Tan, Yi Liang MBBS, FRCR (UK); Eide, Sterling Ellis MBBS, FRCR (UK); Hallinan, James Thomas Patrick Decourcy BSc (Hons), MBChB (Hons), FRCR (UK) Author Information

OBJECTIVE:There is surging interest in using dual-energy computed tomography (DECT) to identify cardiovascular monosodium urate (MSU) deposits in patients with gout. We sought to examine the prevalence and characterization of cardiovascular DECT artifacts using non-electrocardiogram (EKG)-gated DECT pulmonary angiograms. METHODS:We retrospectively reviewed non-EKG-gated DECT pulmonary angiograms performed on patients with and without gout at a single academic center. We noted the presence and locations of vascular green colorization using the default postprocessing two-material decomposition algorithm for MSU. The high- and low-energy grayscale images and advanced DECT measurements were used to determine whether they were true findings or artifacts. We classified artifacts into five categories: streak, contrast medium mixing, misregistration due to motion, foreign body, and noise. RESULTS:Our study included CT scans from 48 patients with gout and 48 age- and sex-matched controls. The majority of patients were male with a mean age of 67 years. Two independent observers attributed all areas of vascular green colorization to artifacts. The most common types of artifacts were streak (56% vs 57% between patients and controls, respectively) and contrast medium mixing (51% vs 65%, respectively). Whereas some of the default DECT measurements of cardiovascular green colorization were consistent with values reported for subcutaneous tophi, advanced DECT measurements were not consistent with that of tophi. CONCLUSION:Artifacts that could be misconstrued as cardiovascular MSU deposits were commonly identified in patients with and without gout on non-EKG-gated DECT pulmonary angiograms. These artifacts can inform future vascular DECT studies on patients with gout to minimize false-positive findings.

IntroductionMetastatic spinal cord compression (MSCC) is a disastrous complication of advanced malignancy. A deep learning (DL) algorithm for MSCC classification on CT could expedite timely diagnosis. In this study, we externally test a DL algorithm for MSCC classification on CT and compare with radiologist assessment.MethodsRetrospective collection of CT and corresponding MRI from patients with suspected MSCC was conducted from September 2007 to September 2020. Exclusion criteria were scans with instrumentation, no intravenous contrast, motion artefacts and non-thoracic coverage. Internal CT dataset split was 84% for training/validation and 16% for testing. An external test set was also utilised. Internal training/validation sets were labelled by radiologists with spine imaging specialization (6 and 11-years post-board certification) and were used to further develop a DL algorithm for MSCC classification. The spine imaging specialist (11-years expertise) labelled the test sets (reference standard). For evaluation of DL algorithm performance, internal and external test data were independently reviewed by four radiologists: two spine specialists (Rad1 and Rad2, 7 and 5-years post-board certification, respectively) and two oncological imaging specialists (Rad3 and Rad4, 3 and 5-years post-board certification, respectively). DL model performance was also compared against the CT report issued by the radiologist in a real clinical setting. Inter-rater agreement (Gwet’s kappa) and sensitivity/specificity/AUCs were calculated.ResultsOverall, 420 CT scans were evaluated (225 patients, mean age=60 ± 11.9[SD]); 354(84%) CTs for training/validation and 66(16%) CTs for internal testing. The DL algorithm showed high inter-rater agreement for three-class MSCC grading with kappas of 0.872 (p<0.001) and 0.844 (p<0.001) on internal and external testing, respectively. On internal testing DL algorithm inter-rater agreement (κ=0.872) was superior to Rad 2 (κ=0.795) and Rad 3 (κ=0.724) (both p<0.001). DL algorithm kappa of 0.844 on external testing was superior to Rad 3 (κ=0.721) (p<0.001). CT report classification of high-grade MSCC disease was poor with only slight inter-rater agreement (κ=0.027) and low sensitivity (44.0), relative to the DL algorithm with almost-perfect inter-rater agreement (κ=0.813) and high sensitivity (94.0) (p<0.001).ConclusionDeep learning algorithm for metastatic spinal cord compression on CT showed superior performance to the CT report issued by experienced radiologists and could aid earlier diagnosis.

To develop a deep learning (DL) model for epidural spinal cord compression (ESCC) on CT, which will aid earlier ESCC diagnosis for less experienced clinicians. We retrospectively collected CT and MRI data from adult patients with suspected ESCC at a tertiary referral institute from 2007 till 2020. A total of 183 patients were used for training/validation of the DL model. A separate test set of 40 patients was used for DL model evaluation and comprised 60 staging CT and matched MRI scans performed with an interval of up to 2 months. DL model performance was compared to eight readers: one musculoskeletal radiologist, two body radiologists, one spine surgeon, and four trainee spine surgeons. Diagnostic performance was evaluated using inter-rater agreement, sensitivity, specificity and AUC. Overall, 3115 axial CT slices were assessed. The DL model showed high kappa of 0.872 for normal, low and high-grade ESCC (trichotomous), which was superior compared to a body radiologist (R4, κ = 0.667) and all four trainee spine surgeons (κ range = 0.625–0.838)(all p < 0.001). In addition, for dichotomous normal versus any grade of ESCC detection, the DL model showed high kappa (κ = 0.879), sensitivity (91.82), specificity (92.01) and AUC (0.919), with the latter AUC superior to all readers (AUC range = 0.732–0.859, all p < 0.001). A deep learning model for the objective assessment of ESCC on CT had comparable or superior performance to radiologists and spine surgeons. Earlier diagnosis of ESCC on CT could reduce treatment delays, which are associated with poor outcomes, increased costs, and reduced survival.

Objective To evaluate the impact of pre-operative contrast-enhanced mammography (CEM) in breast cancer patients with dense breasts. Methods We conducted a retrospective review of 232 histologically proven breast cancers in 200 women (mean age: 53.4 years ± 10.2) who underwent pre-surgical CEM imaging across two Asian institutions (Singapore and Taiwan). Majority (95.5%) of patients had dense breast tissue (BI-RADS category C or D). Surgical decision was recorded in a simulated blinded multi-disciplinary team setting on two separate scenarios: (i) pre-CEM setting with standard imaging, and clinical and histopathological results; and (ii) post-CEM setting with new imaging and corresponding histological findings from CEM. Alterations in surgical plan (if any) because of CEM imaging were recorded. Predictors CEM of patients who benefitted from surgical plan alterations were evaluated using logistic regression. Results CEM resulted in altered surgical plans in 36 (18%) of 200 patients in this study. CEM discovered clinically significant larger tumor size or extent in 24 (12%) patients and additional tumors in 12 (6%) patients. CEM also detected additional benign/false-positive lesions in 13 (6.5%) of the 200 patients. Significant predictors of patients who benefitted from surgical alterations found on multivariate analysis were pre-CEM surgical decision for upfront breast conservation (OR, 7.7; 95% CI, 1.9-32.1; p = 0.005), architectural distortion on mammograms (OR, 7.6; 95% CI, 1.3–42.9; p = .022), and tumor size of ≥ 1.5 cm (OR, 1.5; 95% CI, 1.0-2.2; p = .034). Conclusion CEM is an effective imaging technique for pre-surgical planning for Asian breast cancer patients with dense breasts. Key Points • CEM significantly altered surgical plans in 18% (nearly 1 in 5) of this Asian study cohort with dense breasts. • Significant patient and imaging predictors for surgical plan alteration include (i) patients considered for upfront breast-conserving surgery; (ii) architectural distortion lesions; and (iii) tumor size of ≥ 1.5 cm. • Additional false-positive/benign lesions detected through CEM were uncommon, affecting only 6.5% of the study cohort.

Background: Metastatic epidural spinal cord compression (MESCC) is a disastrous complication of advanced malignancy. Deep learning (DL) models for automatic MESCC classification on staging CT were developed to aid earlier diagnosis. Methods: This retrospective study included 444 CT staging studies from 185 patients with suspected MESCC who underwent MRI spine studies within 60 days of the CT studies. The DL model training/validation dataset consisted of 316/358 (88%) and the test set of 42/358 (12%) CT studies. Training/validation and test datasets were labeled in consensus by two subspecialized radiologists (6 and 11-years-experience) using the MRI studies as the reference standard. Test sets were labeled by the developed DL models and four radiologists (2–7 years of experience) for comparison. Results: DL models showed almost-perfect interobserver agreement for classification of CT spine images into normal, low, and high-grade MESCC, with kappas ranging from 0.873–0.911 (p < 0.001). The DL models (lowest κ = 0.873, 95% CI 0.858–0.887) also showed superior interobserver agreement compared to two of the four radiologists for three-class classification, including a specialist (κ = 0.820, 95% CI 0.803–0.837) and general radiologist (κ = 0.726, 95% CI 0.706–0.747), both p < 0.001. Conclusion: DL models for the MESCC classification on a CT showed comparable to superior interobserver agreement to radiologists and could be used to aid earlier diagnosis.

Background Lumbar spine MRI studies are widely used for back pain assessment. Interpretation involves grading lumbar spinal stenosis, which is repetitive and time consuming. Deep learning (DL) could provide faster and more consistent interpretation. Purpose To assess the speed and interobserver agreement of radiologists for reporting lumbar spinal stenosis with and without DL assistance. Materials and Methods In this retrospective study, a DL model designed to assist radiologists in the interpretation of spinal canal, lateral recess, and neural foraminal stenoses on lumbar spine MRI scans was used. Randomly selected lumbar spine MRI studies obtained in patients with back pain who were 18 years and older over a 3-year period, from September 2015 to September 2018, were included in an internal test data set. Studies with instrumentation and scoliosis were excluded. Eight radiologists, each with 2-13 years of experience in spine MRI interpretation, reviewed studies with and without DL model assistance with a 1-month washout period. Time to diagnosis (in seconds) and interobserver agreement (using Gwet κ) were assessed for stenosis grading for each radiologist with and without the DL model and compared with test data set labels provided by an external musculoskeletal radiologist (with 32 years of experience) as the reference standard. Results Overall, 444 images in 25 patients (mean age, 51 years ± 20 [SD]; 14 women) were evaluated in a test data set. DL-assisted radiologists had a reduced interpretation time per spine MRI study, from a mean of 124-274 seconds (SD, 25-88 seconds) to 47-71 seconds (SD, 24-29 seconds) (P < .001). DL-assisted radiologists had either superior or equivalent interobserver agreement for all stenosis gradings compared with unassisted radiologists. DL-assisted general and in-training radiologists improved their interobserver agreement for four-class neural foraminal stenosis, with κ values of 0.71 and 0.70 (with DL) versus 0.39 and 0.39 (without DL), respectively (both P < .001). Conclusion Radiologists who were assisted by deep learning for interpretation of lumbar spinal stenosis on MRI scans showed a marked reduction in reporting time and superior or equivalent interobserver agreement for all stenosis gradings compared with radiologists who were unassisted by deep learning. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Hayashi in this issue.

A 37-year-old man presented with a history of sudden-onset neck pain, which was worse on movement and neck extension. He reported no history of trauma or infection. He denied any neurological symptoms. A physical examination showed limited neck motion in all directions due to pain. Blood tests showed total white blood cell count to be

Background Assessment of lumbar spinal stenosis at MRI is repetitive and time consuming. Deep learning (DL) could improve -productivity and the consistency of reporting. Purpose To develop a DL model for automated detection and classification of lumbar central canal, lateral recess, and neural -foraminal stenosis. Materials and Methods In this retrospective study, lumbar spine MRI scans obtained from September 2015 to September 2018 were included. Studies of patients with spinal instrumentation or studies with suboptimal image quality, as well as postgadolinium studies and studies of patients with scoliosis, were excluded. Axial T2-weighted and sagittal T1-weighted images were used. Studies were split into an internal training set (80%), validation set (9%), and test set (11%). Training data were labeled by four radiologists using predefined gradings (normal, mild, moderate, and severe). A two-component DL model was developed. First, a convolutional neural network (CNN) was trained to detect the region of interest (ROI), with a second CNN for classification. An internal test set was labeled by a musculoskeletal radiologist with 31 years of experience (reference standard) and two subspecialist radiologists (radiologist 1: A.M., 5 years of experience; radiologist 2: J.T.P.D.H., 9 years of experience). DL model performance on an external test set was evaluated. Detection recall (in percentage), interrater agreement (Gwet κ), sensitivity, and specificity were calculated. Results Overall, 446 MRI lumbar spine studies were analyzed (446 patients; mean age ± standard deviation, 52 years ± 19; 240 women), with 396 patients in the training (80%) and validation (9%) sets and 50 (11%) in the internal test set. For internal testing, DL model and radiologist central canal recall were greater than 99%, with reduced neural foramina recall for the DL model (84.5%) and radiologist 1 (83.9%) compared with radiologist 2 (97.1%) (P < .001). For internal testing, dichotomous classification (normal or mild vs moderate or severe) showed almost-perfect agreement for both radiologists and the DL model, with respective κ values of 0.98, 0.98, and 0.96 for the central canal; 0.92, 0.95, and 0.92 for lateral recesses; and 0.94, 0.95, and 0.89 for neural foramina (P < .001). External testing with 100 MRI scans of lumbar spines showed almost perfect agreement for the DL model for dichotomous classification of all ROIs (κ, 0.95-0.96; P < .001). Conclusion A deep learning model showed comparable agreement with subspecialist radiologists for detection and classification of central canal and lateral recess stenosis, with slightly lower agreement for neural foraminal stenosis at lumbar spine MRI. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Hayashi in this issue.

Algazwi, Diyaa Abdul Rauf MBBS, EDiR; Tian, Quek Swee MBBS, FRCR; Elizabeth, Gartner Louise MBBS, MRCP, FRCR; Ellis, Eide Sterling MBBS, FRCR; Teng, Victor Sheng Yeow MBChB, FRCR; Hallinan, James Thomas Patrick Decourcy MBChB, FRCR Author Information

PURPOSE: To determine the accuracy of a handheld ultrasound-guided optoacoustic tomography (US-OT) probe developed for human deep-tissue imaging in ex vivo assessment of tumor margins postlumpectomy. METHODS: A custom-built two-dimensional (2D) US-OT-handheld probe was used to scan 15 lumpectomy breast specimens. Optoacoustic signals acquired at multiple wavelengths between 700 and 1100 nm were reconstructed using model linear algorithm, followed by spectral unmixing for lipid and deoxyhemoglobin (Hb). Distribution maps of lipid and Hb on the anterior, posterior, superior, inferior, medial, and lateral margins of the specimens were inspected for margin involvement, and results were correlated with histopathologic findings. The agreement in tumor margin assessment between US-OT and histopathology was determined using the Bland-Altman plot. Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of margin assessment using US-OT were calculated. RESULTS: Ninety margins (6 x 15 specimens) were assessed. The US-OT probe resolved blood vessels and lipid up to a depth of 6 mm. Negative and positive margins were discriminated by marked differences in the distribution patterns of lipid and Hb. US-OT assessments were concordant with histopathologic findings in 87 of 89 margins assessed (one margin was uninterpretable and excluded), with diagnostic accuracy of 97.9% (kappa 0.79). The sensitivity, specificity, PPV, and NPV were 100% (4/4), 97.6% (83/85), 66.7% (4/6), and 100% (83/83), respectively. CONCLUSION: US-OT was capable of providing distribution maps of lipid and Hb in lumpectomy specimens that predicted tumor margins with high sensitivity and specificity, making it a potential tool for intraoperative tumor margin assessment.

Osteoblastoma is a rare, benign primary tumor of bone, accounting for < 1% of all bone tumors. We report the case of a 27-year-old female who developed pain and swelling five and a half years after a clavicular fracture and was subsequently found to have an osteoblastoma arising at the fracture site. This is the first reported case of an osteoblastoma developing after a fracture, although osteoid osteomas, which are histologically indistinguishable from osteoblastomas, have been reported at prior fracture sites. This report demonstrates that secondary neoplasms such as osteoblastomas should be considered in the differential diagnosis for pain at a healed fracture site recurring years after the initial trauma.

BACKGROUND:Patients with initial transient ischaemic attack (TIA) subsequently have a higher risk of recurrent TIA or acute ischemic stroke (AIS). The role of scoring intracranial arterial calcification (IAC) in predicting the prevalence of stroke remains unclear. We aim to evaluate if radiological CT calcium score measuring IAC burden could predict future ischemic events in a cohort of TIA patients.METHODS:We studied consecutive patients from July 2014 to December 2015 who presented with first episode of TIA. All patients had noncontrasted CT or CT-angiogram of the brain on admission. CT calcium score (cm3) was quantified by measuring calcium deposition in the bilateral internal carotid arteries, middle cerebral arteries, and vertebrobasilar system. Patients were followed up for 2 years and ischemic events for either recurrent TIA or AIS were recorded. We compared patients in terms of clinical profile at presentation and CT calcium score using appropriate univariate and multivariable analyses.RESULTS:Of 156 TIA patients studied, 22% (n = 35) had recurrent TIA or AIS within 2 years of follow-up. On univariate analyses, recurrent TIA/AIS was associated with gender (OR 0.61; 95%CI 0.40-0.95; P = .038), hypertension (mean difference 2.49; 95%CI 1.08-5.75; P = .030) and higher CT calcium score (mean difference 0.84 95%CI 0.16-1.52 P = .016). On multivariable logistic regression, a higher CT calcium score was significantly associated with recurrent TIA/AIS (adjusted OR 1.25 95%CI 1.01-1.55 P = .042).CONCLUSIONS:In TIA patients, higher IAC burden by measurement of a quantitative CT calcium score may be associated with recurrent ischemic events.

Objective: With increased access to computed tomography(CT) scans, we aimed to evaluate if a radiological CT brain calcium score predicts future transient ischemic attack (TIA) or stroke. Background: Patients with TIA are at higher risks of having a recurrent TIA or stroke. Various clinical scores have been developed to predict stroke risk after TIA. The amount of coronary calcification detected by electrobeam CT is an established prognostic tool for coronary artery disease, but this has not been investigated conclusively for stroke. Design/Methods: Consecutive patients admitted to our stroke centre from January to December 2015 with first presentation of TIA were studied. CT calcium score (cm3) was quantified by measuring calcium deposition in bilateral internal carotids, middle cerebral arteries and vertebrobasilar system. Patients were followed up for 2 years and readmissions for TIA or stroke were recorded as primary outcome. We compared patients in terms of clinical profile at presentation and CT calcium score using appropriate univariate and multivariable analyses. Results: Of the 154 patients studied, 18.1% (n=28) had recurrent TIA or stroke within 2 years. These patients tended to have higher HbA1c (6.7±2.0vs6.2±1.6%, p=0.367) but lower corrected LDL levels (2.45±1.17vs2.80±1.14, p=0.264), but these were not statistically significant. CT brain calcium score was significantly higher in the group with recurrent TIA or stroke (1.37±2.13 vs 0.50±1.11 cm3, p=0.010). Even after adjusting for HbA1c and corrected LDL levels, the CT brain calcium score remained an independent predictor of stroke or TIA recurrence (adjusted OR 1.451, 95%CI 1.054–1.900, p=0.021). Conclusions: CT brain calcium score may prove useful as an objective radiological marker in predicting recurrent TIA or stroke in patients with TIA, with less inter-observer variation. Further studies are warranted for validation and comparison against existing clinical scores. Disclosure: Dr. Kong has nothing to disclose. Dr. Sterling has nothing to disclose. Dr. Ngiam has nothing to disclose. Dr. Yuan has nothing to disclose. Dr. Tan has nothing to disclose. Dr. Ong has nothing to disclose. Dr. Sibi has nothing to disclose. Dr. Tan has nothing to disclose. Dr. Sharma has nothing to disclose. Dr. Chan has nothing to disclose. Dr. Yeo has nothing to disclose.

Multispectral optoacoustic tomography (MSOT) is an innovative state-of-the-art imaging modality that has gained popularity for in vivo breast imaging in recent years. Combining high-resolution images with endogenous differentiation of biochemical contents, MSOT could be an accurate ex vivo imaging modality for assessment of tumor margins during breast-conserving surgery to reduce margin positivity rates. Accurate intraoperative assessment of margins could lead to more precise surgery, allowing a smaller volume of breast tissue to be removed without compromising the resection margins. To the authors' knowledge, there has been no ex vivo study conducted to this day that uses MSOT in the assessment of breast tumor margins after lumpectomy. Hence, we would like to present the first case of breast tumor margin assessment using MSOT in a 55-year-old patient who underwent breast-conserving surgery for invasive ductal carcinoma. (C) 2018 The Authors. Published by Elsevier Inc.

Background and Purpose— We assessed the feasibility of obtaining diagnostic quality images of the heart and thoracic aorta by extending the z axis coverage of a non–ECG-gated computed tomographic angiogram performed in the primary evaluation of acute stroke without increasing the contrast dose. Methods— Twenty consecutive patients with acute ischemic stroke within the 4.5 hours of symptom onset were prospectively recruited. We increased the longitudinal coverage to the domes of the diaphragm to include the heart. Contrast administration (Omnipaque 350) remained unchanged (injected at 3–4 mL/s; total 60–80 mL, triggered by bolus tracking). Images of the heart and aorta, reconstructed at 5 mm slice thickness in 3 orthogonal planes, were read by a radiologist and cardiologist, findings conveyed to the treating neurologist, and correlated with the transthoracic or transesophageal echocardiogram performed within the next 24 hours. Results— Of 20 patients studied, 3 (15%) had abnormal findings: a left ventricular thrombus, a Stanford type A aortic dissection, and a thrombus of the left atrial appendage. Both thrombi were confirmed by transesophageal echocardiography, and anticoagulation was started urgently the following day. None of the patients developed contrast-induced nephropathy on follow-up. The radiation dose was slightly increased from a mean of 4.26 mSV (range, 3.88–4.70 mSV) to 5.17 (range, 3.95 to 6.25 mSV). Conclusions— Including the heart and ascending aorta in a routine non–ECG-gated computed tomographic angiogram enhances an existing imaging modality, with no increased incidence of contrast-induced nephropathy and minimal increase in radiation dose. This may help in the detection of high-risk cardiac and aortic sources of embolism in acute stroke patients.