Abnormal Brachial Plexus Differentiation from Routine Magnetic Resonance Imaging: An AI-based Approach.

Automatic abnormality identification of brachial plexus (BP) from normal magnetic resonance imaging to localize and identify a neurologic injury in clinical practice (MRI) is still a novel topic in brachial plexopathy. This study developed and evaluated an approach to differentiate abnormal BP with artificial intelligence (AI) over three commonly used MRI sequences, i.e. T1, FLUID sensitive and post-gadolinium sequences. A BP dataset was collected by radiological experts and a semi-supervised artificial intelligence method was used to segment the BP (based on nnU-net). Hereafter, a radiomics method was utilized to extract 107 shape and texture features from these ROIs. From various machine learning methods, we selected six widely recognized classifiers for training our Brachial plexus (BP) models and assessing their efficacy. To optimize these models, we introduced a dynamic feature selection approach aimed at discarding redundant and less informative features. Our experimental findings demonstrated that, in the context of identifying abnormal BP cases, shape features displayed heightened sensitivity compared to texture features. Notably, both the Logistic classifier and Bagging classifier outperformed other methods in our study. These evaluations illuminated the exceptional performance of our model trained on FLUID-sensitive sequences, which notably exceeded the results of both T1 and post-gadolinium sequences. Crucially, our analysis highlighted that both its classification accuracies and AUC score (area under the curve of receiver operating characteristics) over FLUID-sensitive sequence exceeded 90%. This outcome served as a robust experimental validation, affirming the substantial potential and strong feasibility of integrating AI into clinical practice.

Deep-learning for automated detection of MSU deposits on DECT: evaluating impact on efficiency and reader confidence.

Introduction: Dual-energy CT (DECT) is a non-invasive way to determine the presence of monosodium urate (MSU) crystals in the workup of gout. Color-coding distinguishes MSU from calcium following material decomposition and post-processing. Manually identifying these foci (most commonly labeled green) is tedious, and an automated detection system could streamline the process. This study aims to evaluate the impact of a deep-learning (DL) algorithm developed for detecting green pixelations on DECT on reader time, accuracy, and confidence. Methods: We collected a sample of positive and negative DECTs, reviewed twice-once with and once without the DL tool-with a 2-week washout period. An attending musculoskeletal radiologist and a fellow separately reviewed the cases, simulating clinical workflow. Metrics such as time taken, confidence in diagnosis, and the tool's helpfulness were recorded and statistically analyzed. Results: We included thirty DECTs from different patients. The DL tool significantly reduced the reading time for the trainee radiologist (p = 0.02), but not for the attending radiologist (p = 0.15). Diagnostic confidence remained unchanged for both (p = 0.45). However, the DL model identified tiny MSU deposits that led to a change in diagnosis in two cases for the in-training radiologist and one case for the attending radiologist. In 3/3 of these cases, the diagnosis was correct when using DL. Conclusions: The implementation of the developed DL model slightly reduced reading time for our less experienced reader and led to improved diagnostic accuracy. There was no statistically significant difference in diagnostic confidence when studies were interpreted without and with the DL model.

A review of epidural and non-epidural contrast flow patterns during fluoroscopic and CT-guided epidural steroid injections.

Epidural steroid injections are commonly performed using fluoroscopic or CT guidance. With both modalities, the injection of contrast material is necessary before steroid administration to ensure adequate epidural flow and exclude non-epidural flow. While fluoroscopic guidance is conventional, CT is utilized at some centers and can be particularly helpful in the setting of challenging or postoperative anatomy. It is important for proceduralists to be adept at evaluating contrast media flow patterns under both modalities. The goal of this review article is to describe and provide examples of epidural and non-epidural flow patterns on both conventional fluoroscopy and CT. Specific non-epidural patterns discussed include intrathecal flow, intradural/subdural flow, vascular uptake, flow into the retrodural space of Okada, inadvertent facet joint flow, and intradiscal flow.

A Machine Learning Model Demonstrates Excellent Performance in Predicting Subscapularis Tears Based on Pre-Operative Imaging Parameters Alone.

PURPOSE: To develop a machine learning model capable of identifying subscapularis tears before surgery based on imaging and physical examination findings. METHODS: Between 2010 and 2020, 202 consecutive shoulders underwent arthroscopic rotator cuff repair by a single surgeon. Patient demographics, physical examination findings (including range of motion, weakness with internal rotation, lift/push-off test, belly press test, and bear hug test), and imaging (including direct and indirect signs of tearing, biceps status, fatty atrophy, cystic changes, and other similar findings) were included for model creation. RESULTS: Sixty percent of the shoulders had partial or full thickness tears of the subscapularis verified during surgery (83% of these were upper third). Using only preoperative imaging-related parameters, the XGBoost model demonstrated excellent performance at predicting subscapularis tears (c-statistic, 0.84; accuracy, 0.85; F1 score, 0.87). The top 5 features included direct signs related to the presence of tearing as evidenced on magnetic resonance imaging (MRI) (changes in tendon morphology and signal), as well as the quality of the MRI and biceps pathology. CONCLUSIONS: In this study, machine learning was successful in predicting subscapularis tears by MRI alone in 85% of patients, and this accuracy did not decrease by isolating the model to the top features. The top five features included direct signs related to the presence of tearing as evidenced on MRI (changes in tendon morphology and signal), as well as the quality of the MRI and biceps pathology. Last, in advanced modeling, the addition of physical examination or patient characteristics did not make a significant difference in the predictive ability of this model. LEVEL OF EVIDENCE: Level III, diagnostic case-control study.

Deep learning-based virtual noncalcium imaging in multiple myeloma using dual-energy CT.

BACKGROUND: Dual-energy CT with virtual noncalcium (VNCa) images allows the evaluation of focal intramedullary bone marrow involvement in patients with multiple myeloma. However, current commercial VNCa techniques suffer from excessive image noise and artifacts due to material decomposition used in synthesizing VNCa images. OBJECTIVES: In this work, we aim to improve VNCa image quality for the assessment of focal multiple myeloma, using an Artificial intelligence based Generalizable Algorithm for mulTi-Energy CT (AGATE) method. MATERIALS AND METHODS: AGATE method used a custom dual-task convolutional neural network (CNN) that concurrently carries out material classification and quantification. The material classification task provided an auxiliary regularization to the material quantification task. CNN parameters were optimized using custom loss functions that involved cross-entropy, physics-informed constraints, structural redundancy in spectral and material images, and texture information in spectral images. For training data, CT phantoms (diameters 30 to 45 cm) with tissue-mimicking inserts were scanned on a third generation dual-source CT system. Scans were performed at routine dose and half of the routine dose. Small image patches (i.e., 40 × 40 pixels) of tissue-mimicking inserts with known basis material densities were extracted for training samples. Numerically simulated insert materials with various shapes increased diversity of training samples. Generalizability of AGATE was evaluated using CT images from phantoms and patients. In phantoms, material decomposition accuracy was estimated using mean-absolute-percent-error (MAPE), using physical inserts that were not used during the training. Noise power spectrum (NPS) and modulation transfer function (MTF) were compared across phantom sizes and radiation dose levels. Five patients with multiple myeloma underwent dual-energy CT, with VNCa images generated using a commercial method and AGATE. Two fellowship-trained musculoskeletal radiologists reviewed the VNCa images (commercial and AGATE) side-by-side using a dual-monitor display, blinded to VNCa type, rating the image quality for focal multiple myeloma lesion visualization using a 5-level Likert comparison scale (-2 = worse visualization and diagnostic confidence, -1 = worse visualization but equivalent diagnostic confidence, 0 = equivalent visualization and diagnostic confidence, 1 = improved visualization but equivalent diagnostic confidence, 2 = improved visualization and diagnostic confidence). A post hoc assignment of comparison ratings was performed to rank AGATE images in comparison to commercial ones. RESULTS: AGATE demonstrated consistent material quantification accuracy across phantom sizes and radiation dose levels, with MAPE ranging from 0.7% to 4.4% across all testing materials. Compared to commercial VNCa images, the AGATE-synthesized VNCa images yielded considerably lower image noise (50-77% noise reduction) without compromising noise texture or spatial resolution across different phantom sizes and two radiation doses. AGATE VNCa images had markedly reduced area under NPS curves and maintained NPS peak frequency (0.7 lp/cm to 1.0 lp/cm), with similar MTF curves (50% MTF at 3.0 lp/cm). In patients, AGATE demonstrated reduced image noise and artifacts with improved delineation of focal multiple myeloma lesions (all readers comparison scores indicating improved overall diagnostic image quality [scores 1 or 2]). CONCLUSIONS: AGATE demonstrated reduced noise and artifacts in VNCa images and ability to improve visualization of bone marrow lesions for assessing multiple myeloma.

Radiology of Intra-Aortic Balloon Pump Catheters.

Radiographs play an important role in ascertaining appropriate placement of the intra-aortic balloon pump catheter. This imaging essay highlights correct and incorrect positioning of these catheters, with emphasis on the variability of radiopaque markers used with different catheter models and on axillary versus femoral catheter placement routes. Keywords: Conventional Radiography, CT, Percutaneous, Cardiac, Vascular, Aorta, Anatomy, Cardiac Assist Devices, Catheters © RSNA, 2022.

Gadolinium retention within multiple rat organs after intra-articular administration of gadolinium-based contrast agents.

OBJECTIVE: To characterize the extent of retention and biodistribution of gadolinium (Gd) following intra-articular (IA) injection of linear and macrocyclic gadolinium-based contrast agents (GBCAs) into the knee joint of a rat model. MATERIALS AND METHODS: Fifteen Wistar rats were divided into five groups and underwent fluoroscopically-guided injections of both knee joints of (1) clinical 1:200 dilution (low dose, LD) gadodiamide (linear GBCA), (2) LD gadobutrol (macrocyclic GBCA), (3) undiluted (high dose, HD) gadodiamide, (4) HD gadobutrol, and (5) saline. Gd concentrations were quantified by inductively coupled plasma mass spectrometry in (1) blood and urine samples obtained over a 72 h period and (2) knee joint tissues, brain, kidney, and bone marrow at 3 days post-injection. RESULTS: Both HD and LD gadodiamide and gadobutrol were rapidly absorbed from the joint with peak serum and urine concentration at 1 h post-injection, with relatively faster clearance of gadobutrol. All GBCA-exposed groups had detectable levels of Gd in the joint tissues, bone marrow, and/or kidneys (median tissue gadolinium range: 0.1-71 µg Gd/g tissue), with higher amounts observed with gadodiamide versus gadobutrol. Retention within brain tissues was only detected following HD gadodiamide administration but not LD gadodiamide nor HD or LD gadobutrol. CONCLUSION: There was rapid systemic absorption, redistribution, and widespread multi-organ retention of Gd following IA injection of both linear and macrocyclic GBCAs, despite substantial amounts of urinary excretion. Higher concentrations of Gd were observed with administration of gadodiamide compared to gadobutrol in most tissues and biofluids.

CT Attenuation Values Do Not Reliably Distinguish Benign Sclerotic Lesions From Osteoblastic Metastases in Patients Undergoing Bone Biopsy.

BACKGROUND. CT attenuation thresholds that accurately distinguish enostoses from untreated osteoblastic metastases have been published. In the Mayo Clinic practices, these thresholds have been applied more broadly to distinguish benign sclerotic bone lesions other than enostoses from osteoblastic metastases. OBJECTIVE. The purpose of this article is to determine if CT attenuation thresholds allow the distinguishing of benign sclerotic bone lesions from osteoblastic metastases in patients undergoing bone biopsy. METHODS. A retrospective search was conducted to identify sclerotic lesions described on CT between October 7, 1998, and July 15, 2018, that underwent subsequent biopsy. Two musculoskeletal radiologists recorded lesions' maximum and mean attenuation. Using previously published attenuation thresholds, sensitivity and specificity for differentiating benign sclerotic lesions from osteoblastic metastases were calculated. ROC curve analysis was performed to determine if more appropriate attenuation thresholds exist. Intraclass correlation coefficients (ICCs) were computed. RESULTS. A total of 280 patients met inclusion criteria. Of those, 162 had malignant biopsy results and 118 had benign biopsy results. Of the 162 malignant lesions, 81 had received prior treatment. Maximum and mean attenuation were not significantly different between benign and malignant lesions for either reader (all p > .05). For reader 1, to distinguish benign from malignant lesions, a maximum attenuation threshold of more than 1060 HU resulted in sensitivity of 23.7%, specificity of 87.0%, and accuracy of 60.6%. A mean attenuation threshold of greater than 885 HU resulted in sensitivity of 19.5%, specificity of 90.7%, and accuracy 60.7%. ROC curve analysis showed AUCs for mean and maximum attenuation thresholds of 51.8% and 54.6%, respectively. Subgroup analyses of benign versus malignant and treated versus untreated lesions had similar results. Similar findings were obtained for reader 2. The two readers' ICC was 0.946 for maximum attenuation and 0.918 for mean attenuation. CONCLUSION. Published attenuation thresholds for distinguishing enostoses from osteoblastic metastases had slightly decreased specificity and markedly decreased sensitivity when applied to the differentiation of benign sclerotic lesions from osteoblastic metastases in our sample of biopsy-proven lesions. ROC analysis showed no high-performing attenuation threshold alternative. CLINICAL IMPACT. Published CT attenuation thresholds intended for distinguishing enostoses from osteoblastic metastases should not be used more broadly. More accurate alternative thresholds could not be derived.

Investigating the Chronology of Meniscus Root Tears: Do Medial Meniscus Posterior Root Tears Cause Extrusion or the Other Way Around?

BACKGROUND: Meniscus root tears are increasingly being recognized. Meniscal extrusion has previously been associated with medial root tears; however, the relationship between secondary meniscal restraints, such as the meniscotibial (MT) ligament, extrusion, and root tears has yet to be formally evaluated. PURPOSE: To better understand the association between MT ligament competence, medial meniscal extrusion, and medial meniscus posterior root tears (MMPRTs) as well as to determine the progression of meniscal extrusion over time. STUDY DESIGN: Case series; Level of evidence, 4. METHODS: Serial magnetic resonance imaging (MRI) scans were reviewed for patients who showed evidence of medial meniscal extrusion and MMPRTs on at least 1 of ≥2 available MRI scans. All patients were symptomatic at the time of diagnosis. All MRI scans were analyzed independently by 2 board-certified musculoskeletal radiologists. MT ligament disruption, medial meniscal extrusion, and MMPRTs were recorded for each MRI scan. The time between MRI scans, presence of insufficiency fractures, and Outerbridge classification for the medial femur and tibia were also evaluated. RESULTS: Overall, 27 knees in 26 patients were included in this study, with a total of 63 MRI scans analyzed (21 knees with 2 MRI scans, 3 with 3 MRI scans, and 3 with 4 MRI scans). All patients demonstrated clear medial meniscal extrusion and MT ligament disruption before the subsequent development of MMPRTs (P < .001). Mean extrusion at the time of initial MRI was 3.3 ± 1.1 mm and increased significantly to 5.5 ± 1.8 mm at the time of first imaging with an identified MMPRT (P < .001). The mean time between initial MRI and the first identification of an MMPRT on later MRI was 1.7 ± 1.6 years. CONCLUSION: In a sample of 27 symptomatic knees with serial MRI scans both before and after an MMPRT diagnosis, all patients demonstrated MT ligament disruption and associated meniscal extrusion before the development of subsequent medial meniscus root tears. These findings suggest that MT ligament disruption and medial meniscal extrusion represent early and predisposing events contributing to MMPRTs. Therefore, this provides a possible explanation of why meniscal extrusion is not corrected with medial meniscus root repair.

Normal Pacinian corpuscles in the hand: radiology-pathology correlation in a cadaver study.

OBJECTIVE: To provide microdissection and histological confirmation of normal Pacinian corpuscles prospectively identified using MRI in a cadaver model. METHODS: 3-T MRI of a cadaveric hand specimen was performed with fiduciary markers on the skin. Based on previous descriptions, subcutaneous nodules representing presumed Pacinian corpuscles were localized with respect to the skin markers, and their sizes and depths were recorded. Focused ultrasound was performed to attempt to visualize the corpuscles. Subsequent microdissection was then performed and the presence and location of Pacinian corpuscles were recorded and compared with the findings on MRI. Histological evaluation for each identified corpuscle was performed. RESULTS: The MRI demonstrated 11 T2-hyperintense palmar subcutaneous nodules around the second through fifth metacarpophalangeal joints. None was visible sonographically. The first eight were dissected and proved to be normal Pacinian corpuscles histologically. In sites devoid of subcutaneous nodules on MRI, subsequent dissection failed to reveal any corpuscles. CONCLUSION: On MRI, normal Pacinian corpuscles appear as round or oval, T2-hyperintense subcutaneous nodules in the palms, clustered around the metacarpophalangeal joints, and should not be mistaken for pathological conditions.

MR imaging of the postsurgical cubital tunnel: an imaging review of the cubital tunnel, cubital tunnel syndrome, and associated surgical techniques.

Ulnar nerve compression at the elbow, specifically the cubital tunnel, is the second most common upper extremity compression neuropathy. Many patients presenting with compression symptoms will subsequently undergo surgical intervention. We review the open surgical treatment of cubital tunnel syndrome and review the expected postoperative imaging appearance of those treatments on magnetic resonance imaging (MR), including: simple or in situ decompression, medial epicondylectomy, and anterior transposition, including subcutaneous, intramuscular, and submuscular variants. We discuss the relevant anatomy of the presurgical cubital tunnel and common sites and causes of ulnar nerve compression at and about the cubital tunnel. The imaging appearance of the preoperative and postoperative ulnar nerve and postoperative complications are reviewed.

MR arthrogram of the postoperative glenoid labrum: normal postoperative appearance versus recurrent tears.

OBJECTIVE: The purpose of this study is to describe the MR arthrogram appearance of the postoperative glenoid labrum and to describe the features consistent with recurrent tear. MATERIALS AND METHODS: We identified 30 patients who had undergone glenoid labral repair, had a subsequent MR arthrogram of his or her shoulder, and went on to repeat shoulder arthroscopy. Each MR arthrogram was reviewed blindly, and the glenoid labrum was described as normal, irregular, or torn. Additional findings recorded included the presence or absence of a paralabral cyst and suture anchors in the glenoid. The operative report was also reviewed for each patient to determine the status of the labrum at arthroscopy. RESULTS: Following consensus review, 18/30 MRIs were felt to demonstrate recurrent glenoid labral tear, 11/30 showed an irregular labrum, and 1/30 was called normal. The radiology impression regarding the presence or absence of a recurrent glenoid labral tear agreed with the operative report in 24/30 (80%) cases, and was discrepant in six. This equals 83.3% sensitivity and 81.8% specificity of MR arthrogram in the diagnosis of recurrent labral tear in this study. A paralabral cyst was present in 3/30 (10%) cases, all three of which were torn. CONCLUSIONS: MR arthrogram findings of signal equal to gadolinium or fluid within or underlying the glenoid labrum and markedly diminutive or absent labrum were the most useful features to diagnose recurrent tear. Some signal underlying the labrum, which is confined to the anterosuperior quadrant, may be normal. The secondary finding of a paralabral cyst was also highly sensitive for recurrent tear.

Pacinian corpuscles: an explanation for subcutaneous palmar nodules routinely encountered on MR examinations.

Pacinian corpuscles, the main touch receptors to pressure and vibration, are ubiquitous in the deep dermis and hypodermis of the fingers and palms. Nevertheless, their existence is largely unknown to most radiologists. We frequently noted hyperintense nodules in the palms of patients on water-sensitive MRI sequences, but were unable to explain their etiology. We recently encountered two patients who had Pacinian corpuscles identified at surgical exploration and pathological analysis. Pre-operative MRI examinations in these patients showed T2 hyperintense subcutaneous palmar nodules corresponding to these corpuscles in a pattern identical to those seen incidentally in other patients. Descriptions from the dermatopathological and orthopedic literature closely correspond to our MRI observations. Based on these data, we hypothesize that the MRI finding that we previously noted represents normal Pacinian corpuscles.

Mismatched perfusion defects secondary to recent methyl methacrylate embolization.

Methyl methacrylate (MMA) pulmonary embolism is a serious potential complication of percutaneous vertebroplasty. We present a case of a 28-year-old man who presented to an outside institution with pleuritic chest pain after MMA vertebroplasty of the T11 thoracic vertebra for chronic back ache secondary to a previous traumatic fracture. Multifocal MMA pulmonary embolism was identified on CT. The patient was referred to our institution following a wedge resection for pulmonary infarction. V/Q scintigraphy was performed and demonstrated normal ventilation with multiple mismatched perfusion defects bilaterally. Patient subsequently underwent central and bilateral segmental pulmonary embolectomies.

Diabetic musculoskeletal complications and their imaging mimics.

Diabetes mellitus is increasingly prevalent and results in various clinically important musculoskeletal disorders affecting the limbs, feet, and spine as well as in widely recognized end-organ complications such as neuropathy, nephropathy, and retinopathy. Diabetic muscle ischemia-a self-limited disorder-may be confused with infectious or inflammatory myositis, venous thrombosis, or compartment syndrome. The absence of fever and leukocytosis, combined with the presence of bilaterally distributed lesions in multiple and often noncontiguous muscles in the legs, including the thighs, is suggestive of ischemia; by contrast, the presence of well-defined intramuscular abscesses with rimlike enhancement favors a diagnosis of infectious pyomyositis. In the diabetic foot, an ulcer, sinus tract, or abscess with an adjacent region of abnormal signal intensity in bone marrow favors the diagnosis of pedal osteomyelitis over that of neuropathic arthropathy. Contrast material-enhanced magnetic resonance imaging is important when planning the treatment of foot infections in diabetic patients because it allows the differentiation of viable tissue from necrotic regions that require surgical débridement in addition to antibiotic therapy. Subtraction images are particularly useful for visualizing nonviable tissue. Dialysis-associated spondyloarthropathy characteristically occurs in diabetic patients with a long history of hemodialysis. Intervertebral disk space narrowing without T2 signal hyperintensity, extensive endplate erosions without endplate remodeling, and facet joint involvement are suggestive of spondyloarthropathy instead of infectious diskitis or degenerative disk disease. Although the clinical features of these conditions may overlap, knowledge of the patient's medical history, coupled with recognition of key imaging characteristics, allows the radiologist to make a prompt and correct diagnosis that leads to appropriate management.