol.231405 . Wang Y, Zhang X, Hu M. The application value of deep learning reconstruction technology in optimizing the scanning time and image quality of T2-weighted MRI scans for prostate imaging. J Magn Reson Imaging 202314(05):48–52. 10.12015/issn.1674-8034.2 05.010. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8199495","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":558901897,"identity":"01fe40a8-49aa-48ff-965c-66c457f00028","order_by":0,"name":"hao wang","email":"","orcid":"","institution":"Tianjin Hospital","correspondingAuthor":false,"prefix":"","firstName":"hao","middleName":"","lastName":"wang","suffix":""},{"id":558901898,"identity":"06b37203-49fe-48cb-b078-dbccfe914606","order_by":1,"name":"xinyi guo","email":"","orcid":"","institution":"Tianjin Hospital","correspondingAuthor":false,"prefix":"","firstName":"xinyi","middleName":"","lastName":"guo","suffix":""},{"id":558901899,"identity":"53f1c196-e92a-43b1-a567-695c30e987a3","order_by":2,"name":"naiqian liu","email":"","orcid":"","institution":"Tianjin University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"naiqian","middleName":"","lastName":"liu","suffix":""},{"id":558901900,"identity":"f8677b1b-dadf-4821-becd-014743884125","order_by":3,"name":"jinghong wang","email":"","orcid":"","institution":"Tianjin Hospital","correspondingAuthor":false,"prefix":"","firstName":"jinghong","middleName":"","lastName":"wang","suffix":""},{"id":558901901,"identity":"ef0aff21-7987-49be-b3fa-beff448b04de","order_by":4,"name":"yi cao","email":"","orcid":"","institution":"Tianjin Hospital","correspondingAuthor":false,"prefix":"","firstName":"yi","middleName":"","lastName":"cao","suffix":""},{"id":558901902,"identity":"c93b40a9-6fa1-42e8-84b3-d849ff5959b6","order_by":5,"name":"lin guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYFCCA2BSho2B+cCBDxVEaOCBauFhY2BLPDjjDFFa4DSP8WHeFiK02DOeMZP82mbDwyfd8+EAbwODPL/YAUK2nDGTlm1L42GTObvhgOQOBsOZsxOI0CLZdpiHTSJ3wwHDMwwJBreJ15Lz4EBiG5FaJD9CtDAcOEiUlgPHiq0ZzgH9IpFmcLDhjARhv7DPOLzx5o8yGzn5GcmPP/+psJHnlyaghUHihIk0DxKXgHIQ4G9//PEHEepGwSgYBaNgBAMA8t9EkDt0GiAAAAAASUVORK5CYII=","orcid":"","institution":"Tianjin Hospital","correspondingAuthor":true,"prefix":"","firstName":"lin","middleName":"","lastName":"guo","suffix":""}],"badges":[],"createdAt":"2025-11-25 06:23:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8199495/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8199495/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98432419,"identity":"d41fc9f7-f800-4ed1-90b9-e754f36e45b7","added_by":"auto","created_at":"2025-12-17 16:49:32","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":703965,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/35601012a68652fcd0f85468.docx"},{"id":98217944,"identity":"b4b3a979-77d8-4223-ad4f-6f971dc5c803","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8623,"visible":true,"origin":"","legend":"","description":"","filename":"c551dd0f56de44cca366ae012ee2f902.json","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/f06daef05e004646b2a03003.json"},{"id":98217945,"identity":"fa925495-32d3-4062-952d-85989472df03","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129084,"visible":true,"origin":"","legend":"","description":"","filename":"c551dd0f56de44cca366ae012ee2f9021enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/adb266024240e93ffbbceef8.xml"},{"id":98217933,"identity":"b74a4600-85a0-4ed3-816a-ebf08609f6ab","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/3f1956a09edb04eb6df22c7c.jpeg"},{"id":98433827,"identity":"03ae5c25-731a-43c6-b7a9-a98e9c7ca075","added_by":"auto","created_at":"2025-12-17 16:51:09","extension":"wmf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":566,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/7c12302d86ec89fd84545263.wmf"},{"id":98431860,"identity":"44161e61-06bf-4dde-bba2-a15517c1b642","added_by":"auto","created_at":"2025-12-17 16:48:29","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":194964,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/8bd8d28991995f1a76c89876.png"},{"id":98217934,"identity":"3c688a48-f8b8-48a7-8354-4f363711e786","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29086,"visible":true,"origin":"","legend":"","description":"","filename":"groupimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/7129b779351199f476bca202.jpeg"},{"id":98217935,"identity":"d9fbee91-3622-497c-88db-858bbb144092","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"wmf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":566,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.wmf","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/ba598bb17cf23ec5ed313e21.wmf"},{"id":98217940,"identity":"c77d7c9a-3f18-43a3-a6e4-8bc7fb00575c","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58846,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/79c3927b1e3c526fed7fde51.png"},{"id":98217947,"identity":"5cdf420d-0763-46f6-8147-cfbb233ef41a","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/b0edea569490e27707d7148f.png"},{"id":98432906,"identity":"434f2b78-e017-4b19-8d18-262d34d7bea3","added_by":"auto","created_at":"2025-12-17 16:50:06","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":383,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/f2334fb1f75f45ebf3deb2be.png"},{"id":98432455,"identity":"9980423f-9d5f-44dd-ae07-5973c196ac48","added_by":"auto","created_at":"2025-12-17 16:49:35","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91424,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/c40c44f5ea67b45e0c029026.png"},{"id":98432202,"identity":"cd99d200-889a-4108-a248-88854520c1ab","added_by":"auto","created_at":"2025-12-17 16:49:13","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17911,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinegroupimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/30d62f597390ac7592c3945a.png"},{"id":98433841,"identity":"a44cebd9-2bec-41cb-9ad3-241422afe977","added_by":"auto","created_at":"2025-12-17 16:51:09","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":383,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/d2a53fb17ed182c146166cbc.png"},{"id":98217946,"identity":"4e448871-3b1f-4d7e-867a-50eaa81dd6ec","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":123307,"visible":true,"origin":"","legend":"","description":"","filename":"c551dd0f56de44cca366ae012ee2f9021structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/31b70767d4af2feeb5a91d4c.xml"},{"id":98217948,"identity":"2842deba-a635-4b91-bcc0-f393a0998f94","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"html","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138554,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/afb27b380eebb417f4bff9c0.html"},{"id":98217932,"identity":"d9931b75-1cd2-4f13-ba41-b4a62ed21b20","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":171229,"visible":true,"origin":"","legend":"\u003cp\u003eThe measurement\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/deb943eca802ea536a0ab144.jpeg"},{"id":98433392,"identity":"fc40fdf0-ed01-4e3b-8cb0-cabc862ff7f3","added_by":"auto","created_at":"2025-12-17 16:50:42","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":156504,"visible":true,"origin":"","legend":"\u003cp\u003eShows four groups of different reconstructed images of the same patient,which are respectively:the original image with NEX=1(2A),the IQMR-reconstructed image with NEX=1(2B),the original image with NEX=2(2C), and the IQMR-reconstructed image with NEX=2(2D).\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/e9974fc0776533c53de07cf9.jpg"},{"id":98217930,"identity":"17dfbd97-5120-497c-af30-c6c1f9cbbba9","added_by":"auto","created_at":"2025-12-15 10:59:22","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":100466,"visible":true,"origin":"","legend":"\u003cp\u003eSignal intensity measurements of regions of interest in 100 patients, including (A) SNR tibial plateau, (B) SNR anterior cruciate ligament, (C) SNR posterior cruciate ligament, (D) SNR cartilage, (E) CNR tibial plateau, (F) CNR anterior cruciate ligament, (G) CNR posterior cruciate ligament, and (H) CNR cartilage. Results are presented as scatter plots with boxplots.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/07d22069757ec4cc510eba6f.jpg"},{"id":99315106,"identity":"f6231c88-5042-48ed-83f4-53dcb9ede8d4","added_by":"auto","created_at":"2025-12-31 16:26:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1298524,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8199495/v1/61ababe1-f97d-4a95-8bd3-aaa68bd07548.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Clinical study on the application of intelligent and rapid magnetic resonance technology in improving the quality of knee joint scan images","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMagnetic Resonance Imaging (MRI) is a crucial tool in the field of modern medical diagnosis. Due to its advantages such as non-radiation, high resolution, and multi-parameter imaging, it has been widely used in the medical field \u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. Especially in the evaluation of musculoskeletal system diseases, such as ligament and meniscus injuries, cartilage injuries, fractures, degenerative changes, and tumors, MRI has demonstrated its unique advantages \u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. However, the traditional MRI scanning process is time-consuming and has high requirements for patients' cooperation. Particularly for patients with knee joint diseases, slight movements of the lower limbs and the need to maintain a static posture for a long time are likely to cause motion artifacts, which in turn affect the image quality and accurate diagnosis. With the development of magnetic resonance rapid imaging technology, MRI technology has made breakthroughs in image resolution, scanning speed, and clinical applications \u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn terms of fast imaging algorithms for MRI, the earliest approach was to achieve fast magnetic resonance imaging by reducing the number of sampling points in the k-space. For example, the partial Fourier technique only acquires the central part and some high-frequency signals of the k-space, and then uses symmetry to complete the k-space data and generate a full image \u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. With the widespread application of phased-array coils, parallel acquisition techniques based on k-space undersampling have gradually been applied in clinical practice \u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. However, this hardware-based acceleration method has physical limitations. The higher the acceleration factor, the more the noise is amplified. Moreover, due to the requirement of incoherent sampling, the number of coils needs to be much larger than the acceleration factor, which undoubtedly greatly increases the hardware cost and limits the acceleration performance. Therefore, when hardware acceleration approaches its limit, the compressed sensing technology was subsequently proposed to further reduce the MR scanning time. It can reconstruct high-quality images with fewer acquired signals, achieving image reconstruction under high-multiple undersampling. While ensuring image quality, it greatly increases the acceleration factor and has been widely used \u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. However, at extremely high acceleration factors, aliasing artifacts are likely to occur, and the reconstruction technology required for compressed sensing has extremely high requirements for computing resources, which limits its clinical application \u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. In 2016, with the application of artificial intelligence-artificial neural networks in the field of magnetic resonance, the fast imaging technology and image reconstruction technology of magnetic resonance have witnessed new breakthroughs. Using artificial intelligence can not only restore the details and fine structures of MR images but also be compatible with online reconstruction algorithms to achieve more efficient image reconstruction \u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn recent years, Intelligent Quick Magnetic Resonance (IQMR), as an emerging technology, combines deep learning algorithms, iterative reconstruction modules, and k-space correction modules. It is an image iterative reconstruction technology assisted by artificial intelligence (AI). The principle is to jointly estimate and separate noise and signals by calculating the features of each image information from the input raw data. The iteration process stops and the optimal value is output when the best image in the same MRI is obtained \u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. Previously, some studies have used IQMR to improve the image quality of shoulder joint, cervical spine, and cranial MRI \u003csup\u003e[\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Although good results and performance have been achieved, whether the IQMR technology can improve the MRI image quality for different sequences and different body parts needs further exploration. Moreover, how to make the IQMR technology better exert its advantages through the adjustment and coordination of different parameters is also a key point for further research. This study focuses on the application of IQMR technology in knee joint MRI scanning. The aim is to explore the clinical application potential of IQMR technology by comparing and analyzing the quality of IQMR sequence images with different numbers of excitations and conventional sequence images. By improving the image quality, the IQMR technology is expected to provide a more comfortable and efficient examination experience for patients with knee joint diseases, and at the same time improve the accuracy and efficiency of diagnosis.\u003c/p\u003e"},{"header":"1. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e1.1 General information\u003c/h2\u003e\u003cp\u003eA total of 100 patients who underwent MR examinations due to clinically suspected knee joint lesions from August 2024 to September 2024 at Tianjin Hospital in Tianjin were selected as the research subjects. Among them, there were 55 males and 45 females, aged from 23 to 85 years old, with an average age of (57.43\u0026thinsp;\u0026plusmn;\u0026thinsp;14.85) years.\u003c/p\u003e\u003cp\u003e Inclusion criteria: ① Patients presented with symptoms such as knee joint pain and limited knee joint mobility, and clinically there was suspicion of knee joint lesions; ② Patients needed to undergo knee joint MRI examination for definite diagnosis or condition assessment; ③ Patients or their legal guardians signed the informed consent form and agreed to use their imaging data in this study. Exclusion criteria: ① Patients received treatments or surgeries that might affect the diagnostic results of knee joint MRI during the study period; ② Patients had severe mental or cognitive disorders.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e1.2 MRI Acquisition\u003c/h2\u003e\u003cp\u003eScanning was performed using a GEDiscovery750 3.0T MRI scanner with an 8-channel dedicated knee coil. The patients were placed in a supine position, with their feet first into the scanner, and their hands naturally relaxed on both sides of the body. It was ensured that the mid-sagittal plane of the human body was accurately aligned with the midline of the examination bed. A fast sequence with NEX\u0026thinsp;=\u0026thinsp;1 and an original sequence with NEX\u0026thinsp;=\u0026thinsp;2 were used for PD-FS sagittal scanning. After the scanning was completed, the images were uploaded, and IQMR post-processing reconstruction was automatically carried out. All four sets of image data were transferred to the Picture Archiving and Communication System (PACS). All the image data were downloaded from this system in the standard DICOM format for subsequent analysis.Sequence scanning parameters(Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e):\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSequence scanning parameters\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParameters\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFast-MR(NEX\u0026thinsp;=\u0026thinsp;1)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFast-IQMR(NEX\u0026thinsp;=\u0026thinsp;1)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eOriginal-MR(NEX\u0026thinsp;=\u0026thinsp;2)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOriginal-IQMR(NEX\u0026thinsp;=\u0026thinsp;2)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTR/ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2200\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTE/ms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFOV/mm2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e160\u0026times;160\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e160\u0026times;160\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e160\u0026times;160\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e160\u0026times;160\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMatrix\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e256\u0026times;224\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e256\u0026times;224\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e256\u0026times;224\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e256\u0026times;224\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eThickness/mm\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNEX\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of layers\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eScanning time/s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1min17s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1min17s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2min30s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2min30s\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"2. Image analysis and indicators","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Subjective evaluation\u003c/h2\u003e\u003cp\u003eThree radiologists independently conducted subjective evaluations of the images. Among them, Radiologist 1 has over 12 years of work experience and holds an associate senior professional technical title; Radiologist 2 has over 8 years of work experience and also holds an associate senior professional technical title; Radiologist 3 has 4 years of work experience and holds a junior professional technical title. The above three radiologists independently scored the following three key indicators of the four groups of images: lesion details, anatomical structure clarity, overall image artifacts, and overall image quality. The specific scoring criteria(Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSubjective Scoring Criteria Table\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eScore points\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eEvaluation content\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLesion details, clarity of anatomical structures\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eOverall image artifacts\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eOverall image quality\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1-point\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThe details of the lesion and the anatomical structure are extremely unclear and cannot be used for diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSevere artifacts,making diagnosis impossible.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNon-visualization\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2 points\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThe details of the lesions and anatomical structures are not clear, which has a significant impact on the diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate artifacts,affect diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBad\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3 - point\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThe details of the lesion and the clarity of the anatomical structure are acceptable, which has a certain impact on the diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate artifacts, which do not affect the diagnosis\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMedium\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4-point\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThe details of the lesions and the clarity of the anatomical structures are relatively good, which basically has no impact on the diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMild artifacts, which do not affect the diagnosis.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eGood\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5 - point\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eThe details of the lesion and the anatomical structure are very clear, which has no impact on the diagnosis at all.\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eArtifact-free\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eExcellent\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Objective evaluation\u003c/h2\u003e\u003cp\u003eAfter the image data is transmitted to the GE post-processing workstation, two radiologists are respectively tasked with delineating the Region of Interest (ROI) and measuring relevant parameters. In the sagittal plane of the knee joint PD-FS in four groups of images, the most clearly displayed and signal-uniform middle layer is selected. ROIs are delineated for the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, cartilage, and fat to measure the Signal Intensity (SI) values and the background noise intensity (Standard Deviation, SDnoise). The sizes of the ROIs are 60\u0026ndash;70 mm\u0026sup2;, 20\u0026ndash;30 mm\u0026sup2;, 20\u0026ndash;30 mm\u0026sup2;, 20\u0026ndash;30 mm\u0026sup2;, 10\u0026ndash;20 mm\u0026sup2;, and 60\u0026ndash;70 mm\u0026sup2; respectively. Each measurement is repeated three times and the average value is taken to ensure the accuracy of the data. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage are calculated.\u003c/p\u003e\u003cp\u003eThe measurement method of background noise intensity is as follows: Select smaller regions at the four corners of the background area outside the image, measure the standard deviations of the signal intensities in the four regions, and take the average value as the noise (SDnoise). The calculation formulas are as follows \u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e:\u003c/p\u003e\u003cp\u003eSNR\u003csub\u003etibial plateau\u003c/sub\u003e=SI\u003csub\u003etibial plateau\u003c/sub\u003e/SD\u003csub\u003enoise\u003c/sub\u003e (1)\u003c/p\u003e\u003cp\u003eSNR\u003csub\u003eanterior cruciate ligament\u003c/sub\u003e=SI\u003csub\u003eanterior cruciate ligament\u003c/sub\u003e/SD\u003csub\u003enoise\u003c/sub\u003e (2)\u003c/p\u003e\u003cp\u003eSNR\u003csub\u003eposterior cruciate ligament\u003c/sub\u003e=SI\u003csub\u003eposterior cruciate ligament\u003c/sub\u003e/SD\u003csub\u003enoise\u003c/sub\u003e (3)\u003c/p\u003e\u003cp\u003eSNR\u003csub\u003ecartilage\u003c/sub\u003e=SI\u003csub\u003ecartilage\u003c/sub\u003e/SD\u003csub\u003enoise\u003c/sub\u003e (4)\u003c/p\u003e\u003cp\u003eCNR\u003csub\u003etibial plateau\u003c/sub\u003e= (SI\u003csub\u003etibial plateau\u003c/sub\u003e-SI\u003csub\u003efat\u003c/sub\u003e)/SD\u003csub\u003enoise\u003c/sub\u003e (5)\u003c/p\u003e\u003cp\u003eCNR\u003csub\u003eanterior cruciate ligament\u003c/sub\u003e=(SI\u003csub\u003eanterior cruciate ligament\u003c/sub\u003e-SI\u003csub\u003efat\u003c/sub\u003e)/SD\u003csub\u003enoise\u003c/sub\u003e (6)\u003c/p\u003e\u003cp\u003eCNR\u003csub\u003eposterior cruciate ligament\u003c/sub\u003e=(SI\u003csub\u003eposterior cruciate ligament\u003c/sub\u003e-SI\u003csub\u003efat\u003c/sub\u003e)/SD\u003csub\u003enoise\u003c/sub\u003e (7)\u003c/p\u003e\u003cp\u003eCNR\u003csub\u003ecartilage\u003c/sub\u003e=(SI\u003csub\u003ecartilage\u003c/sub\u003e-SI\u003csub\u003efat\u003c/sub\u003e)/SD\u003csub\u003enoise\u003c/sub\u003e (8)\u003c/p\u003e\u003cp\u003eSI is the measured signal intensity value, and SDnoise is the calculated average value of the background noise.The measurement(Figure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e):\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Statistical analysis","content":"\u003cp\u003eFirst, the Shapiro-Wilk test for normal distribution was performed on the measurement data (such as signal intensity, noise, signal-to-noise ratio, contrast-to-noise ratio,etc.). For data conforming to the normal distribution, the mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation was used for representation; for data not conforming to the normal distribution and rank data (such as subjective scores), the median (upper and lower quartiles) was used for representation.The data in this study conformed to the normal distribution.The ANOVA test was used to compare the statistical differences among the four groups of scanned images in terms of signal intensity (SI), average background noise (SD),signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR), and the Bonferroni method was used for multiple comparisons \u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e.The Kendall\u0026rsquo;s W test was used to evaluate the consistency of the subjective scores of the images by three imaging diagnostic physicians.\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 significant consistency was considered to exist.The Friedman test was used to compare the overall subjective score differences of the four types of scanned images.This test is applicable to ordered categorical variables or continuous variables,and the Nemenyi test was used for multiple comparisons. All statistical analyses were performed using the R 4.4.2 software. \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicated that the results were statistically significant.\u003c/p\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003e4.1 Scanning time\u003c/h2\u003e\n \u003cp\u003eIn this study, the scanning time of the 18-slice PDWI-FS Fast-MR was 1 minute and 17 seconds. By adjusting the sequence parameters and reducing the number of excitations (NEX\u0026thinsp;=\u0026thinsp;1), the scanning time was reduced. The scanning time of the Fast - MR was 49% shorter than that of the Original-MR sequence (2 minutes and 30 seconds). The SNR and CNR of the Fast-IQMR sequence were significantly improved compared with those of the conventional Fast-MR and Original-MR groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e4.2 Comparative analysis of subjective scores\u003c/h2\u003e\n \u003cp\u003eIn the four groups of images, three subjective scores (lesion details, anatomical structure clarity, overall image artifacts, and overall image quality) performed by three physicians all showed significant consistency (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). There were statistical differences in the three subjective scores among the four groups of images (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e、4、5). Further post-hoc pairwise comparisons revealed that compared with the Original-MR images, the differences in the three subjective scores of the Fast-IQMR group images were all statistically significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Compared with the Original-MR images, the differences in the three subjective scores of the Original-IQMR image group were all statistically significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In the subjective evaluation, when comparing the four different reconstructed images of the bone marrow edema of the knee patella of the same patient (Fig.\u0026nbsp;2), the IQMR reconstructed images improved the display of the clarity and sharpness of the boundary of the bone marrow edema area compared with the conventional images.\u003c/p\u003e\n \u003cp\u003eNote: The Nemenyi test was used for pairwise comparisons after the event, and all the differences were statistically significant, with all\u0026nbsp;\u003cem\u003ep\u003c/em\u003e values less than 0.05.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u003cbr\u003e\u003cbr\u003e\n \u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eComparison of uniformity scores of four groups of images Note: The Nemenyi test was used for pairwise comparisons after the event, and all the differences were statistically significant, with all \u003cem\u003ep\u003c/em\u003e values less than 0.05.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eUniformity\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eQ\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20(21%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e77(79%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" rowspan=\"4\"\u003e\n \u003cp\u003e158.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" rowspan=\"4\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;2.2e-16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e75(77%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22(23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1(1.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e31(32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e65(67%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003csup\u003e\u003cem\u003e1\u003c/em\u003e\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e35(36%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e62(64%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab4\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eComparison of artifact scores of four groups of images\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eArtifact\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eQ\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18(19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e72(74%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7(7.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" rowspan=\"4\"\u003e\n \u003cp\u003e156.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" rowspan=\"4\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;2.2e-16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e60(62%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e37(38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e23(24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e72(74%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2(2.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16(16%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e81(84%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv align=\"left\" class=\"colspec\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003ctable id=\"Tab5\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eComparison of the overall image quality scores among the four groups\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOverall image quality\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eQ\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11(11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e83(86%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3(3.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" rowspan=\"4\"\u003e\n \u003cp\u003e155.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" rowspan=\"4\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;2.2e-16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN1Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45(46%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e52(54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7(7.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e88(91%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2(2.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN2Q\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0(0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6(6.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e91(94%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003e4.3 Comparative analysis of objective scoring\u003c/h2\u003e\n \u003cp\u003eIn the four sets of images, significant differences were observed in the SNR and CNR data of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage ROI (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001)(Table \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e). Further pairwise comparisons revealed that the SNR and CNR values in the Fast-IQMR and Original-IQMR groups were significantly higher than those in the two conventional groups (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Specifically, compared to the Fast-MR images, the SNR of the tibial plateau, SNR of the anterior cruciate ligament, SNR of the posterior cruciate ligament, SNR of cartilage, CNR of the tibial plateau, CNR of the anterior cruciate ligament, CNR of the posterior cruciate ligament, and CNR of cartilage in the Fast-IQMR images under the PD-FSSag sequence increased by 144.68%, 141.33%, 133.51%, 138.38%, 151.28%, 142.94%, 137.46%, and 124.36%, respectively.Compared with Original-MR images, the Original-IQMR group demonstrated improvements of 165.63%, 157.4%, 150.59%, 159.74%, 161.7%, 182.86%, 144.31%, and 158.2%, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003eNote: Pairwise comparisons after the event were conducted using the Nemenyi test, and all the differences were statistically significant, with all P values less than 0.05.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab6\" border=\"1\" class=\"fr-table-selection-hover\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eComparison of the differences in SNR and CNR among the four groups of images\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eGroup\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSNR\u003csub\u003eTibial plateau\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSNR\u003csub\u003eAnterior cruciate ligament\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSNR\u003csub\u003ePosterior cruciate ligament\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSNR\u003csub\u003eCartilage\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCNR\u003csub\u003eTibial plateau\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCNR\u003csub\u003eAnterior cruciate ligament\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCNR\u003csub\u003ePosterior cruciate ligament\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCNR\u003csub\u003eCartilage\u003c/sub\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNEX\u0026thinsp;=\u0026thinsp;1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e33.37(15.38, 70.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e45.12(15.66, 100.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e35.57(1.00, 79.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e130.52(32.21, 262.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8.93(0.06, 60.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10.99(0.00, 76.94)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11.13(0.07, 68.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e88.16(15.05, 193.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNEX\u0026thinsp;=\u0026thinsp;1 IQMR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e81.65(20.21, 193.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e108.89(36.68, 296.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e83.06(1.00, 218.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e311.14(85.52, 793.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e22.44(0.23, 165.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26.70(0.00, 233.10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26.43(0.03, 189.95)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e197.80(44.11, 515.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNEX\u0026thinsp;=\u0026thinsp;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e46.42(20.35, 101.94)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e63.13(17.38, 157.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e46.90(7.07, 121.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e176.96(67.04, 340.35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11.96(0.06, 65.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13.89(0.16, 104.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e16.88(0.41, 75.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e118.35(10.07, 256.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNEX\u0026thinsp;=\u0026thinsp;2 IQMR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e123.31(29.72, 278.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e162.50(39.65, 547.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e117.53(28.16, 395.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e459.65(121.23, 1,091.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e31.30(0.72, 230.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e39.29(0.42, 369.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e41.24(1.44, 192.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e305.58(24.25, 814.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e285.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e270.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e275.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e282.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e218.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e153.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e212.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e274.46\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cp\u003eAs a non-invasive and radiation-free examination method, magnetic resonance imaging (MRI) has been widely used in the precise diagnosis of the nervous system and musculoskeletal system. However, MRI examinations of the limb joints in the musculoskeletal system often face numerous challenges due to the relatively long acquisition time and the requirement for high patient cooperation. Currently, most studies on shortening the scanning time adopt methods such as reducing the number of excitations, half-scan technology, or parallel acquisition technology, but all of these come at the cost of reducing image quality\u003csup\u003e[19]\u003c/sup\u003e. Therefore, in clinical diagnosis and treatment, the focus is on balancing image quality, meeting clinical diagnostic requirements, and ensuring an appropriate scanning time. The emergence of IQMR technology provides new ideas and methods for solving this problem. It has significant advantages in improving the image quality of knee joint magnetic resonance scans and can significantly enhance the detail expression and clarity of the images.\u003c/p\u003e\n\u003cp\u003eBy leveraging the statistical prior knowledge of noise distribution and signal-to-noise ratio enhancement in MRI images, the low-quality images generated by short-time scanning sequences are processed through a deep learning module to determine the iterative image reconstruction parameters, and K-space correction is used to reduce noise and specific artifacts.\u003csup\u003e[14]\u003c/sup\u003eUltimately, high-quality images can be obtained while significantly shortening the scanning time, which is of great significance for improving patient comfort, reducing motion artifacts, and enhancing diagnostic accuracy\u003csup\u003e[20\u0026ndash;21]\u003c/sup\u003e. Currently, the widely studied and popular denoising method based on convolutional neural network (CNN) has been applied in MRI image reconstruction\u003csup\u003e[22\u0026ndash;25]\u003c/sup\u003e. However, Fan et al.\u003csup\u003e[26]\u003c/sup\u003ebelieve that there are certain limitations because most of the current training uses simulated noise data, and a large number of real images have not been used for denoising training by sequences and body parts. In this study, the scores of image clarity, artifacts, and overall image quality of the images optimized by IQMR are better than those of the fast sequence and the conventional sequence. The reason is mainly related to the core algorithm of IQMR. The convolutional neural network formed by AI learning and training in the deep learning module of IQMR, which includes a collection of high-quality image feature parameters and optimal parameters for image post-processing, is the basis for the iterative reconstruction module to reconstruct and restore the input low-resolution images to high-resolution images\u003csup\u003e[27\u0026ndash;28]\u003c/sup\u003e. Secondly, the multiple iterative reconstruction steps of IQMR technology have a good effect on noise filtering and contrast enhancement\u003csup\u003e[29\u0026ndash;30]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThis study aims to explore the clinical application value of intelligent quick magnetic resonance (IQMR) technology in improving the image quality of knee joint magnetic resonance scans. Through comparative analysis of four groups of PD-FS sag sequence images, we found that the IQMR technology showed significant advantages in improving image quality, enhancing the expressiveness of image details, reducing artifacts, and increasing the signal-to - noise ratio (SNR) and contrast - to - noise ratio (CNR). First of all, from the perspective of image quality scores, the results of this study showed that the scores of the IQMR group in lesion details, anatomical structure clarity, overall image artifacts, and overall image quality were significantly higher than those of the conventional group. This result is consistent with the findings of previous studies by Kanemaru et al.\u003csup\u003e[14]\u003c/sup\u003e, which pointed out that the IQMR technology can significantly improve the clarity and contrast of images by optimizing scanning parameters and image processing algorithms, thereby enhancing the readability and diagnostic value of images. In this study, the IQMR technology also demonstrated its remarkable effect in improving the quality of knee joint magnetic resonance images, providing doctors with more accurate and reliable imaging reference. Secondly, in terms of SNR and CNR, the results of this study showed that the SNR of the tibial plateau, SNR of the anterior cruciate ligament, SNR of the posterior cruciate ligament, SNR of the cartilage, CNR of the tibial plateau, CNR of the anterior cruciate ligament, CNR of the posterior cruciate ligament, and CNR of the cartilage in each sequence of the IQMR group images were higher than those of the conventional group. This finding echoes the results of a study by Xu et al\u003csup\u003e[13]\u003c/sup\u003e, which also reported the significant effect of the IQMR technology in improving SNR and CNR. The improvement of SNR and CNR helps doctors to more accurately identify lesions and distinguish normal tissues from abnormal tissues, thereby improving the accuracy and reliability of diagnosis.\u003c/p\u003e\n\u003cp\u003eIn addition, it is worth noting that although the detection of lesions in conventional images is consistent with that in IQMR images, the IQMR reconstructed images show the scope of lesions more accurately. This finding may be closely related to the ability of IQMR technology to effectively reduce image noise and artifacts. Noise and artifacts have always been key factors affecting image quality\u003csup\u003e[31\u0026ndash;32]\u003c/sup\u003e, and they can interfere with doctors\u0026apos; diagnostic judgments. Compared with traditional MRI denoising techniques\u003csup\u003e[33]\u003c/sup\u003e, IQMR technology can more effectively reduce the interference of noise and artifacts by optimizing scanning parameters and image processing algorithms, thereby improving the clarity and contrast of images\u003csup\u003e[34]\u003c/sup\u003e, enabling doctors to observe the fine structures of lumbar lesions more clearly.\u003c/p\u003e\n\u003cp\u003eThe advantage of this study lies in verifying that while IQMR improves image quality, it also determines how to use IQMR more efficiently by adjusting the key parameter NEX, which is directly related to the scanning time and image quality.Multiple studies have proven\u003csup\u003e[25, 35\u0026ndash;36]\u003c/sup\u003e that the MR deep-learning image reconstruction technology can improve the quality of the input low-quality MR images by using the MR image feature parameters learned through machine learning. Therefore, the overall scanning time of the sequence can be reduced by minimizing image data acquisition and sequence design. Wang et al.\u0026apos;s research showed\u003csup\u003e[37]\u003c/sup\u003e that by reducing the data acquisition in the K-space and then using the deep-learning reconstruction technology, high-quality prostate images can be obtained with a deep-learning sequence. Moreover, the deep-learning sequence shortens the scanning time by 32.1% compared to the conventional sequence. However, the deep-learning sequence adopted by their team requires complex installation and is only compatible with the MR model and specific sequence they used. In contrast, the IQMR system used in this study, as a third-party image post-processing system, only applies deep-learning technology at the image reconstruction end. It can achieve fully automated image processing in the background and image transmission, with simpler installation and wider applicability to different MR models. In this study, the scanning time of the 18-layer PDWI-FS accelerated sequence is 1 minute and 17 seconds. By adjusting the sequence parameters and reducing the number of excitations (NEX\u0026thinsp;=\u0026thinsp;1), the scanning time is decreased. The scanning time of the fast sequence is 49% shorter than that of the conventional sequence (2minutes and 30seconds). Less scanning time can reduce the discomfort of patients with knee joint injuries, effectively reduce motion artifacts, and significantly improve the efficiency of magnetic resonance examinations. However, the reduction in the number of excitations also leads to a decrease in the data filling the K-space, resulting in a decline in important image quality indicators such as the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and resolution. Therefore, it is necessary to introduce the IQMR system to perform post-processing on the images of the accelerated scanning sequence to improve the image quality. In this study, the SNR and CNR of the Fast-IQMR sequence are significantly higher than those of the Fast-MR group and the Original-MR group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\n\u003cp\u003eHowever, this study also has certain deficiencies and limitations. First, the sample size of this study is limited, and it may not comprehensively reflect the application effects of the IQMR technology in all patients with knee joint diseases. Future studies need to expand the sample size to further verify the general applicability and accuracy of the IQMR technology. Second, this study only conducted a preliminary exploration of the application of the IQMR technology in knee joint magnetic resonance scanning under a single acceleration factor (changing NEX), and did not involve the evaluation and analysis of the effects of different acceleration parameters on image quality. Relevant research is currently underway. Finally, this study did not conduct an in-depth analysis of the cost - effectiveness of the IQMR technology. Although the IQMR technology improves the image quality, its cost may be relatively high. Therefore, future studies need to comprehensively consider the cost-effectiveness ratio of the IQMR technology to evaluate its feasibility in clinical practice.\u003c/p\u003e\n\u003cp\u003eIn conclusion, the IQMR technique has significant advantages in improving the image quality of knee joint magnetic resonance scans. It can not only significantly improve key indicators such as image quality scores, SNR, and CNR, but also more accurately display the scope of lesions and reduce the interference of noise and artifacts. With these remarkable advantages, the IQMR technique shows broad application potential and important clinical practical value in the accurate diagnosis of knee joint diseases. In the future, we will continue to explore the application effects of the IQMR technique in images of other body parts and other sequences to expand its clinical application scope and provide better medical services for patients.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAuthors\u0026apos; declaration of conflicts of interest: All authors declare no conflicts of interest.\u003c/p\u003e\n\u003cp\u003eIn accordance with the Declaration of Helsinki, this research has been approved by the Ethics Committee of Tianjin Hospital. The ethics approval number is: 2025 MedEthics Review 118.\u003c/p\u003e\n\u003cp\u003eAll the participants in the study agreed to have their results published.\u003c/p\u003e\n\u003cp\u003eThis research did not receive any financial support from any institution.\u003c/p\u003e\n\u003ch2\u003eDeclaration of interests\u003c/h2\u003e\n\u003cp\u003e☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eHao Wang:Writing-Original-Draft,Writing-Review.\u0026amp;amp;\u0026middot;Editing.XinYi Guo:Investigation,Validation,Methodology.NaiQian Liu:Investigation,Methodology,Conceptualization.JingHong Wang:Investigation,Methodology.Yi Cao:Preparation, creation and/or presentation of the published workLin Guo:ProjectAdministration,Supervision,Review\u0026amp;amp;Editing.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe data are, however, available upon request and with the permission of Norwegian Social Research (NOVA)\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSukerkar PA, Doyle Z. Imaging of Osteoarthritis of the Knee, Radiologic Clinics of North America 60 (2022) 605\u0026ndash;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rcl.2022.03.004\u003c/span\u003e\u003cspan address=\"10.1016/j.rcl.2022.03.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEck BL, Yang M, Elias JJ, et al. Quantitative MRI for Evaluation of Musculoskeletal Disease: Cartilage and Muscle Composition, Joint Inflammation, and Biomechanics in Osteoarthritis. Invest Radiol. 2023;58:60\u0026ndash;75. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/RLI.0000000000000909\u003c/span\u003e\u003cspan address=\"10.1097/RLI.0000000000000909\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYoon MA, Gold GE, Chaudhari AS. Accelerated Musculoskeletal Magnetic Resonance Imaging. J Magn Reson Imaging. 2024;60:1806\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jmri.29205\u003c/span\u003e\u003cspan address=\"10.1002/jmri.29205\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSafari M, Eidex Z, Chang C-W et al. Fast MRI Reconstruction Using Deep Learning-based Compressed Sensing: A Systematic Review, ArXiv (2024) arXiv:2405.00241v1.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFeinberg DA, Hale JD, Watts JC, et al. Halving MR imaging time by conjugation: demonstration at 3.5 kG. Radiology. 1986;161:527\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1148/radiology.161.2.3763926\u003c/span\u003e\u003cspan address=\"10.1148/radiology.161.2.3763926\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChang Y, Pham HA, Li Z. A dual-interpolator method for improving parallel MRI Reconstruction. Magn Reson Imaging. 2022;92:108\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.mri.2022.06.009\u003c/span\u003e\u003cspan address=\"10.1016/j.mri.2022.06.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee Y, Yoon S, Paek M, et al. Advanced MRI techniques in abdominal imaging. Abdom Radiol (NY). 2024;49:3615\u0026ndash;36. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00261-024-04369-7\u003c/span\u003e\u003cspan address=\"10.1007/s00261-024-04369-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDieckmeyer M, Roy AG, Senapati J, et al. Effect of MRI acquisition acceleration via compressed sensing and parallel imaging on brain volumetry. MAGMA. 2021;34:487\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10334-020-00906-9\u003c/span\u003e\u003cspan address=\"10.1007/s10334-020-00906-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoori N, Yamamoto S, Kamekawa H, et al. Comparison of image quality evaluation methods for magnetic resonance imaging using compressed sensing-sensitivity encoding (CS-SENSE). Radiol Phys Technol. 2025;18:597\u0026ndash;605. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12194-025-00911-4\u003c/span\u003e\u003cspan address=\"10.1007/s12194-025-00911-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang S, Su Z, Ying L, ACCELERATING MAGNETIC RESONANCE IMAGING VIA DEEP LEARNING. Proc IEEE Int Symp Biomed Imaging,, 2016 (2016) 514\u0026ndash;517. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/ISBI.2016.7493320\u003c/span\u003e\u003cspan address=\"10.1109/ISBI.2016.7493320\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDo W-J, Seo S, Han Y, et al. Reconstruction of multicontrast MR images through deep learning. Med Phys. 2020;47:983\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/mp.14006\u003c/span\u003e\u003cspan address=\"10.1002/mp.14006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXie X, Wang J, Yan X. The Application Value of Intelligent Rapid MRI Technology in the Treatment of Ischial Tendon Injuries. Chin J Magn Reson Imaging. 2024.;15(10):148\u0026ndash;52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.12015/issn.1674-8034.2024.10.025\u003c/span\u003e\u003cspan address=\"10.12015/issn.1674-8034.2024.10.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXU M, WU Y, LIU J. Application value of intelligent quick magnetic resonance technique in magnetic resonance scanning of cervical vertebra[J]. Chin J Magn Reson Imaging. 2023;14(10):111\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.12015/issn.1674-8034.2023.10.019\u003c/span\u003e\u003cspan address=\"10.12015/issn.1674-8034.2023.10.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKanemaru N, Takao H, Amemiya S. The effect of a post-scan processing denoising system on image quality and morphometric analysis. J Neuroradiol. 2022;49:205\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neurad.2021.11.007\u003c/span\u003e\u003cspan address=\"10.1016/j.neurad.2021.11.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBratke G, Rau R, Weiss K, et al. Accelerated MRI of the Lumbar Spine Using Compressed Sensing: Quality and Efficiency. J Magn Reson Imaging. 2019;49:e164\u0026ndash;75. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jmri.26526\u003c/span\u003e\u003cspan address=\"10.1002/jmri.26526\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eN\u0026ouml;lte I, Gerigk L, Brockmann MA, et al. MRI of degenerative lumbar spine disease: comparison of non-accelerated and parallel imaging. Neuroradiology. 2008;50:403\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00234-008-0363-0\u003c/span\u003e\u003cspan address=\"10.1007/s00234-008-0363-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaeder Y, Dunet V, Richard R, et al. Bone Marrow Metastases: T2-weighted Dixon Spin-Echo Fat Images Can Replace T1-weighted Spin-Echo Images. Radiology. 2018;286:948\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1148/radiol.2017170325\u003c/span\u003e\u003cspan address=\"10.1148/radiol.2017170325\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUeda T, Ohno Y, Yamamoto K, et al. Deep Learning Reconstruction of Diffusion-weighted MRI Improves Image Quality for Prostatic Imaging. Radiology. 2022;303:373\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1148/radiol.204097\u003c/span\u003e\u003cspan address=\"10.1148/radiol.204097\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCho SJ, Choi YJ, Chung SR, et al. High-resolution MRI using compressed sensing-sensitivity encoding (CS-SENSE) for patients with suspected neurovascular compression syndrome: comparison with the conventional SENSE parallel acquisition technique. Clin Radiol. 2019;74:e8179\u0026ndash;81714. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.crad.2019.06.023\u003c/span\u003e\u003cspan address=\"10.1016/j.crad.2019.06.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePadole A, Digumarthy S, Flores E, et al. Assessment of chest CT at CTDIvol less than 1 mGy with iterative reconstruction techniques. Br J Radiol. 2017;90:20160625. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1259/bjr.20160625\u003c/span\u003e\u003cspan address=\"10.1259/bjr.20160625\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePadole A, Singh S, Ackman JB, et al. Submillisievert chest CT with filtered back projection and iterative reconstruction techniques. AJR Am J Roentgenol. 2014;203:772\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2214/AJR.13.12312\u003c/span\u003e\u003cspan address=\"10.2214/AJR.13.12312\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu X, Liu WV, Yan Y, et al. Evaluation of deep learning-based reconstruction late gadolinium enhancement images for identifying patients with clinically unrecognized myocardial infarction. BMC Med Imaging. 2024;24:127. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12880-024-01308-2\u003c/span\u003e\u003cspan address=\"10.1186/s12880-024-01308-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRen J, Eriksen JG, Nijkamp J, et al. Comparing different CT, PET and MRI multi-modality image combinations for deep learning-based head and neck tumor segmentation. Acta Oncol. 2021;60:1399\u0026ndash;406. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/0284186X.2021.1949034\u003c/span\u003e\u003cspan address=\"10.1080/0284186X.2021.1949034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEo T, Jun Y, Kim T, et al. KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images. Magn Reson Med. 2018;80:2188\u0026ndash;201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/mrm.27201\u003c/span\u003e\u003cspan address=\"10.1002/mrm.27201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYoo H, Yoo R-E, Choi SH, et al. Deep learning-based reconstruction for acceleration of lumbar spine MRI: a prospective comparison with standard MRI. Eur Radiol. 2023;33:8656\u0026ndash;68. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00330-023-09918-0\u003c/span\u003e\u003cspan address=\"10.1007/s00330-023-09918-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan L, Zhang F, Fan H, et al. Brief review of image denoising techniques. Vis Comput Ind Biomed Art. 2019;2:7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s42492-019-0016-7\u003c/span\u003e\u003cspan address=\"10.1186/s42492-019-0016-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen Q, Fang S, Yuchen Y, et al. Clinical feasibility of deep learning reconstruction in liver diffusion-weighted imaging: Improvement of image quality and impact on apparent diffusion coefficient value. Eur J Radiol. 2023;168:111149. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejrad.2023.111149\u003c/span\u003e\u003cspan address=\"10.1016/j.ejrad.2023.111149\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang R, Zou Y, Liu WV, et al. High-Resolution Single-Shot Fast Spin-Echo MR Imaging with Deep Learning Reconstruction Algorithm Can Improve Repeatability and Reproducibility of Follicle Counting. J Clin Med. 2023;12:3234. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/jcm12093234\u003c/span\u003e\u003cspan address=\"10.3390/jcm12093234\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang X, Wang Y, Xu X, et al. Bladder MRI with deep learning-based reconstruction: a prospective evaluation of muscle invasiveness using VI-RADS. Abdom Radiol (NY). 2024;49:1615\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00261-024-04280-1\u003c/span\u003e\u003cspan address=\"10.1007/s00261-024-04280-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGao Y, Liu WV, Li L, et al. Usefulness of T2-Weighted Images with Deep-Learning-Based Reconstruction in Nasal Cartilage. Diagnostics (Basel). 2023;13:3044. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/diagnostics13193044\u003c/span\u003e\u003cspan address=\"10.3390/diagnostics13193044\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSerai SD, Ho M-L, Artunduaga M, et al. Components of a magnetic resonance imaging system and their relationship to safety and image quality. Pediatr Radiol. 2021;51:716\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00247-020-04894-9\u003c/span\u003e\u003cspan address=\"10.1007/s00247-020-04894-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFeuerriegel GC, Sutter R. Managing hardware-related metal artifacts in MRI: current and evolving techniques. Skeletal Radiol. 2024;53:1737\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00256-024-04624-4\u003c/span\u003e\u003cspan address=\"10.1007/s00256-024-04624-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhan SU, Ullah N, Ahmed I, Imaging MRI, et al. Comparison of MRI with other Modalities, Noise in MRI Images and Machine Learning Techniques for Noise Removal: A Review. Curr Med Imaging Rev. 2019;15:243\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/1573405614666180726124952\u003c/span\u003e\u003cspan address=\"10.2174/1573405614666180726124952\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNoordman CR, Yakar D, Bosma J, et al. Complexities of deep learning-based undersampled MR image reconstruction. Eur Radiol Exp. 2023;7:58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s41747-023-00372-7\u003c/span\u003e\u003cspan address=\"10.1186/s41747-023-00372-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Z, Wen B, Wang Z, et al. Deep learning-based reconstruction enhances image quality and improves diagnosis in magnetic resonance imaging of the shoulder joint. Quant Imaging Med Surg. 2024;14:2840\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21037/qims-23-1412\u003c/span\u003e\u003cspan address=\"10.21037/qims-23-1412\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXie Y, Tao H, Li X, et al. Prospective Comparison of Standard and Deep Learning-reconstructed Turbo Spin-Echo MRI of the Shoulder. Radiology. 2024;310:e231405. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1148/radiol.231405\u003c/span\u003e\u003cspan address=\"10.1148/radiol.231405\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Y, Zhang X, Hu M. The application value of deep learning reconstruction technology in optimizing the scanning time and image quality of T2-weighted MRI scans for prostate imaging. J Magn Reson Imaging 202314(05):48\u0026ndash;52.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.12015/issn.1674-8034.2\u003c/span\u003e\u003cspan address=\"10.12015/issn.1674-8034.2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e05.010.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Intelligent Quick Magnetic Resonance technology, Number of excitations, Knee joint, Magnetic resonance scan, Image quality","lastPublishedDoi":"10.21203/rs.3.rs-8199495/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8199495/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective:\u003c/strong\u003e To explore the application value of Intelligent Quick Magnetic Resonance (IQMR) in improving the image quality of knee joint scans.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eMethods:\u003c/strong\u003e A total of 100 patients suspected of having knee joint lesions underwent conventional and accelerated knee joint magnetic resonance examinations on a GE Discovery 750 3.0T MRI scanner. The sagittal PDWI-FS sequence of the knee joint was selected, and the fast PDWI-FS sequence (Fast-MR, NEX=1) and the original PDWI-FS sequence (Original-MR, NEX=2) were scanned respectively. The two groups of sequences were automatically transferred to the IQMR post-processing system to reconstruct Fast-IQMR and Original-IQMR images. Three radiologists independently gave subjective scores on a five-point scale for the lesion details, anatomical structure clarity, overall image artifacts, and overall image quality of the four groups of images in the PD-FS sequence. The signal values of the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, cartilage, and fat, as well as the background noise intensity were measured. With the muscle as the background, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of each tissue were calculated, and the above indicators of the four groups of images were compared and analyzed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eThe Fast-IQMR sequence reduced the average examination time by 49% compared with the Original-IQMR sequence. By Kendall's W test, in the four groups of images, the three subjective scores given by the three physicians showed significant consistency (\u003cem\u003ep\u003c/em\u003e\u0026lt;0.05). There were statistical differences in the three subjective scores among the four groups of images (\u003cem\u003ep\u003c/em\u003e\u0026lt;0.001). The SNR and CNR of the Fast-IQMR and Original-IQMR images in the tibial plateau, anterior cruciate ligament, posterior cruciate ligament, and cartilage were significantly higher than those of the Fast - MR and Original - MR images respectively (p \u0026lt; 0.001), and the SNR and CNR of the Fast - IQMR images were significantly higher than those of the Original - MR group images (p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eThe images reconstructed by IQMR have certain improvements in both scanning speed and image quality compared with conventional images. The IQMR technology has significant advantages in improving the image quality of knee joint magnetic resonance scans, which can significantly improve the detail expression and clarity of the images, increase the SNR and CNR of the images, and more accurately show the scope of the lesions. Compared with the Original - MR images, the improvement effect of the image quality of the Fast - IQMR images is significant. That is, the Fast - IQMR sequence with NEX = 1 can not only ensure the image quality but also improve the scanning efficiency.\u003c/p\u003e","manuscriptTitle":"Clinical study on the application of intelligent and rapid magnetic resonance technology in improving the quality of knee joint scan images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-15 10:59:17","doi":"10.21203/rs.3.rs-8199495/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"964c5214-ea26-49ed-90d7-ae4d27cd059a","owner":[],"postedDate":"December 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-29T10:24:32+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-15 10:59:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8199495","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8199495","identity":"rs-8199495","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}