Preoperative Magnetic Resonance Imaging Research Articles

Trochlear Dysplasia Is Associated With Increased Sagittal Tibial Tubercle Trochlear-Groove Distance in Patients With Patellar Instability

To compare the sagittal position of the tibial tubercle in relation the trochlea groove in patients with and without trochlear dysplasia (TD). Patients with high-grade TD show a significantly increased sagittal position of the tibial tubercle in relation to the trochlear groove (sTTTG) compared with patients without TD. This may affect patellofemoral loading and contribute to the increased prevalence of cartilage lesions seen in the patellofemoral joint of patients with dysplasia of the trochlear groove. All patients between January 2017 and December 2020 with high-grade TD (Dejour type B, C, and D) who underwent patellar-stabilizing surgery for patellar instability at a single institution were included in the current study. Patients without preoperative magnetic resonance imaging (MRI), any previous osteotomy on the affected lower extremity, or cruciate ligament insufficiency were excluded. Patients who underwent knee arthroscopy for meniscal repair/debridement without any signs of TD or any of the aforementioned criteria served as the control group. Preoperative MRI was retrospectively assessed to compare common patellofemoral anatomic parameters including patellar angle, patellar tilt, patella morphology according to Wiberg, Caton-Deschamps index, PF index, trochlear sulcus angle, sulcus depth, lateral inclination angle of the trochlea, tibiofemoral rotation, TTTG, and sTTTG distance between both groups. The sTTTG is measured as the distance between the nadir point of the cartilaginous trochlear groove and the most anterior point of the tibial tubercle on an axial MRI. Independent predictors for the sTTTG were assessed for patients with TD. Patients with high-grade TD (n= 82) showed an increased patellar tilt, Caton-Deschamps index, trochlear sulcus angle, lateral tibiofemoral rotation angle, TTTG, and sTTTG (9.16 ± 4.47 mm vs 2.66 ± 4.21 mm) compared with the control group (n=83) (P < .001). Patellar angle, PF index, sulcus depth, and lateral inclination angle of the trochlear were significantly decreased in the TD group (P < .001). The sTTTG was similar in all TD groups (n.s.). Among patients with TD, both tibiofemoral rotation and patellar height were independent predictors of the sTTTG (P < .05). Patients with high-grade TD show not only abnormal values in common patellofemoral instability risk factors but also a significantly increased sTTTG compared with patients without TD. Level III, retrospective case comparative study.

Tibial-sided Anterolateral Ligament Injuries Are Associated With Poor Healing and Residual Pivot Shift Testing After Anterior Cruciate Ligament Reconstruction

Purpose; to investigate if tibial sided anterolateral ligament (ALL) injuries are associated with poor healing and residual pivot shift. Methods; Patients who underwent anterior cruciate ligament reconstructions within postinjury 6 weeks from January 2008 to March 2021 were included. They had concomitant ALL injury confirmed by preoperative magnetic resonance imaging (MRI) and were followed minimum 20 months. Preoperative ALL injury was graded by the Muramatsu classification and postoperative ALL healing were graded by a modification of the Lee classification (good, partial, and non-visualized). The patients were allocated to an improved group (IG) and an unimproved group (UG) based on a comparison of pre- and postoperative MRI. ALL tear site, postoperative knee stability, Lysholm score, and Tegner activity scale (PROs) were compared between the two groups. Results; 128 patients were enrolled. 94.5 % patients achieved the minimal clinically important difference (MCID) for PROs, respectively. The ALL was torn at the femoral side in 46.9% patients, at mid-substance in 31.2.%, and at tibial side in 21.9%, Preoperatively, 86 (67.2%) patients had a partial tear and 42 (32.8%) patients had a complete tear. Based on the postoperative MRI appearances, 38 (29.7%) and 90 (70.3%) patients were allocated to the IG and UG, respectively. The tibial side tears were significantly frequent in UG (p= .032). Pivot shift showed a significantly higher incidence in UG than IG (p=.004). Lachman test and PROs did not differ between the two groups. Conclusions; Tibial-sided ALL tears occurred in 18.7% of the UG compared to 3.2% of the IG, and 45.6% of the UG had a postoperative grade 1 or 2 pivot shift, compared to 15.8% of the IG. Percentages of patients with the MCID for PROs did not differ between the two groups. However, the clinical relevance is limited by a high transfer bias.

Surgical treatment of cystic pituitary adenomas: literature-based definitions and postoperative outcomes.

To survey the applied definitions of 'cystic' among pituitary adenomas and evaluate whether postoperative outcomes differ relative to non-cystic counterparts. A literature search and meta-analysis was performed using PRISMA guidelines. Studies were eligible if novel data were reported regarding the applied definition of 'cystic' and postoperative outcomes among cases of surgically treated pituitary adenomas. Data were pooled with random effects meta-analysis models into cohorts based on the applied definition of 'cystic'. Categorical meta-regressions were used to investigate differences between cohorts. Among studies comparing cystic and non-cystic pituitary adenomas, meta-analysis models were applied to determine the Odds Ratio [95% Confidence Interval]. Statistical analyses were performed using Comprehensive Meta-Analysis (CMA, 4.0), with a priori significance defined as P < 0.05. Ten studies were eligible yielding 283 patients with cystic pituitary adenomas. The definitions of 'cystic' mainly varied between the visual appearance of cystic components on preoperative magnetic resonance imaging and a volumetric definition requiring 50% or greater of tumor volume exhibiting cystic components. Tumor diameter was seldom reported with an associated standard deviation/error, limiting meta-analyses. When the data were pooled in accordance with the definition applied, there were no significant differences in the rates of gross total resection (P = 0.830), endocrinologic remission (P = 0.563), and tumor recurrence (P = 0.320). Meta-analyses on studies comparing cystic versus non-cystic pituitary adenomas indicated no significant difference in the rates of gross total resection (P = 0.729), endocrinologic remission (P = 0.857), and tumor recurrence (P = 0.465). Despite some individual studies describing a significant influence of pituitary adenoma texture on postoperative outcomes, meta-analyses revealed no such differences between cystic and non-cystic pituitary adenomas. This discrepancy may be explained in part by the inconsistent definition of 'cystic' and between-group differences in tumor size. A notion of a field-standard definition of 'cystic' among pituitary adenomas should be established to facilitate inter-study comparisons.

Delta-radiomics Analysis Based on Multi-phase Contrast-enhanced MRI to Predict Early Recurrence in Hepatocellular Carcinoma After Percutaneous Thermal Ablation

Rationale and ObjectivesIt is critical to predict early recurrence (ER) after percutaneous thermal ablation (PTA) for hepatocellular carcinoma (HCC). We aimed to develop and validate a delta-radiomics nomogram based on multi-phase contrast-enhanced magnetic resonance imaging (MRI) to preoperatively predict ER of HCC after PTA. Materials and MethodsWe retrospectively enrolled 164 patients with HCC and divided them into training, temporal validation, and other-scanner validation cohorts (n=110, 29, and 25, respectively). The volumes of interest of the intratumoral and/or peritumoral regions were delineated on preoperative multi-phase MR images. Original radiomics features were extracted from each phase, and delta-radiomics features were calculated. Logistic regression was used to train the corresponding radiomics models. The clinical and radiological characteristics were evaluated and combined to establish a clinical-radiological model. A fusion model comprising the best radiomics scores and clinical-radiological risk factors was constructed and presented as a nomogram. The performance of each model was evaluated and recurrence-free survival (RFS) was assessed. ResultsChild-Pugh grade B, high-risk tumor location, and an incomplete/absent tumor capsule were independent predictors of ER. The optimal radiomics model comprised 12 delta-radiomics features with areas under the curve (AUCs) of 0.834, 0.795, and 0.769 in the training, temporal validation, and other-scanner validation cohorts, respectively. The nomogram showed the best predictive performance with AUCs as 0.893, 0.854, and 0.827 in the three datasets. There was a statistically significant difference in RFS between the risk groups calculated using the delta-radiomics model and nomogram. ConclusionsThe nomogram combined with the delta-radiomic score and clinical-radiological risk factors could non-invasively predict ER of HCC after PTA.

Open and Endoscopic Gluteus Medius and/or Minimus Repair Achieves Clinical Success Regardless of Tear Grade: High-Grade Fatty Infiltration Portends Worse Outcomes

To evaluate minimum 2-year gluteus medius and/or minimus repair clinical success rates stratified by the 3-grade magnetic resonance imaging (MRI)-based classification (MRI grade) and to evaluate clinical success rates by the surgical approach used at each MRI grade and by the Goutallier-Fuchs (GF) classification. A retrospective review identified patients who underwent primary endoscopic or open gluteus medius and/or minimus repair from 2012 to 2021 performed by a single surgeon. Preoperative MRI scans were classified using the MRI grade and GF classification. Patient-reported outcomes were collected preoperatively and at minimum 2-year follow-up. Cohort-specific minimal clinically important difference and patient acceptable symptom state achievement was recorded. Rates of clinical success, defined as achievement of the 2-year minimal clinically important difference or patient acceptable symptom state with avoidance of revision surgery, were compared by MRI grade, by surgical approach at each MRI grade, and by GF classification. A total of 112 patients (71 with MRI grade 1, 19 with grade 2, and 22 with grade 3) were included. MRI grade 1 patients underwent endoscopic repair (P < .001) more often than the other groups. The overall clinical success rate was 90%. Clinical success rates by MRI grade were 93% for grade 1, 95% for grade 2, and 77% for grade 3 (P= .087). Clinical success rates by the endoscopic and open surgical approaches used at each MRI grade were 93% versus 90% for grade 1 (P= .543), 91% versus 100% for grade 2 (P > .999), and 60% versus 92% for grade 3 (P= .135). GF grade 1 tears achieved a higher rate of clinical success than GF grade 4 tears (100% vs 71%, P= .030). Primary repair of gluteus medius and/or minimus tears resulted in clinical success in most patients irrespective of MRI grade and irrespective of the surgical approach used at each MRI grade, yet GF grade 1 tears showed a significantly higher clinical success rate than GF grade 4 tears. Level IV, prognostic retrospective case series.

A tight anterosuperior intercondylar notch may increase the risk of cyclops syndrome after anterior cruciate ligament reconstruction using a quadruple semi-tendinosus short autograft

BackgroundCyclops syndrome is loss of terminal knee extension caused by a fibrous nodule developed in the anterior intercondylar notch. The many known risk factors include preoperative motion-range limitation, tibial tunnel malposition, and tight hamstrings. The primary objective of this study was to assess whether intercondylar notch size was associated with the risk of cyclops syndrome or graft tear after anterior cruciate ligament (ACL) reconstruction using a quadruple semi-tendinosis autograft. The secondary objective was to determine whether intercondylar notch size was associated with functional outcomes. HypothesisA narrow intercondylar notch is associated with higher risks of cyclops syndrome and poor functional outcomes. MethodsConsecutive patients who underwent ACL reconstruction by quadruple semi-tendinosus autograft were included retrospectively. Preoperative magnetic resonance imaging scans were assessed by a single senior surgeon, who determined the conventional notch width index (NWI) and the anterior NWI (aNWI) for each patient. ResultsThe 120 included patients had a mean follow-up of 2.4±0.8 years. Among them, 20 (16.7%) experienced cyclops syndrome and 7 (5.8%) graft rupture. At last follow-up, 26 (21.7%) had not returned to sports and only 47 (39.2%) had returned to sports at the pre-injury level. The mean Lysholm score was 87.9±13.5 and the main subjective IKDC score was 84±13. A narrow notch was significantly associated with lower likelihoods of returning to sports (p=0.001), returning to the same sport (p<0.0001), and returning to the pre-injury sport level (p=0.004). By multivariate analysis, only the aNWI index was significantly associated with the risk of cyclops syndrome (p<0.0001). An aNWI index lower than 0.18 had 85% sensitivity and 78% specificity for predicting cyclops syndrome. ConclusionA narrow anterosuperior intercondylar notch may increase the risk of cyclops syndrome after ACL reconstruction using a quadruple semi-tendinosus graft but is not associated with the risk of graft rupture. Level of evidenceIV, retrospective observational cohort study.

SurvNet: A low-complexity convolutional neural network for survival time classification of patients with glioblastoma

Background and objectiveMalignant primary brain tumors cause the greatest number of years of life lost than any other cancer. Grade 4 glioma is particularly devastating: The median survival without any treatment is less than six months and with standard-of-care treatment is only 14.6 months. Accurate identification of the overall survival time of patients with brain tumors is of profound importance in many clinical applications. Automated image analytics with magnetic resonance imaging (MRI) can provide insights into the prognosis of patients with brain tumors. MethodsIn this paper, We propose SurvNet, a low-complexity deep learning architecture based on the convolutional neural network to classify the overall survival time of patients with brain tumors into long-time and short-time survival cohorts. Through the incorporation of diverse MRI modalities as inputs, we facilitate deep feature extraction at various anatomical sites, thereby augmenting the precision of predictive modeling. We compare SurvNet with the Inception V3, VGG 16 and ensemble CNN models on pre-operative magnetic resonance image datasets. We also analyzed the effect of segmented brain tumors and training data on the system performance. ResultsSeveral measures, such as accuracy, precision, and recall, are calculated to examine the perfor-mance of SurvNet on three-fold cross-validation. SurvNet with T1 MRI modality achieved a 62.7 % accuracy, compared with 52.9 % accuracy of the Inception V3 model, 58.5 % accuracy of the VGG 16 model, and 54.9 % of the ensemble CNN model. By increasing the MRI input modalities, SurvNet becomes more accurate and achieves 76.5 % accuracy with four MRI modalities. Combining the segmented data, SurvNet achieved the highest accuracy of 82.4 %. ConclusionsThe research results show that SurvNet achieves higher metrics such as accuracy and f1-score than the comparisons. Our research also proves that by using multiparametric MRI modalities, SurvNet is able to learn more image features and performs a better classification accuracy. We can conclude that SurvNet with the complete scenario, i.e., segmented data and four MRI modalities, achieved the best accuracy, showing the validity of segmentation information during the survival time prediction process.

Prevalence of carotid artery stenosis and intra-cranial lesions in patients with aortic arch aneurysm and its association with intraoperative regional cerebral oxygen saturation and postoperative neurological outcomes.

Although aortic aneurysm is associated with vascular aging and atherosclerosis, carotid and intracranial vascular disease prevalence in patients with aortic arch aneurysm remains unclear. Similarly, the effect of carotid and intracranial lesions on postoperative outcomes is unknown. This study aimed to investigate the prevalence of carotid artery stenosis and intracranial lesions in patients with aortic arch aneurysm and its association with intraoperative regional cerebral oxygen saturation (rScO2) and postoperative neurological outcomes, including delirium and cerebral infarction. This retrospective observational study included 133 patients with true aortic arch aneurysm who underwent preoperative magnetic resonance imaging (MRI). We evaluated the prevalence of carotid and intracranial arterial lesions. Symptomatic cerebral infarction and delirium, defined by the confusion assessment method for the intensive care unit, were evaluated for their association with preoperative cerebrovascular lesions. Additionally, changes in regional saturation of the cerebral tissue at different surgical phases were evaluated for patients with and without cerebrovascular lesions. Fifteen (11.3%) patients experienced symptomatic cerebral infarction, and 64 (48.1%) had postoperative delirium. Preoperative MRI showed old infarction, microbleeds, significant carotid artery stenosis, and intracranial lesions in 21.1%, 14.3%, 10.5%, and 7.5% of the patients, respectively. White matter hyperintensities with Fazekas scale 2 were observed in 40.6% of the patients, while Fazekas scale 3 were observed in 18.8% of the patients. Preoperative MRI findings and postoperative neurological outcomes were not significantly different. Seventy-six patients underwent rScO2 monitoring intraoperatively. Changes in rScO2 in patients with and without carotid/cerebrovascular lesions were not significantly different. However, rScO2 was significantly lower in patients who developed cerebral infarction. Significant carotid artery stenosis and intracranial lesions were observed in 10.5% and 7.5% of the patients, respectively. Although preoperative MRI findings and changes in rScO2 or postoperative outcomes showed no significant association, patients with postoperative cerebral infarction showed significantly lower rScO2 intraoperatively.

Pineal Tumor Surgery—The Choice of the Approach Related to Tumor Characteristics and Posterior Fossa Anatomy

This research aimed to determine whether an adequate surgical approach can be chosen based on clearly defined values of anatomical landmarks (tentorial angle) and tumor size and extension. We conducted a retrospective analysis of patients operated on because of pineal tumors. The cohort was divided depending on the surgical approach. On preoperative magnetic resonance imaging, we measured maximal diameters, tumor volume, and tumor propagation. In the group of patients operated with the supracerebellar infratentorial approach, we also tested the correlation of tentorial angle with residual tumor. Differences among groups in resection, complications rate, and outcome were tested by the χ2 test. Finally, in both groups, the correlation of residual tumor with tumor volume, propagation, and diameters was tested using the receiver operating characteristic curve. In the group operated with a supracerebellar approach, total resection was achieved in 78% of the patients. The critical value of cranio-caudal diameter correlated with tumor residue was 31mm, for lateral-lateral diameter 25mm, for the lateral extension 14mm, and tumor volume 12cm3. Tentorial angle did not influence the extent of the resection. In the group operated with an occipital transtentorial approach, the critical tumor volume related to tumor residue was 9mm3, anterior-posterior diameter 29mm, and cranio-caudal diameter 28mm. The extent of the resection was significantly higher in the supracerebellar group. In both approaches, tumors larger than 3cm show an increased risk of subtotal resection. Except when most tumor volume is localized above the venous system, we advocate a supracerebellar corridor as an effective approach that is not limited by tentorial angle.

Membranous urethral length is the single independent predictor of urinary continence recovery at 12 months following Retzius-sparing robot-assisted radical prostatectomy

The influence of anatomical parameters on urinary continence (UC) after Retzius-sparing robot-assisted radical prostatectomy (RS-RARP) remains uncharted. Our objective was to evaluate their association with UC at 3, 6 and 12 months post-operatively. Data from patients who underwent RS-RARP were prospectively collected. Continence was defined as no pad use. Anatomic variables were measured on preoperative magnetic resonance imaging (MRI). Regression analyses were performed to identify predictors of UC at each time point. We included 158 patients with a median age of 60 years, most of whom had a localized tumor (≤ cT2). On multivariate analyses, at 3 months post-surgery, urinary incontinence (UI) rises with age, odds ratio (OR) 1.07 [95% confidence interval (CI) 1.004−1.142] and with prostate volume (PV), OR 1.029 (95% CI 1.006−1.052); it reduces with longer membranous urethral length (MUL), OR 0.875 (95% CI 0.780−0.983) and with higher membranous urethral volume (MUV), OR 0.299 (95% CI 0.121−0.737). At 6 months, UI rises with PV, OR 1.033 (95% CI 1.011−1.056) and decreases with MUV, OR 0.1504 (95% CI 0.050−0.444). Significantly, at 12 months post-surgery, the only predictor of UI is MUL, OR 0.830 (95% CI 0.706−0.975), establishing a threshold associated with a risk of UI of 5% (MUL > 15 mm) in opposition to a risk of 25% (MUL < 10 mm). This single institutional study requires external validation. To our knowledge, this is the first prospective cohort study supporting MUL as the single independent predictor of UC at 12 months post-surgery. By establishing MUL thresholds, we enable precise patient counseling.

Preoperative Physical Therapy is Associated with Decreased Length of Stay and Improved Postoperative Mobility in Patients with Sarcopenia Undergoing Adult Spinal Deformity Surgery.

Retrospective cohort study. To determine the relationship between preoperative physical therapy (PT) and postoperative mobility, adverse events (AEs), and length of stay (LOS) among patients with low normalized total psoas area (NTPA) undergoing ASD surgery. Sarcopenia as defined by low NTPA has been shown to predict poor perioperative outcomes following adult spinal deformity (ASD) surgery. However, there is limited evidence correlating the benefits of PT within the sarcopenic patient population. NTPA was analyzed at the L3 and L4 mid-vertebral body on preoperative magnetic resonance imaging (MRI). Receiver operating characteristic (ROC) curve analysis was used to determine gender-specific NTPA cut-off values for predicting perioperative AEs. Patients were categorized as having low NTPA if both L3 and L4 NTPA were below these cut-off values. Perioperative outcomes were compared between patients with low NTPA that underwent documented formal PT within 6 months prior to ASD surgery with those that did not. 103 patients (42 males, 61 females) met criteria for low NTPA for inclusion in the study, of which 42 underwent preoperative PT and 61 did not. The preoperative PT group had a shorter LOS (111.2±37.5 vs. 162.1±97.0h, P<0.001), higher ambulation distances (feet) on postoperative day (POD) 1 (61.7±50.3 vs. 26.1±69.0, P<0.001), POD 2 (113.2±81.8 vs. 62.1±73.1, P=0.003), and POD 3 (126.0±61.2 vs. 91.2±72.6, P=0.029), and lower rates of total AEs (31.0% vs. 54.1%, P=0.003) when excluding anemia requiring transfusion. Multivariable analysis found preoperative PT to be the most significant predictor of decreased LOS (OR 0.32, P=0.013). Sarcopenic patients may benefit from formal preoperative PT prior to undergoing ASD surgery to improve early postoperative mobility, decrease AEs, and decrease LOS. 3.

Comparative analysis of intraoperative and imaging features of invasive growth in pituitary adenomas.

Most pituitary adenomas (PAs), also termed pituitary neuroendocrine tumors, are benign in nature and can be treated effectively by surgical resection, medical treatment, and in special cases by radiotherapy. However, invasive growth can be an important feature of a more aggressive behavior and adverse prognosis. The extension of PAs into the cavernous sinus can be categorized according to the Knosp criteria on magnetic resonance imaging (MRI). Comparative analyses of MRI features and intraoperative findings of invasive growth regarding different clinical factors are still scarce. We performed a retrospective single-center analysis of 764 PAs that were surgically treated between October 2004 and April 2018. Invasive growth was assessed according to the surgical reports and preoperative MRI (Knosp criteria). Clinical data, such as patient age at diagnosis and gender, histopathological adenoma type, and extent of resection, were collected. Invasive features on MRI were seen in 24.4% (Knosp grades 3A-4, 186/764) of the cases. Intraoperatively, invasion was present in 42.4% (324/764). Complete resection was achieved in 80.0% of adenomas and subtotal resection, in 20.1%. By multivariate analysis, invasion according to intraoperative findings was associated with the sparsely granulated corticotroph (SGCA, P = .0026) and sparsely granulated somatotroph (SGSA, P = .0103) adenoma type as well as age (P = .0287). Radiographic invasion according to Knosp grades 3A-4 correlated with age (P = .0098), SGCAs (P = .0005), SGSAs (P = .0351), and gonadotroph adenomas (P = .0478). Both criteria of invasion correlated with subtotal resection (P = .0001, respectively). Both intraoperative and radiographic signs of invasive growth are high-risk lesions for incomplete extent of resection and occur more frequently in older patients. A particularly high prevalence of invasion can be found in the SGCA and SGSA types. Cavernous sinus invasion is also more common in gonadotroph adenomas. Usage of the Knosp classification is a valuable preoperative estimation tool.

A predictive model using MRI and clinicopathologic features for breast cancer recurrence in young women treated with upfront surgery.

To identify preoperative breast MR imaging and clinicopathological variables related to recurrence and develop a risk prediction model for recurrence in young women with breast cancer treated with upfront surgery. This retrospective study analyzed 438 consecutive women with breast cancer aged 35 years or younger between January 2007 and December 2016. Breast MR images before surgery were independently reviewed by breast radiologists blinded to patient outcomes. The clinicopathological data including patient demographics, clinical features, and tumor characteristics were reviewed. Univariate and multivariate logistic regression analyses were used to identify the independent factors associated with recurrence. The risk prediction model for recurrence was developed, and the discrimination and calibration abilities were assessed. Of 438 patients, 95 (21.7%) developed recurrence after a median follow-up of 65 months. Tumor size at MR imaging (HR = 1.158, p = 0.006), multifocal or multicentric disease (HR = 1.676, p = 0.017), and peritumoral edema on T2WI (HR = 2.166, p = 0.001) were identified as independent predictors of recurrence, while adjuvant endocrine therapy (HR = 0.624, p = 0.035) was inversely associated with recurrence. The prediction model showed good discrimination ability in predicting 5-year recurrence (C index, 0.707 in the development cohort; 0.686 in the validation cohort) and overall recurrence (C index, 0.699 in the development cohort; 0.678 in the validation cohort). The calibration plot demonstrated an excellent correlation (concordance correlation coefficient, 0.903). A prediction model based on breast MR imaging and clinicopathological features showed good discrimination to predict recurrence in young women with breast cancer treated with upfront surgery, which could contribute to individualized risk stratification. Our prediction model, incorporating preoperative breast MR imaging and clinicopathological features, predicts recurrence in young women with breast cancer undergoing upfront surgery, facilitating personalized risk stratification and informing tailored management strategies. Younger women with breast cancer have worse outcomes than those diagnosed at more typical ages. The described prediction model showed good discrimination performance in predicting 5-year and overall recurrence. Incorporating better risk stratification tools in this population may help improve outcomes.

Deformable registration of magnetic resonance images using unsupervised deep learning in neuro-/radiation oncology

PurposeAccurate deformable registration of magnetic resonance imaging (MRI) scans containing pathologies is challenging due to changes in tissue appearance. In this paper, we developed a novel automated three-dimensional (3D) convolutional U-Net based deformable image registration (ConvUNet-DIR) method using unsupervised learning to establish correspondence between baseline pre-operative and follow-up MRI scans of patients with brain glioma.MethodsThis study involved multi-parametric brain MRI scans (T1, T1-contrast enhanced, T2, FLAIR) acquired at pre-operative and follow-up time for 160 patients diagnosed with glioma, representing the BraTS-Reg 2022 challenge dataset. ConvUNet-DIR, a deep learning-based deformable registration workflow using 3D U-Net style architecture as a core, was developed to establish correspondence between the MRI scans. The workflow consists of three components: (1) the U-Net learns features from pairs of MRI scans and estimates a mapping between them, (2) the grid generator computes the sampling grid based on the derived transformation parameters, and (3) the spatial transformation layer generates a warped image by applying the sampling operation using interpolation. A similarity measure was used as a loss function for the network with a regularization parameter limiting the deformation. The model was trained via unsupervised learning using pairs of MRI scans on a training data set (n = 102) and validated on a validation data set (n = 26) to assess its generalizability. Its performance was evaluated on a test set (n = 32) by computing the Dice score and structural similarity index (SSIM) quantitative metrics. The model’s performance also was compared with the baseline state-of-the-art VoxelMorph (VM1 and VM2) learning-based algorithms.ResultsThe ConvUNet-DIR model showed promising competency in performing accurate 3D deformable registration. It achieved a mean Dice score of 0.975 ± 0.003 and SSIM of 0.908 ± 0.011 on the test set (n = 32). Experimental results also demonstrated that ConvUNet-DIR outperformed the VoxelMorph algorithms concerning Dice (VM1: 0.969 ± 0.006 and VM2: 0.957 ± 0.008) and SSIM (VM1: 0.893 ± 0.012 and VM2: 0.857 ± 0.017) metrics. The time required to perform a registration for a pair of MRI scans is about 1 s on the CPU.ConclusionsThe developed deep learning-based model can perform an end-to-end deformable registration of a pair of 3D MRI scans for glioma patients without human intervention. The model could provide accurate, efficient, and robust deformable registration without needing pre-alignment and labeling. It outperformed the state-of-the-art VoxelMorph learning-based deformable registration algorithms and other supervised/unsupervised deep learning-based methods reported in the literature.

Differentiating mixed epithelial and stromal tumor family from predominantly cystic renal cell carcinoma using magnetic resonance imaging-based Bosniak classification system version 2019.

To differentiate mixed epithelial and stromal tumor family (MESTF) of the kidney from predominantly cystic renal cell carcinoma (RCC) using the magnetic resonance imaging (MRI)-based Bosniak classification system version 2019 (v2019). The study included 36 consecutive patients with MESTF and 77 with predominantly cystic RCC who underwent preoperative renal MRI. One radiologist evaluated and documented the clinical and MRI characteristics (age, sex, laterality, R.E.N.A.L. Nephrometry Score [RNS], surgical approach, the signal intensity on T2-weighted imaging, restricted diffusion and enhancement features in corticomedullary phase). Blinded to clinical and pathological information, another two radiologists independently evaluated Bosniak category of all masses. Interobserver agreement based on Bosniak classification system v2019 was measured by the weighted Cohen/Conger's Kappa coefficient. Furthermore, predominantly cystic RCCs and MESTFs were divided into low (categories I, II, and IIF) and high-class (categories III, and IV) tumors. The independent sample t test (Mann-Whitney U test) or Pearson Chi-square test (Fisher's exact probability test) was utilized to compare clinical and imaging characteristics between MESTFs and predominantly cystic RCCs. The performance of the Bosniak classification system v2019 in distinguishing MESTF from predominantly cystic RCC was investigated via receiver operating characteristic curve analysis. MESTF and predominantly cystic RCC groups significantly differed in terms of age, lesion size, RNS, restricted diffusion, and obvious enhancement in corticomedullary phase, but not sex, laterality, surgical approach, and the signal intensity on T2WI. Interobserver agreement was substantially based on the Bosniak classification system v2019. There were 24 low-class tumors and 12 high-class tumors in the MESTF group. Meanwhile, 13 low-class tumors and 64 high-class tumors were observed in the predominantly cystic RCC group. The distribution of low- or high-class tumors significantly differed between the MESTF and predominantly cystic RCC groups. Bosniak classification system v2019 had excellent discrimination (cutoff value = category III), and an area under curve value was 0.81; accuracy, 80.5%; sensitivity, 87.0%; and specificity, 66.7%. The MRI-based Bosniak classification system v2019 can effectively distinguish MESTF from predominantly cystic RCC if category III was used as a cutoff reference.

Magnetic resonance imaging should be the first-line imaging modality for investigating suspected proximal hamstring avulsion injuries.

The purpose of the study was to evaluate how the initial diagnostic tool used-specifically ultrasound (US) or magnetic resonance imaging (MRI)-for diagnosing proximal hamstring avulsion injury affects the delay before surgery and, secondarily, the outcomes of these injuries. This was a retrospective analysis of prospectively collected data (2012-2020). It targeted patients primarily treated for proximal hamstring avulsion injury. It included all patients with a preoperative US and/or MRI. Patients were divided into two groups based on the initial diagnostic tool used (US-first vs. MRI-first groups). The primary outcomes measure was the time from initial injury to surgical intervention (surgical delay). The secondary outcomes were the Parisien Hamstring Avulsion Score, as well as the activity level as measured by the University of California, Los Angeles (UCLA)Activity Scale and Tegner Activity Scale. The analysis included 392 patients with a mean age of 43.8 ± 13.6 years for the MRI-firstgroup and 47.6 ± 12.0 years for US-first group. Patients in the MRI-first group had a significantly shorter median time from injury to surgery of 20.0 days (interquartile range [IQR]: 11.0-61.0) compared to 30 days (IQR: 18-74) in the US-first group. At the final follow-up (4.2 ± 2.2 years for the MRI-first group and 5.1 ± 1.9 years for the US-first group), the MRI-first group had significantly higher mean Tegner Activity Scale and UCLA scores than the US-first group: The Tegner Activity Scale was median 5 (IQR: 3-7) for the MRI-first group versus median 4 (IQR: 2-6) for the US-first group (p < 0.05). The UCLA scores were 7.9 ± 2.4 for the MRI-first group compared to 7.3 ± 2.4 for the US-first group (p < 0.05). This difference was more pronounced when comparing the MRI-first group with the patient-false negative initial ultrasound. No difference was found regarding the Parisien Hamstring Avulsion Score. MRI as the initial diagnostic tool for proximal hamstring avulsion injury is associated with a shorter time to surgery and better postoperative outcomes in Tegner Activity Scale and UCLA scores, compared to US. Level III.

Body mass weighted prostate-specific antigen levels, new markers to predict locally advanced prostate cancer after prostatectomy.

Prostate-specific antigen (PSA) remains the most useful marker for screening, risk categorization, and follow-up in patients with prostate cancer. In the obese population, several studies have revealed that obesity may not only inversely interfere with the concentration of PSA, but also increase the risk of prostate cancer. Thus, we considered using the body mass weighted PSA levels, presented as serum PSA concentration multiplied by body weight or body mass index (BMI), instead of traditional PSA concentration, as potential markers to predict locally advanced prostate cancer after prostatectomy. We retrospectively collected and analyzed data acquired from a single institute at which robot-assisted laparoscopic radical prostatectomy was performed. A total of 174 patients underwent radical prostatectomy, and the collected data included age, PSA level, body weight, BMI, and pathology results. A total of 174 patients were diagnosed with adenocarcinoma of the prostate by needle biopsy, and most (N = 165) were considered to have localized disease on preoperative multiparameter magnetic resonance imaging. After prostatectomy, 73% (N = 127) of the patients remained in the localized disease group (group A) and 27% (N = 47) of the patients were reclassified to the locally advanced prostate cancer (group B). The value of PSA was higher in group B (16.9 vs 11.2 ng/dL; p = 0.062), but there was no statistically significant difference between the two groups. After using the numerical values of PSA × body weight and PSA × BMI, a statistically significant difference emerged between the two groups ( p = 0.0198 in PSA × BW; p = 0.0110 in PSA × BMI). The body mass-weighted PSA levels, instead of the traditional PSA concentration, may be better markers for predicting non-organ-confined disease after surgery. It may also be useful in screening and risk categorization.

MRI of pelvic endometriosis: evaluation of the mr#Enzian classification and the importance of adenomyosis subtypes

PurposeThis study aimed to investigate the utility of the #Enzian classification in magnetic resonance imaging (MRI) for endometriosis assessment, focusing on inter-reader agreement, diagnostic accuracy, and the correlation of adenomyosis with deep endometriosis (DE).MethodsThis IRB- approved retrospective single-center study included 412 women who underwent MRI evaluation for endometriosis between February 2017 and June 2022. Two experienced radiologists independently analyzed MRI images using the #Enzian classification and assessed the type of adenomyosis, if any. The surgical #Enzian classification served as the gold standard for evaluating preoperative MRI results of 45 patients. Statistical analysis was performed to assess inter-reader agreement and diagnostic accuracy.ResultsInter-reader agreement was substantial to excellent (Cohen’s kappa 0.75–0.96) for most compartments except peritoneal involvement (0.39). The preoperative MRI showed mostly substantial to excellent accuracy (0.84–0.98), sensitivity (0.62–1.00), specificity (0.87–1.00), positive (0.58–1.00) and negative predictive values (0.86–1.00) for most compartments, except for peritoneal lesions (0.36, 0.17, 1.00, 1.00, 0.26 respectively). A trend with a higher prevalence of concordant DE in women with MR features of external adenomyosis compared to those with internal adenomyosis was visible (p = 0.067).ConclusionsThe mr#Enzian showed mostly high inter-reader agreement and good diagnostic accuracy for various endometriosis compartments. MRI’s role is particularly significant in the context of the current paradigm shift towards medical endometriosis treatment. The inclusion of information about the type of adenomyosis in the mr#Enzian classification could enhance diagnostic accuracy and inform treatment planning.Graphical

Assessing the Impact of Preoperative MRI-Based Vertebral Bone Quality Scores on 5-Year Prognosis in Lumbar Spine Surgery.

Retrospective cohort study. To assess the utility of preoperative vertebral bone quality (VBQ) scores in predicting the 5-year clinical outcomes following lumbar spine surgery. Osteoporosis poses a significant concern in older adults undergoing spinal surgery. The VBQ score, assessed through preoperative magnetic resonance imaging (MRI), is associated with subsequent osteoporotic fractures and postoperative complications. However, previous report on the impact of VBQ score on mid-term clinical outcomes after lumbar spine surgery remains lacking. A total of 189 patients who underwent lumbar surgery (≤3-disc levels) for lumbar spinal stenosis between 2010 and 2016 were enrolled. Patients were classified into high (>3.35), middle (2.75 to 3.35), and low (<2.73) VBQ score groups based on tertiles. Clinical scores, including Japanese Orthopedic Association Back Pain Evaluation Questionnaire (JOABPEQ) and Short Form-36 (SF-36), were recorded preoperatively and 1, 2, and 5 years postoperatively. Comparative analysis showed significant differences among the VBQ groups throughout the study period in low back pain (P=0.013), walking ability (P=0.005), social life function (P=0.010) of JOABPEQ, and physical component summary of the SF-36 (P=0.018) following lumbar spine surgery. A higher VBQ score was significantly correlated with worse 5-year postoperative outcomes for all domains except for lumbar function of the JOABPEQ using multiple linear regression analysis, adjusting for age, sex, BMI, hyperlipidemia, surgical procedures, and each preoperative score. A high preoperative VBQ score is a risk factor for poor 5-year clinical outcomes after lumbar spine surgery. Evaluation of the VBQ score through routine preoperative MRI facilitates osteoporotic screening in lumbar patients without radiation exposure and healthcare costs, while also demonstrating its potential as a prognostic indicator of postoperative clinical outcomes. 3.

The presence of a deep lateral femoral notch sign in ACL-injured patients is associated with a 2.7° steeper posterior tibial slope and a 19% higher frequency of lateral meniscal injuries.

The purpose of this study was to study the relationship between the presence of a deep lateral femoral notch sign (DLFNS) in anterior cruciate ligament (ACL)-injured patients and a higher posterior lateral tibial slope (LPTS), a reduced meniscal bone angle (MBA), a higher LPTS/MBA ratio and a higher incidence of concomitant injuries in primary ACL tears. A retrospective case-control study was performed in patients submitted to primary ACL reconstruction with an available preoperative magnetic resonance imaging (MRI) scan. Patients with ACL tears and a femoral impactation with a depth ≥2 mm were assorted to the DLFNS groupand patients with ACL tear and without a DLFNS to the control group. LPTS and MBA were measured in MRI. The presence of concomitant injuries (meniscal, posterior cruciate ligament, medial collateral ligament, lateral collateral ligament and bone injuries) was assessed in MRI. Quantitative data are presented in the median ± interquartile range (IQR). There were 206 patients included in the study, with 46 patients assorted to the DLFNS group and 160 patients to the control group. In the DLFNS group, the median LPTS was 6.7° (IQR: 4.0-8.2) versus 4.0° in the control group (IQR: 2.2-6.5) (p = 0.003). The LPTS/MBA ratio was significantly higher in the DLFNS group, with a median of 0.32 (IQR: 0.19-0.44), in comparison to the control group, with a median of 0.19 (IQR: 0.11-0.31) (p < 0.001). The multivariable logistic regression analysis showed that the LPTS is an independent risk factor to having a DLFNS (odds ratio [OR] = 1.161; 95% confidence interval [CI]: 1.042-1.293, p = 0.007). There was a higher incidence of concomitant lateral meniscal injuries in the DLFNS group (67% vs. 48%, p = 0.017). In patients with ACL tears, the presence of a DLFNS is associated with a steeper lateral posterior tibial slope, as well as a higher incidence of concomitant lateral meniscal injuries. Level III.