Deep learning-based magnetic resonance imaging reconstruction for improving the image quality of reduced-field-of-view diffusionweighted imaging of the pancreas

BACKGROUND It has been reported that deep learning-based reconstruction(DLR)can reduce image noise and artifacts,thereby improving the signal-to-noise ratio and image sharpness.However,no previous studies have evaluated the efficacy of DLR in improving image quality in reduced-field-of-view(reduced-FOV)diffusionweighted imaging(DWI)[field-of-view optimized and constrained undistorted single-shot(FOCUS)]of the pancreas.We hypothesized that a combination of these techniques would improve DWI image quality without prolonging the scan time but would influence the apparent diffusion coefficient calculation.AIM To evaluate the efficacy of DLR for image quality improvement of FOCUS of the pancreas.METHODS This was a retrospective study evaluated 37 patients with pancreatic cystic lesions who underwent magnetic resonance imaging between August 2021 and October 2021.We evaluated three types of FOCUS examinations:FOCUS with DLR(FOCUS-DLR+),FOCUS without DLR(FOCUS-DLR−),and conventional FOCUS(FOCUS-conv).The three types of FOCUS and their apparent diffusion coefficient(ADC)maps were compared qualitatively and quantitatively.RESULTS FOCUS-DLR+(3.62,average score of two radiologists)showed significantly better qualitative scores for image noise than FOCUS-DLR−(2.62)and FOCUS-conv(2.88)(P<0.05).Furthermore,FOCUS-DLR+showed the highest contrast ratio and 600 s/mm^(2)(0.72±0.08 and 0.68±0.08)and FOCUS-DLR−showed the highest CR between cystic lesions and the pancreatic parenchyma for the b-values of 0 and 600 s/mm2(0.62±0.21 and 0.62±0.21)(P<0.05),respectively.FOCUS-DLR+provided significantly higher ADCs of the pancreas and lesion(1.44±0.24 and 3.00±0.66)compared to FOCUS-DLR−(1.39±0.22 and 2.86±0.61)and significantly lower ADCs compared to FOCUS-conv(1.84±0.45 and 3.32±0.70)(P<0.05),respectively.CONCLUSION This study evaluated the efficacy of DLR for image quality improvement in reduced-FOV DWI of the pancreas.DLR can significantly denoise images without prolonging the scan time or decreasing the spatial resolution.The denoising level of DWI can be controlled to make the images appear more natural to the human eye.However,this study revealed that DLR did not ameliorate pancreatic distortion.Additionally,physicians should pay attention to the interpretation of ADCs after DLR application because ADCs are significantly changed by DLR.

Three-dimensional reconstructed magnetic resonance scans:Accuracy in identifying and defining knee meniscal tears

AIM To determine whether three-dimensional(3D) reconstruction from conventional magnetic resonance imaging(MRI) is able to accurately detect a meniscal tear, and define the configuration.METHODS Thirty-three patients' 3T MRI scan data were collected and sagittal uni-planar 3D reconstructions performed from the preoperative MRI. There were 24 meniscal tears in 24 patients, and nine controls. All patients had arthroscopic corroboration of MRI findings. Two independent observers prospectively reported on all 33 reconstructions. Meniscal tear presence or absence was noted, and tear configuration subsequently categorised as either radial, bucket-handle, parrot beak, horizontal or complex.RESULTS Identification of control menisci or meniscal tear presence was excellent(Accuracy: observer 1 = 90.9%; observer 2 = 81.8%). Of the tear configurations, bucket handle tears were accurately identified(Accuracy observer 1 and 2 = 80%). The remaining tear configurations were notaccurately discernable.CONCLUSION Uni-planar 3D reconstruction from 3T MRI knee scan sequences are useful in identifying normal menisci and menisci with bucket-handle tears. Advances in MRI sequencing and reconstruction software are awaited for accurate identification of the remaining meniscal tear configurations.

Magnetic resonance imaging with three-dimensional fast imaging employing steady-state acquisition with phase-cycled and short T1 inversion recovery pulse sequence for evaluating brachial plexus injury

There is a large amount of fat in the postganglionic segment of the brachial plexus nerve.The use of short T1 inversion recovery pulse sequence may improve signal strength of the brachial plexus postganglionic segment.The present study revealed that the combination of three-dimensional fast imaging employing steady-state acquisition with phase-cycled and short T1 inversion recovery pulse sequence clearly displayed the anatomical morphology and structure of the brachial plexus nerve,together with maximum intensity projection,volume rendering and other three-dimensional reconstruction techniques.Our results suggested that this method is also suitable for providing accurate assessment and diagnosis of the site,severity and scope of brachial plexus injury.

Magnetic resonance imaging after anterior cruciate ligament reconstruction:A practical guide

Anterior cruciate ligament(ACL) reconstruction is one of the most common orthopedic procedures performed worldwide. In this regard, magnetic resonance imaging(MRI) represents a useful pre-operative tool to confirm a disruption of the ACL and to assess for potential associated injuries. However, MRI is also valuable postoperatively, as it is able to identify, in a non-invasive way, a number of aspects and situations that could suggest potential problems to clinicians. Graft signal and integrity, correct tunnel placement, tunnel widening, and problems with fixation devices or the donor site could all compromise the surgical outcomes and potentially predict the failure of the ACL reconstruction. Furthermore, several anatomical features of the knee could be associated to worst outcomes or higher risk of failure. This review provides a practical guide for the clinician to evaluate the post-surgical ACL through MRI, and to analyze all the parameters and features directly or indirectly related to ACL reconstruction, in order to assess for normal or pathologic conditions.

Application of Three-Dimensional Magnetic Resonance Imaging in the Diagnosis of Perianal Abscess

Perianal abscess is a common disease in anorectal surgery. If the diagnosis is not clear and the cure is thoroughly cleared, the recurrence and spread of anal fistula will cause life-long pain. Objective: To investigate the application of 3.0T MRI 3D CUBE T2WI lipid suppression sequence in the diagnosis of perianal abscess. Methods: Thirty-six patients with perianal abscess confirmed by operation were examined with 2D T2WI and 3D CUBE T2WI lipid suppression sequences before operation. Two imaging techniques were evaluated to show the types of perianal abscess, the number of abscesses, the number of internal orifices of abscess, and the number of fistula branches with anal fistula in abscess. Results: Among 36 cases of perianal abscess, there were 5 cases of anal subcutaneous abscess, 12 cases of ischiorectal space abscess (8 cases complicated with anal fistula), 6 cases of posterior anal space abscess, 5 cases of anal sphincter abscess (3 cases complicated with anal fistula), 2 cases of high intermuscular abscess, 2 cases of rectal submucosal abscess, 3 cases of complex abscess (3 cases complicated with anal fistula), 1 case of misdiagnosis, 2D T2WI lipid suppression sequence and 3D CUBE T2WI suppression. The accuracy of lipid sequence abscess typing was 80.6% (29/36) and 88.9% (32/36), respectively, with no significant difference (P > 0.05). Thirty-six patients were surgically diagnosed as having 32 internal orifices, 68.8% (22/32) and 93.8% (30/32) of 2D T2WI and 3D CUBE T2WI lipid-suppressing sequences, respectively, with significant difference (P Conclusion: 3D CUBE T2WI lipid suppression sequence is superior to 2D T2WI lipid suppression sequence in the classification of perianal abscess, the number of internal orifices of abscess and the number of fistula branches of abscess complicated with anal fistula. It can also determine the number of internal orifices of abscess complicated with anal fistula, the number of fistula branches, the shape of primary and branch fistula and the relationship among pelvic floor muscle tissues. It can provide more accurate images for preoperative and intraoperative clinical surgery.

Magnetic resonance imaging-three-dimensional printing technology fabricates customized scaffolds for brain tissue engineering

Conventional fabrication methods lack the ability to control both macro- and micro-structures of generated scaffolds. Three-dimensional printing is a solid free-form fabrication method that provides novel ways to create customized scaffolds with high precision and accuracy. In this study, an electrically controlled cortical impactor was used to induce randomized brain tissue defects. The overall shape of scaffolds was designed using rat-specific anatomical data obtained from magnetic resonance imaging, and the internal structure was created by computer- aided design. As the result of limitations arising from insufficient resolution of the manufacturing process, we magnified the size of the cavity model prototype five-fold to successfully fabricate customized collagen-chitosan scaffolds using three-dimensional printing. Results demonstrated that scaffolds have three-dimensional porous structures, high porosity, highly specific surface areas, pore connectivity and good internal characteristics. Neural stem cells co-cultured with scaffolds showed good viability, indicating good biocompatibility and biodegradability. This technique may be a promising new strategy for regenerating complex damaged brain tissues, and helps pave the way toward personalized medicine.

Three-dimensional time-of-flight magnetic resonance angiography combined with high resolution T2-weighted imaging in preoperative evaluation of microvascular decompression

BACKGROUND Neurovascular compression(NVC) is the main cause of primary trigeminal neuralgia(TN) and hemifacial spasm(HFS). Microvascular decompression(MVD) is an effective surgical method for the treatment of TN and HFS caused by NVC. The judgement of NVC is a critical step in the preoperative evaluation of MVD, which is related to the effect of MVD treatment. Magnetic resonance imaging(MRI) technology has been used to detect NVC prior to MVD for several years. Among many MRI sequences, three-dimensional time-of-flight magnetic resonance angiography(3D TOF MRA) is the most widely used. However, 3D TOF MRA has some shortcomings in detecting NVC. Therefore, 3D TOF MRA combined with high resolution T2-weighted imaging(HR T2WI) is considered to be a more effective method to detect NVC.AIM To determine the value of 3D TOF MRA combined with HR T2WI in the judgment of NVC, and thus to assess its value in the preoperative evaluation of MVD.METHODS Related studies published from inception to September 2022 based on PubMed, Embase, Web of Science, and the Cochrane Library were retrieved. Studies that investigated 3D TOF MRA combined with HR T2WI to judge NVC in patients with TN or HFS were included according to the inclusion criteria. Studies without complete data or not relevant to the research topics were excluded. The Quality Assessment of Diagnostic Accuracy Studies checklist was used to assess the quality of included studies. The publication bias of the included literature was examined by Deeks’ test. An exact binomial rendition of the bivariate mixed-effects regression model was used to synthesize data. Data analysis was performed using the MIDAS module of statistical software Stata 16.0. Two independent investigators extracted patient and study characteristics, and discrepancies were resolved by consensus. Individual and pooled sensitivities and specificities were calculated. The I_(2) statistic and Q test were used to test heterogeneity. The study was registered on the website of PROSERO(registration No. CRD42022357158).RESULTS Our search identified 595 articles, of which 12(including 855 patients) fulfilled the inclusion criteria. Bivariate analysis showed that the pooled sensitivity and specificity of 3D TOF MRA combined with HR T2WI for detecting NVC were 0.96 [95% confidence interval(CI): 0.92-0.98] and 0.92(95%CI: 0.74-0.98), respectively. The pooled positive likelihood ratio was 12.4(95%CI: 3.2-47.8), pooled negative likelihood ratio was 0.04(95%CI: 0.02-0.09), and pooled diagnostic odds ratio was 283(95%CI: 50-1620). The area under the receiver operating characteristic curve was 0.98(95%CI: 0.97-0.99). The studies showed no substantial heterogeneity(I2 = 0, Q = 0.001 P = 0.50).CONCLUSION Our results suggest that 3D TOF MRA combined with HR T2WI has excellent sensitivity and specificity for judging NVC in patients with TN or HFS. This method can be used as an effective tool for preoperative evaluation of MVD.

Non-iterative image reconstruction from sparse magnetic resonance imaging radial data without priors

The state-of-the-art approaches for image reconstruction using under-sampled k-space data are compressed sensing based.They are iterative algorithms that optimize objective functions with spatial and/or temporal constraints.This paper proposes a non-iterative algorithm to estimate the un-measured data and then to reconstruct the image with the efficient filtered backprojection algorithm.The feasibility of the proposed method is demonstrated with a patient magnetic resonance imaging study.The proposed method is also compared with the state-of-the-art iterative compressed-sensing image reconstruction method using the total-variation optimization norm.

Reduced Imaging Time and Improved Image Quality of 3D Isotropic T2-Weighted Magnetic Resonance Imaging with Compressed Sensing for the Female Pelvis

This study is to compare three-dimensional(3D)isotropic T2-weighted magnetic resonance imaging(MRI)with compressed sensing-sampling perfection with application optimized contrast(CS-SPACE)and the conventional image(3D-SPACE)sequence in terms of image quality,estimated signal-to-noise ratio(SNR),relative contrast-to-noise ratio(CNR),and the lesions’conspicuous of the female pelvis.Thirty-six females(age:51,28-73)with cervical carcinoma(n=20),rectal carcinoma(n=7),or uterine fibroid(n=9)were included.Patients underwent magnetic resonance(MR)imaging at a 3T scanner with the sequences of 3D-SPACE,CS-SPACE,and twodimensional(2D)T2-weighted turbo-spin echo(TSE).Quantitative analyses of estimated SNR and relative CNR between tumors and other tissues,image quality,and tissue conspicuity were performed.Two radiologists assessed the difference in diagnostic findings for carcinoma.Quantitative values and qualitative scores were analyzed,respectively.The estimated SNR and the relative CNR of tumor-to-muscle obturator internus,tumor-to-myometrium,and myometrium-to-muscle obturator internus was comparable between 3D-SPACE and CS-SPACE.The overall image quality and the conspicuity of the lesion scores of the CS-SPACE were higher than that of the 3D-SPACE(P<0.01).The CS-SPACE sequence offers shorter scan time,fewer artifacts,and comparable SNR and CNR to conventional 3D-SPACE,and has the potential to improve the performance of T2-weighted images.

Linearized Proximal Alternating Direction Method of Multipliers for Parallel Magnetic Resonance Imaging

In this study, we propose a linearized proximal alternating direction method with variable stepsize for solving total variation image reconstruction problems. Our method uses a linearized technique and the proximal function such that the closed form solutions of the subproblem can be easily derived.In the subproblem, we apply a variable stepsize, that is like Barzilai-Borwein stepsize, to accelerate the algorithm. Numerical results with parallel magnetic resonance imaging demonstrate the efficiency of the proposed algorithm.

Cerebral blood flow volume measurements of the carotid artery and ipsilateral branches using two-dimensional phase-contrast magnetic resonance angiography

The optimal velocity encoding of phase-contrast magnetic resonance angiography （PC MRA） in measuring cerebral blood flow volume （BFV） ranges from 60 to 80 cm/s. To verify the accuracy of two-dimensional （2D） PC MRA, the present study localized the region of interest at blood vessels of the neck using PC MRA based on three-dimensional time-of-flight sequences, and the velocity encoding was set to 80 cm/s. Results of the measurements showed that the error rate was 7.0±6.0% in the estimation of BFV in the internal carotid artery, the external carotid artery and the ipsilateral common carotid artery. There was no significant difference, and a significant correlation in BFV between internal carotid artery ＋ external carotid artery and ipsilateral common carotid artery. In addition, the BFV of the common carotid artery was correlated with that of the ipsilateral internal carotid artery. The main error was attributed to the external carotid artery and its branches. Therefore, after selecting the appropriate scanning parameters and protocols, 2D PC MRA is more accurate in the determination of BFV in the carotid arteries.

Initial experience with circumferential pulmonary vein ablation guided by fusion of magnetic resonance imaging with three-dimensional electroanatomic mapping

Catheter ablation for the treatment of atrial fibrillation （AF） has been a focal target ofelectrophysiological study in recent years. Up to date, circumferential pulmonary vein ablation （CPVA） guided by three-dimensional （3-D） electreanatomic mapping （Carto, USA） has been one of the most favourable procedures for the treatment of AF. However, it is still difficult to acquire the detailed information on number, location, and branching pattern of all pulmonary veins （PVs） when the 3-D electroanatomic mapping system is used alone.

Magnetic Resonance Three-dimensional Cube Technique in the Measurement of Piglet Femoral Anteversion

Background： The accurate measurement of the femoral anteversion （FA） angle is always a topic of much debate in the orthopedic surgery and radiology research. We aimed to explore a new FA measurement method to acquire accurate results without radiation damage using piglet model. Methods： A total of thirty piglets were assigned to two groups based on the age. Bilateral femora were imaged with 3.0-T magnetic resonance （MR） and 64-slice computed tomography （CT） examinations on all piglets. FA was measured on MR-three-dimensional （3D） postprocessing software with a four-step method： initial validation of the femoral condylar axis, validation of the condylar plane, validation of the femoral neck axis, and line-plane angle measurement of FA. After MR and CT examinations, all piglets were sacrificed and their degree of FA was measured using their excised, dried femora. MR, CT, and dried-femur measurement results were analyzed statistically; M R and CT measurements were compared for accuracy against each other and against the gold standard dried femur measurement. Results： In both groups, the mean FA value measured by MR was lower than that measured by CT. A statistically significant difference was observed between CT- and dried-femur measurements but not between MR- and dried-femur measurements. A higher correlation （0.783 vs. 0.408） and a higher consistency （0.863 vs. 0.578） with dried-femur measurement results were seen for MR measurements than CT measurements in the 1 -week age group. However, in the 8-week age group, similar correlations （0.707 vs. 0.669） and consistencies （0.864 vs. 0.82 l ） were observed. Conclusions： Noninvasive MR-3D-Cube reconstruction was able to accurately measure FA in piglets. Particularly in the 1-week age group with a larger proportion of cartilaginous structures, the correlation and consistency between MR- and dried-femur measurement results were higher than those between CT- and dried-femur measurements, suggesting that MR may be a new useful examination tool for FA-related diseases in children.

A three-dimensional digital visualization model of cervical nerves in a healthy person

Three-dimensional reconstruction nerve models are classically obtained from two-dimensional images of ＂visible human＂ frozen sections. However, because of the flexibility of nerve tissues and small color differences compared with surrounding tissues, the integrity and validity of nerve tissues can be impaired during milling. Thus, in the present study, we obtained two-dimensional data from a healthy volunteer based on continuous CT angiography and magnetic resonance myelography. Semi-automatic segmentation and reconstruction were then conducted at different thresholds in different tissues using Mimics software. Small anatomical structures such as muscles and cervical nerves were reconstructed using the medical computer aided design module. Three-dimensional digital models of the cervical nerves and their surrounding structures were successfully developed, which allowed visualization of the spatial relation of anatomical structures with a strong three-dimensional effect, distinct appearance, clear distribution, and good continuity, precision, and integrality. These results indicate the validity of a three-dimensional digital visualization model of healthy human cervical nerves, which overcomes the disadvantages of milling, avoids data loss, and exhibits a realistic appearance and three-dimensional image.

A three-dimensional matrix system containing melatonin and neural stem cells repairs damage from traumatic brain injury in rats

Brain lesions can cause neural stem cells to activate,proliferate,diffe rentiate,and migrate to the injured area.However,after traumatic brain injury,brain tissue defects and microenvironment changes greatly affect the survival and growth of neural stem cells;the resulting reduction in the number of neural stem cells impedes effective repair of the injured area.Melatonin can promote the survival,proliferation,and differentiation of neural stem cells under adverse conditions such as oxidative stress or hypoxia that can occur after traumatic brain injury.Therefore,we investigated the therapeutic effects of melatonin combined with neural stem cells on traumatic brain injury in rats.First,in vitro studies confirmed that melatonin promoted the survival of neural stem cells deprived of oxygen and glucose.Then,we established a three-dimensional Matrigel-based transplantation system containing melatonin and neural stem cells and then used it to treat traumatic brain injury in rats.We found that treatment with the Matrigel system containing melatonin and neural stem cells decreased brain lesion volume,increased the number of surviving neuro ns,and improved recove ry of neurological function compared with treatment with Matrigel alone,neural stem cells alone,Matrigel and neural stem cells combined,and Matrigel and melatonin combined.Our findings suggest that the three-dimensional Matrigelbased transplantation system containing melatonin and neural stem cells is a potential treatment for traumatic brain injury.

TV Sparsifying MR Image Reconstruction in Compressive Sensing

In this paper, we apply alternating minimization method to sparse image reconstruction in compressed sensing. This approach can exactly reconstruct the MR image from under-sampled k-space data, i.e., the partial Fourier data. The convergence analysis of the fast method is also given. Some MR images are employed to test in the numerical experi-ments, and the results demonstrate that our method is very efficient in MRI reconstruction.

Current status of deep learning in abdominal image reconstruction

Abdominal magnetic resonance imaging(MRI)and computed tomography(CT)are commonly used for disease screening,diagnosis,and treatment guidance.However,abdominal MRI has disadvantages including slow speed and vulnerability to motions,while CT suffers from problems of radiation.It has been reported that deep learning reconstruction can solve such problems while maintaining good image quality.Recently,deep learning-based image reconstruction has become a hot topic in the field of medical imaging.This study reviews the latest research on deep learning reconstruction in abdominal imaging,including the widely used convolutional neural network,generative adversarial network,and recurrent neural network.

MRI深度学习图像重建技术在肌骨系统疾病诊断的应用进展

深度学习图像重建(deep learning-based reconstruction,DLR)技术是目前MRI图像重建领域最为前沿的技术进展之一。相对于常规MRI图像重建技术而言,DLR技术重新定义了MRI的信噪比、空间分辨率和扫描时间之间新的边界,其突出的技术优势是有效去除图像噪声及伪影,大幅缩短扫描时间,且在提高病灶的检出率和定性准确率方面也具有潜在优势。随着算法的不断优化和模型泛化性的提升,DLR目前已被广泛应用于神经系统、肌骨系统及心脏等多部位的MRI检查,其适用的扫描序列及临床应用场景也在不断拓展。DLR技术在维持原有空间分辨率条件下,通过减少信号采集次数联合增加并行采集加速因子,将成像时间缩短50%以上,实现肌骨系统快速成像,且所获得图像质量明显优于传统重建图像。目前,DLR在膝关节、肩关节、手腕关节及脊柱等肌骨系统的MRI检查中被广泛应用,并证实了其在缩短成像时间、提升图像信噪比和提高分辨率方面具有卓越表现。

前交叉韧带重建术后1年腱骨愈合的MRI纵向随访30例分析

目的:探索采用腘绳肌腱前交叉韧带重建术(ACLR)后1年内腱骨愈合的磁共振成像(MRI)信号变化规律,并探讨腱骨愈合情况与临床主观功能评分的关系。方法:回顾性分析30例男性患者于2019年1月至2020年12月在深圳市第二人民医院采用腘绳肌腱行ACLR术后的3、6、12个月不同时间节点MRI上腱骨结合部的信号变化;其中腱骨愈合程度以胫骨段及股骨段骨道内移植物的信噪比(SNQ)表示。采用重复测量方差分析比较不同时间节点的骨道内移植物SNQ值的差异,采用t检验比较同一时间节点胫骨、股骨骨道内移植物的SNQ值的差异。结果:在术后3、6、12个月,股骨段及胫骨段移植物的SNQ值在3个月开始逐渐升高,6个月达峰,随后逐渐下降;股骨段内移植物的SNQ值在3、6、12个月之间的差异具有统计学意义(F=6.88,P<0.05);胫骨段移植物的SNQ值在3、6、12个月不同时间节点组间差异无统计学意义(F=2.87,P>0.05)。在术后1年内,同一时间节点,股骨段移植物的SNQ值显著高于胫骨段,差异具有统计学意义(P<0.05)。结论:前交叉韧带腱骨结合部信号在6月达峰,随后逐渐下降,股骨段愈合差于胫骨段。

基于深度学习重建技术的头部增强T1WI序列在垂体神经内分泌肿瘤病变成像中的应用

目的对比分析基于深度学习重建(deep learning reconstruction,DL Recon)技术的头部T1WI增强序列与常规T1WI增强序列对垂体神经内分泌肿瘤病变的成像质量。材料与方法前瞻性纳入行头部MRI增强扫描的50例垂体神经内分泌肿瘤患者,均在注射对比剂后行定制的T1WI(试验组)及常规T1WI(对照组)轴位扫描,其中试验组出两组图像,经DL Recon处理的图像记作A组,未经DL处理的原始图像记作B组,对照组记作C组,对比分析各组图像在灰质、白质和病灶区域的信噪比(signal to noise ratio,SNR)及对比噪声比(contrast to noise ratio,CNR),并由两位诊断医师分析图像总体质量和诊断置信度。结果试验组T1WI扫描时间(42 s)比传统T1WI扫描时间短(76 s)。A组的SNR灰质、SNR白质和SNR病灶显著高于B组和C组(P<0.001);A组的CNR灰质/白质以及CNR病灶/白质均高于B组和C组(P<0.001);A组的图像总体质量评分(5 vs.3和4)显著高于B组和C组(P<0.001),但是诊断置信度无显著差异(P>0.05)。结论垂体神经内分泌肿瘤成像时,基于DL重建技术的头部T1WI增强序列相比于常规T1WI增强序列在缩短了扫描时间的情况下具有更好的图像质量和同等的诊断置信度。