Radiologists In Clinical Practice Research Articles

Automated classification of liver fibrosis stages using ultrasound imaging.

Ultrasound imaging is the most frequently performed for the patients with chronic hepatitis or liver cirrhosis. However, ultrasound imaging is highly operator dependent and interpretation of ultrasound images is subjective, thus well-trained radiologist is required for evaluation. Automated classification of liver fibrosis could alleviate the shortage of skilled radiologist especially in low-to-middle income countries. The purposed of this study is to evaluate deep convolutional neural networks (DCNNs) for classifying the degree of liver fibrosis according to the METAVIR score using US images. We used ultrasound (US) images from two tertiary university hospitals. A total of 7920 US images from 933 patients were used for training/validation of DCNNs. All patient were underwent liver biopsy or hepatectomy, and liver fibrosis was categorized based on pathology results using the METAVIR score. Five well-established DCNNs (VGGNet, ResNet, DenseNet, EfficientNet and ViT) was implemented to predict the METAVIR score. The performance of DCNNs for five-level (F0/F1/F2/F3/F4) classification was evaluated through area under the receiver operating characteristic curve (AUC) with 95% confidential interval, accuracy, sensitivity, specificity, positive and negative likelihood ratio. Similar mean AUC values were achieved for five models; VGGNet (0.96), ResNet (0.96), DenseNet (0.95), EfficientNet (0.96), and ViT (0.95). The same mean accuracy (0.94) and specificity values (0.96) were yielded for all models. In terms of sensitivity, EffcientNet achieved highest mean value (0.85) while the other models produced slightly lower values range from 0.82 to 0.84. In this study, we demonstrated that DCNNs can classify the staging of liver fibrosis according to METAVIR score with high performance using conventional B-mode images. Among them, EfficientNET that have fewer parameters and computation cost produced highest performance. From the results, we believe that DCNNs based classification of liver fibrosis may allow fast and accurate diagnosis of liver fibrosis without needs of additional equipment for add-on test and may be powerful tool for supporting radiologists in clinical practice.

Evaluation of American College of Radiology Ovarian-Adnexal Reporting and Data System ultrasound to predict malignancy risk in adnexal lesions.

To validate the diagnostic performance of Ovarian-Adnexal Reporting and Data System (O-RADS) ultrasound for preoperative adnexal lesions in an external center. The secondary aim was to evaluate the performance of a strategy test including O-RADS ultrasound evaluation and subjective assessment of higher malignant risk lesions. One hundred thirty patients with 158 ovarian-adnexal lesions were enrolled in the study. Each lesion was assigned an O-RADS score after real-time ultrasound examination by one experienced radiologist. A second subjective assessment by an expert was performed for O-RADS 4 and O-RADS 5 lesions. The histopathological diagnosis was used as the reference standard. A total of 126 benign and 32 malignant adnexal masses were included in the study. The area under the receiver operating characteristic curve of O-RADS ultrasound was 0.950, with a cutoff value > O-RADS 3. The sensitivity, specificity, and negative and positive predictive values were 100% (95% confidence interval [CI], 0.867-1), 83.3% (95% CI, 0.754-0.892), 60.4% (95% CI, 0.460-0.732), and 100% (95% CI, 0.956-1), respectively. For the strategy test, the sensitivity, specificity, negative and positive predictive values were 100% (95% CI, 0.867-1), 92.1% (95% CI, 0.855-0.959), 76.2% (95% CI, 0.602-0.874), and 100% (95% CI, 0.960-1), respectively. In comparison with O-RADS ultrasound, the specificity and negative predictive value of the strategy test were slightly higher (p < 0.05). Good diagnostic performance of the O-RADS ultrasound in adnexal lesions can be achieved by experienced radiologists in clinical practice. A second subjective assessment of sonographic findings can be applied to O-RADS 4 and 5 lesions.

Establishment and validation of an AI-aid method in the diagnosis of myocardial perfusion imaging

BackgroundThis study aimed to develop and validate an AI (artificial intelligence)-aid method in myocardial perfusion imaging (MPI) to differentiate ischemia in coronary artery disease.MethodsWe retrospectively selected 599 patients who had received gated-MPI protocol. Images were acquired using hybrid SPECT-CT systems. A training set was used to train and develop the neural network and a validation set was used to test the predictive ability of the neural network. We used a learning technique named “YOLO” to carry out the training process. We compared the predictive accuracy of AI with that of physician interpreters (beginner, inexperienced, and experienced interpreters).ResultsTraining performance showed that the accuracy ranged from 66.20% to 94.64%, the recall rate ranged from 76.96% to 98.76%, and the average precision ranged from 80.17% to 98.15%. In the ROC analysis of the validation set, the sensitivity range was 88.9 ~ 93.8%, the specificity range was 93.0 ~ 97.6%, and the AUC range was 94.1 ~ 96.1%. In the comparison between AI and different interpreters, AI outperformed the other interpreters (most P-value < 0.05).ConclusionThe AI system of our study showed excellent predictive accuracy in the diagnosis of MPI protocols, and therefore might be potentially helpful to aid radiologists in clinical practice and develop more sophisticated models.

Review of the Pathogenesis, Diagnosis, and Management of Osteoradionecrosis of the Femoral Head.

Osteoradionecrosis (ORN) of the femoral head is an important issue for orthopedists and radiologists in clinical practice. With the rapid development of technological advances in radiation therapy and the improvement in cancer survival rates, the incidence of ORN is rising, and there is an unmet need for basic and clinical research. The pathogenesis of ORN is complex, and includes vascular injury, mesenchymal stem cell injury, bone loss, reactive oxygen species, radiation-induced fibrosis, and cell senescence. The diagnosis of ORN is challenging and requires multiple considerations, including exposure to ionizing radiation, clinical manifestations, and findings on physical examination and imaging. Differential diagnosis is essential, as clinical symptoms of ORN of the femoral head can resemble many other hip conditions. Hyperbaric oxygen therapy, total hip arthroplasty, and Girdlestone resection arthroplasty are effective treatments, each with their own advantages and disadvantages. The literature on ORN of the femoral head is incomplete and there is no criterion standard or clear consensus on management. Clinicians should gain a better and more comprehensive understanding on this disease to facilitate its early and better prevention, diagnosis, and treatment. This article aims to review the pathogenesis, diagnosis, and management of osteoradionecrosis of the femoral head.

Deep learning for automated, interpretable classification of lumbar spinal stenosis and facet arthropathy from axial MRI.

To evaluate a deep learning model for automated and interpretable classification of central canal stenosis, neural foraminal stenosis, and facet arthropathy from lumbar spine MRI. T2-weighted axial MRI studies of the lumbar spine acquired between 2008 and 2019 were retrospectively selected (n = 200) and graded for central canal stenosis, neural foraminal stenosis, and facet arthropathy. Studies were partitioned into patient-level train (n = 150), validation (n = 20), and test (n = 30) splits. V-Net models were first trained to segment the dural sac and the intervertebral disk, and localize facet and foramen using geometric rules. Subsequently, Big Transfer (BiT) models were trained for downstream classification tasks. An interpretable model for central canal stenosis was also trained using a decision tree classifier. Evaluation metrics included linearly weighted Cohen's kappa score for multi-grade classification and area under the receiver operator characteristic curve (AUROC) for binarized classification. Segmentation of the dural sac and intervertebral disk achieved Dice scores of 0.93 and 0.94. Localization of foramen and facet achieved intersection over union of 0.72 and 0.83. Multi-class grading of central canal stenosis achieved a kappa score of 0.54. The interpretable decision tree classifier had a kappa score of 0.80. Pairwise agreement between readers (R1, R2), (R1, R3), and (R2, R3) was 0.86, 0.80, and 0.74. Binary classification of neural foraminal stenosis and facet arthropathy achieved AUROCs of 0.92 and 0.93. Deep learning systems can be performant as well as interpretable for automated evaluation of lumbar spine MRI including classification of central canal stenosis, neural foraminal stenosis, and facet arthropathy. • Interpretable deep-learning systems can be developed for the evaluation of clinical lumbar spine MRI. Multi-grade classification of central canal stenosis with a kappa of 0.80 was comparable to inter-reader agreement scores (0.74, 0.80, 0.86). Binary classification of neural foraminal stenosis and facet arthropathy achieved favorable and accurate AUROCs of 0.92 and 0.93, respectively. • While existing deep-learning systems are opaque, leading to clinical deployment challenges, the proposed system is accurate as well as interpretable, providing valuable information to a radiologist in clinical practice.

Cardiovascular CT, MRI, and PET/CT in 2021: Review of Key Articles.

This review focuses on three key noninvasive cardiac imaging modalities-cardiac CT angiography (CTA), MRI, and PET/CT-and summarizes key publications in 2021 relevant to radiologists in clinical practice. Although this review focuses primarily on articles published in Radiology, important studies from other major journals are included to highlight "must-know" articles in the field of cardiovascular imaging. Cardiac CTA has been established as the first-line test for patients with stable chest pain and no known coronary artery disease, and its value remains central to the assessment of surgical or transcatheter aortic valve replacement. Artificial intelligence continues to evolve in a number of applications in cardiovascular disease. In cardiac MRI studies, 2021 has seen an emphasis on nonischemic cardiomyopathies, valvular heart disease, and COVID-19 disease cardiac manifestations and the authors highlight the key articles on these topics. A section featuring the increasing role of cardiac PET/CT in the assessment of cardiac sarcoidosis and prosthetic valves is also provided.

Multiscale unsupervised domain adaptation for automatic pancreas segmentation in CT volumes using adversarial learning.

Computer-aided automatic pancreas segmentation is essential for early diagnosis and treatment of pancreatic diseases. However, the annotation of pancreas images requires professional doctors and considerable expenditure. Due to imaging differences among various institution population, scanning devices, imaging protocols, and so on, significant degradation in the performance of model inference results is prone to occur when models trained with domain-specific (usually institution-specific) datasets are directly applied to new (other centers/institutions) domain data. In this paper, we propose a novel unsupervised domain adaptation method based on adversarial learning to address pancreas segmentation challenges with the lack of annotations and domain shift interference. A 3D semantic segmentation model with attention module and residual module is designed as the backbone pancreas segmentation model. In both segmentation model and domain adaptation discriminator network, a multiscale progressively weighted structure is introduced to acquire different field of views. Features of labeled data and unlabeled data are fed in pairs into the proposed multiscale discriminator to learn domain-specific characteristics. Then the unlabeled data features with pseudodomain label are fed to the discriminator to acquire domain-ambiguous information. With this adversarial learning strategy, the performance of the segmentation network is enhanced to segment unseen unlabeleddata. Experiments were conducted on two public annotated datasets as source datasets, respectively, and one private dataset as target dataset, where annotations were not used for the training process but only for evaluation. The 3D segmentation model achieves comparative performance with state-of-the-art pancreas segmentation methods on source domain. After implementing our domain adaptation architecture, the average dice similarity coefficient (DSC) of the segmentation model trained on the NIH-TCIA source dataset increases from 58.79% to 72.73% on the local hospital dataset, while the performance of the target domain segmentation model transferred from the medical segmentation decathlon (MSD) source dataset rises from 62.34% to 71.17%. Correlations of features across data domains are utilized to train the pancreas segmentation model on unlabeled data domain, improving the generalization of the model. Our results demonstrate that the proposed method enables the segmentation model to make meaningful segmentation for unseen data of the training set. In the future, the proposed method has the potential to apply segmentation model trained on public dataset to clinical unannotated CT images from local hospital, effectively assisting radiologists in clinicalpractice.

Development of a Deep Learning System to Detect Esophageal Cancer by Barium Esophagram.

BackgroundImplementation of deep learning systems (DLSs) for analysis of barium esophagram, a cost-effective diagnostic test for esophageal cancer detection, is expected to reduce the burden to radiologists while ensuring the accuracy of diagnosis.ObjectiveTo develop an automated DLS to detect esophageal cancer on barium esophagram.MethodsThis was a retrospective study using deep learning for esophageal cancer detection. A two-stage DLS (including a Selection network and a Classification network) was developed. Five datasets based on barium esophagram were used for stepwise training, validation, and testing of the DLS. Datasets 1 and 2 were used to respectively train and test the Selection network, while Datasets 3, 4, and 5 were respectively used to train, validate, and test the Classification network. Finally, a positioning box with a probability value was outputted by the DLS. A region of interest delineated by experienced radiologists was selected as the ground truth to evaluate the detection and classification efficiency of the DLS. Standard machine learning metrics (accuracy, recall, precision, sensitivity, and specificity) were calculated. A comparison with the conventional visual inspection approach was also conducted.ResultsThe accuracy, sensitivity, and specificity of our DLS in detecting esophageal cancer were 90.3%, 92.5%, and 88.7%, respectively. With the aid of DLS, the radiologists’ interpretation time was significantly shortened (Reader1, 45.7 s vs. 72.2 s without DLS aid; Reader2, 54.1 s vs. 108.7 s without DLS aid). Respective diagnostic efficiencies for Reader1 with and without DLS aid were 96.8% vs. 89.3% for accuracy, 97.5% vs. 87.5% for sensitivity, 96.2% vs. 90.6% for specificity, and 0.969 vs. 0.890 for AUC. Respective diagnostic efficiencies for Reader2 with and without DLS aid were 95.7% vs. 88.2% for accuracy, 92.5% vs. 77.5% for sensitivity, 98.1% vs. 96.2% for specificity, and 0.953 vs. 0.869 for AUC. Of note, the positioning boxes outputted by the DLS almost overlapped with those manually labeled by the radiologists on Dataset 5.ConclusionsThe proposed two-stage DLS for detecting esophageal cancer on barium esophagram could effectively shorten the interpretation time with an excellent diagnostic performance. It may well assist radiologists in clinical practice to reduce their burden.

Automated Prediction of the Response to Neoadjuvant Chemoradiotherapy in Patients Affected by Rectal Cancer.

Simple SummaryColorectal cancer is the second most malignant tumor per number of deaths after lung cancer and the third per number of new cases after breast and lung cancer. The correct and rapid identification (i.e., segmentation of the cancer regions) is a fundamental task for correct patient diagnosis. In this study, we propose a novel automated pipeline for the segmentation of MRI scans of patients with LARC in order to predict the response to nCRT using radiomic features. This study involved the retrospective analysis of T2-weighted MRI scans of 43 patients affected by LARC. The segmentation of tumor areas was on par or better than the state-of-the-art results, but required smaller sample sizes. The analysis of radiomic features allowed us to predict the TRG score, which agreed with the state-of-the-art results.Background: Rectal cancer is a malignant neoplasm of the large intestine resulting from the uncontrolled proliferation of the rectal tract. Predicting the pathologic response of neoadjuvant chemoradiotherapy at an MRI primary staging scan in patients affected by locally advanced rectal cancer (LARC) could lead to significant improvement in the survival and quality of life of the patients. In this study, the possibility of automatizing this estimation from a primary staging MRI scan, using a fully automated artificial intelligence-based model for the segmentation and consequent characterization of the tumor areas using radiomic features was evaluated. The TRG score was used to evaluate the clinical outcome. Methods: Forty-three patients under treatment in the IRCCS Sant’Orsola-Malpighi Polyclinic were retrospectively selected for the study; a U-Net model was trained for the automated segmentation of the tumor areas; the radiomic features were collected and used to predict the tumor regression grade (TRG) score. Results: The segmentation of tumor areas outperformed the state-of-the-art results in terms of the Dice score coefficient or was comparable to them but with the advantage of considering mucinous cases. Analysis of the radiomic features extracted from the lesion areas allowed us to predict the TRG score, with the results agreeing with the state-of-the-art results. Conclusions: The results obtained regarding TRG prediction using the proposed fully automated pipeline prove its possible usage as a viable decision support system for radiologists in clinical practice.

Fully Automated Thrombus Segmentation on CT Images of Patients with Acute Ischemic Stroke.

Thrombus imaging characteristics are associated with treatment success and functional outcomes in stroke patients. However, assessing these characteristics based on manual annotations is labor intensive and subject to observer bias. Therefore, we aimed to create an automated pipeline for consistent and fast full thrombus segmentation. We used multi-center, multi-scanner datasets of anterior circulation stroke patients with baseline NCCT and CTA for training (n = 228) and testing (n = 100). We first found the occlusion location using StrokeViewer LVO and created a bounding box around it. Subsequently, we trained dual modality U-Net based convolutional neural networks (CNNs) to segment the thrombus inside this bounding box. We experimented with: (1) U-Net with two input channels for NCCT and CTA, and U-Nets with two encoders where (2) concatenate, (3) add, and (4) weighted-sum operators were used for feature fusion. Furthermore, we proposed a dynamic bounding box algorithm to adjust the bounding box. The dynamic bounding box algorithm reduces the missed cases but does not improve Dice. The two-encoder U-Net with a weighted-sum feature fusion shows the best performance (surface Dice 0.78, Dice 0.62, and 4% missed cases). Final segmentation results have high spatial accuracies and can therefore be used to determine thrombus characteristics and potentially benefit radiologists in clinical practice.

Cancer Risk in Nodules Detected at Follow-Up Lung Cancer Screening CT.

BACKGROUND. Nodules may have different lung cancer risks when new on follow-up CT versus when present on previously performed CT (i.e., existing nodules). Diameter-based Lung-RADS and volume-based NELSON (Nederlands-Leuvens Longkanker Screenings ONderzoek trial) categories have shown variable performance in nodule risk assessment. OBJECTIVE. The purpose of this study was to assess Lung-RADS and NELSON classifications of nodules detected on follow-up lung cancer screening CT examinations. METHODS. This retrospective study included 185 patients (100 women and 85 men; median age, 66 years) who underwent a lung cancer screening CT examination for which a prior CT examination was available. Stratified random sampling was performed to enrich the sample with suspicious nodules, yielding 50, 45, 47, 30, and 13 nodules with Lung-RADS categories 2, 3, 4A, 4B, and 4X, respectively. Lung-RADS categories were recorded from clinical reports. The linear measurements of the nodules were extracted from clinical reports to generate Lung-RADS categories by use of strict criteria from Lung-RADS version 1.1. Two radiologists used a semiautomated tool to obtain nodule volumes, which were used to generate NELSON categories. Lung cancer risk was assessed. ROC analysis was performed. Percentages and AUCs were weighted on the basis of Lung-RADS category frequencies in the underlying screening cohort. RESULTS. Twenty-nine cancers were diagnosed. The weighted cancer risk was 5% for new nodules, 1% for stable existing nodules, and 44% for growing existing nodules. None of the clinical Lung-RADS category 2 nodules were cancer. With use of strict Lung-RADS version 1.1 criteria, 34 nodules, including seven cancers, were downgraded to category 2. The AUC for cancer was 0.96 for clinical Lung-RADS, 0.81 for strict Lung-RADS, 0.71-0.84 for the NELSON algorithm (two readers), and 0.89 for nodule diameter measurement. Clinical Lung-RADS achieved weighted sensitivity and specificity, respectively, of 100% and 85% for the entire sample, 100% and 41% for new nodules, and 100% and 94% for existing nodules. The optimal diameter threshold was 8 mm for existing nodules versus 6 mm for new nodules. CONCLUSION. Lung-RADS, as applied by radiologists in clinical practice, achieved excellent performance on follow-up screening examinations. Strict Lung-RADS resulted in the downgrading of some cancers to category 2. Volumetric assessments had weaker performance than clinical Lung-RADS. New nodules warrant smaller size thresholds than existing nodules. CLINICAL IMPACT. The findings of the present study provide insight into radiologists' management of nodules detected on follow-up screening examinations.

An Accurate and Explainable Deep Learning System Improves Interobserver Agreement in the Interpretation of Chest Radiograph

Interpretation of chest radiographs (CXR) is a difficult but essential task for detecting thoracic abnormalities. Recent artificial intelligence (AI) algorithms have achieved radiologist-level performance on various medical classification tasks. However, only a few studies addressed the localization of abnormal findings from CXR scans, which is essential in explaining the image-level classification to radiologists. Additionally, the actual impact of AI algorithms on the diagnostic performance of radiologists in clinical practice remains relatively unclear. To bridge these gaps, we developed an explainable deep learning system called VinDr-CXR that can classify a CXR scan into multiple thoracic diseases and, at the same time, localize most types of critical findings on the image. VinDr-CXR was trained on 51,485 CXR scans with radiologist-provided bounding box annotations. It demonstrated a comparable performance to experienced radiologists in classifying 6 common thoracic diseases on a retrospective validation set of 3,000 CXR scans, with a mean area under the receiver operating characteristic curve (AUROC) of 0.967 (95% confidence interval [CI]: 0.958–0.975). The VinDr-CXR was also externally validated in independent patient cohorts and showed its robustness. For the localization task with 14 types of lesions, our free-response receiver operating characteristic (FROC) analysis showed that the VinDr-CXR achieved a sensitivity of 80.2% at the rate of 1.0 false-positive lesion identified per scan. A prospective study was also conducted to measure the clinical impact of the VinDr-CXR in assisting six experienced radiologists. The results indicated that the proposed system, when used as a diagnosis supporting tool, significantly improved the agreement between radiologists themselves with an increase of 1.5% in mean Fleiss’ Kappa. We also observed that, after the radiologists consulted VinDr-CXR’s suggestions, the agreement between each of them and the system was remarkably increased by 3.3% in mean Cohen’s Kappa. Altogether, our results highlight the potentials of the proposed deep learning system as an effective assistant to radiologists in clinical practice. Part of the dataset used for developing the VinDr-CXR system has been made publicly available at https://physionet.org/content/vindr-cxr/1.0.0/.

How artificial intelligence improves radiological interpretation in suspected pulmonary embolism.

ObjectivesTo evaluate and compare the diagnostic performances of a commercialized artificial intelligence (AI) algorithm for diagnosing pulmonary embolism (PE) on CT pulmonary angiogram (CTPA) with those of emergency radiologists in routine clinical practice.MethodsThis was an IRB-approved retrospective multicentric study including patients with suspected PE from September to December 2019 (i.e., during a preliminary evaluation period of an approved AI algorithm). CTPA quality and conclusions by emergency radiologists were retrieved from radiological reports. The gold standard was a retrospective review of CTPA, radiological and clinical reports, AI outputs, and patient outcomes. Diagnostic performance metrics for AI and radiologists were assessed in the entire cohort and depending on CTPA quality.ResultsOverall, 1202 patients were included (median age: 66.2 years). PE prevalence was 15.8% (190/1202). The AI algorithm detected 219 suspicious PEs, of which 176 were true PEs, including 19 true PEs missed by radiologists. In the cohort, the highest sensitivity and negative predictive values (NPVs) were obtained with AI (92.6% versus 90% and 98.6% versus 98.1%, respectively), while the highest specificity and positive predictive value (PPV) were found with radiologists (99.1% versus 95.8% and 95% versus 80.4%, respectively). Accuracy, specificity, and PPV were significantly higher for radiologists except in subcohorts with poor-to-average injection quality. Radiologists positively evaluated the AI algorithm to improve their diagnostic comfort (55/79 [69.6%]).ConclusionInstead of replacing radiologists, AI for PE detection appears to be a safety net in emergency radiology practice due to high sensitivity and NPV, thereby increasing the self-confidence of radiologists.Key Points • Both the AI algorithm and emergency radiologists showed excellent performance in diagnosing PE on CTPA (sensitivity and specificity ≥ 90%; accuracy ≥ 95%). • The AI algorithm for PE detection can help increase the sensitivity and NPV of emergency radiologists in clinical practice, especially in cases of poor-to-moderate injection quality. • Emergency radiologists recommended the use of AI for PE detection in satisfaction surveys to increase their confidence and comfort in their final diagnosis. Supplementary InformationThe online version contains supplementary material available at 10.1007/s00330-022-08645-2.

Detection and PI-RADS classification of focal lesions in prostate MRI: Performance comparison between a deep learning-based algorithm (DLA) and radiologists with various levels of experience

PurposeTo compare the performance of lesion detection and Prostate Imaging-Reporting and Data System (PI-RADS) classification between a deep learning-based algorithm (DLA), clinical reports and radiologists with different levels of experience in prostate MRI. MethodsThis retrospective study included 121 patients who underwent prebiopsy MRI and prostate biopsy. More than five radiologists (Reader groups 1, 2: residents; Readers 3, 4: less-experienced radiologists; Reader 5: expert) independently reviewed biparametric MRI (bpMRI). The DLA results were obtained using bpMRI. The reference standard was based on pathologic reports. The diagnostic performance of the PI-RADS classification of DLA, clinical reports, and radiologists was analyzed using AUROC. Dichotomous analysis (PI-RADS cutoff value ≥ 3 or 4) was performed, and the sensitivities and specificities were compared using McNemar’s test. ResultsClinically significant cancer [CSC, Gleason score ≥ 7] was confirmed in 43 patients (35.5%). The AUROC of the DLA (0.828) for diagnosing CSC was significantly higher than that of Reader 1 (AUROC, 0.706; p = 0.011), significantly lower than that of Reader 5 (AUROC, 0.914; p = 0.013), and similar to clinical reports and other readers (p = 0.060–0.661). The sensitivity of DLA (76.7%) was comparable to those of all readers and the clinical reports at a PI-RADS cutoff value ≥ 4. The specificity of the DLA (85.9%) was significantly higher than those of clinical reports and Readers 2–3 and comparable to all others at a PI-RADS cutoff value ≥ 4. ConclusionsThe DLA showed moderate diagnostic performance at a level between those of residents and an expert in detecting and classifying according to PI-RADS. The performance of DLA was similar to that of clinical reports from various radiologists in clinical practice.

Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography

Computed tomography (CT) is the preferred imaging method for diagnosing 2019 novel coronavirus (COVID19) pneumonia. We aimed to construct a system based on deep learning for detecting COVID-19 pneumonia on high resolution CT. For model development and validation, 46,096 anonymous images from 106 admitted patients, including 51 patients of laboratory confirmed COVID-19 pneumonia and 55 control patients of other diseases in Renmin Hospital of Wuhan University were retrospectively collected. Twenty-seven prospective consecutive patients in Renmin Hospital of Wuhan University were collected to evaluate the efficiency of radiologists against 2019-CoV pneumonia with that of the model. An external test was conducted in Qianjiang Central Hospital to estimate the system’s robustness. The model achieved a per-patient accuracy of 95.24% and a per-image accuracy of 98.85% in internal retrospective dataset. For 27 internal prospective patients, the system achieved a comparable performance to that of expert radiologist. In external dataset, it achieved an accuracy of 96%. With the assistance of the model, the reading time of radiologists was greatly decreased by 65%. The deep learning model showed a comparable performance with expert radiologist, and greatly improved the efficiency of radiologists in clinical practice.

A robust convolutional neural network for lung nodule detection in the presence of foreign bodies

Lung cancer is a major cause of death worldwide. As early detection can improve outcome, regular screening is of great interest, especially for certain risk groups. Besides low-dose computed tomography, chest X-ray is a potential option for screening. Convolutional network (CNN) based computer aided diagnosis systems have proven their ability of identifying nodules in radiographies and thus may assist radiologists in clinical practice. Based on segmented pulmonary nodules, we trained a CNN based one-stage detector (RetinaNet) with 257 annotated radiographs and 154 additional radiographs from a public dataset. We compared the performance of the convolutional network with the performance of two radiologists by conducting a reader study with 75 cases. Furthermore, the potential use for screening on patient level and the impact of foreign bodies with respect to false-positive detections was investigated. For nodule location detection, the architecture achieved a performance of 43 true-positives, 26 false-positives and 22 false-negatives. In comparison, performance of the two readers was 42 ± 2 true-positives, 28 ± 0 false-positives and 23 ± 2 false-negatives. For the screening task, we retrieved a ROC AUC value of 0.87 for the reader study test set. We found the trained RetinaNet architecture to be only slightly prone to foreign bodies in terms of misclassifications: out of 59 additional radiographs containing foreign bodies, false-positives in two radiographs were falsely detected due to foreign bodies.

Atypical Appearance of Hepatocellular Carcinoma and Its Mimickers: How to Solve Challenging Cases Using Gadoxetic Acid-Enhanced Liver Magnetic Resonance Imaging.

Hepatocellular carcinoma (HCC) can be diagnosed noninvasively with contrast-enhanced dynamic computed tomography, magnetic resonance imaging, or ultrasonography on the basis of its hallmark imaging features of arterial phase hyperenhancement and washout on portal or delayed phase images. However, approximately 40% of HCCs show atypical imaging features, posing a significant diagnostic challenge for radiologists. Another challenge for radiologists in clinical practice is the presentation of many HCC mimickers such as intrahepatic cholangiocarcinoma, combined HCC-cholangiocarcinoma, arterioportal shunt, and hemangioma in the cirrhotic liver. The differentiation of HCCs from these mimickers on preoperative imaging studies is of critical importance. Hence, we will review the typical and atypical imaging features of HCCs and the imaging features of its common mimickers. In addition, we will discuss how to solve these challenges in practice.

Diş Hekimlerinin Konik Işınlı Bilgisayarlı Tomografi Raporlarına Yönelik Algı ve Tutumları

Objectives: Radiology reports are the most important method of communication between the clinician and the radiologist. In dentomaxillofacial radiology, cone-beam computed tomography (CBCT) reporting is a new subject. The purpose of this study was to evaluate the satisfaction and expectations of dentists from CBCT reporting as well as contributing to standardization and improvement in the quality of CBCT reports. Materials and Methods: Dentists were invited to participate in the survey by e-mail. The participants filled out a survey with their demographic data and responded to 14 questions regarding CBCT reports. The responses regarding gender, age, title, institution, and department were analysed and compared with chi-square tests. Results: In total, 185 dentists (97 females and 88 males) participated in the study. Participants reported that the adequacy level of the reports were mostly moderate (47%) and that the source of adequate reports was university hospitals (49.2%). Most dentists (57%) reported that they needed a consultant radiologist in clinical practice on a part time basis. There was a statistically significant difference (p&amp;lt;0.05) between participants’ genders, age groups, titles, and departments regarding the source of the adequate reports. Conclusions: The results of this study showed that most of the dentists were not satisfied about the proficiency of CBCT reports. More than half of those surveyed thought that “not reading” the radiology reports might give them a legal liability. Most dentists wanted to consult with the radiologist before and after patient examinations. 

3:27 PM Abstract No. 144 The effects of aprons, suspended protection systems, and mobile shields on radiation exposures to the chests and heads of interventional radiologists in clinical practice

3:27 PM Abstract No. 144 The effects of aprons, suspended protection systems, and mobile shields on radiation exposures to the chests and heads of interventional radiologists in clinical practice

Annals for Educators - 4 October 2016

Web Exclusives4 October 2016Annals for Educators - 4 October 2016FREEDarren B. Taichman, MD, PhDDarren B. Taichman, MD, PhDSearch for more papers by this authorAuthor, Article, and Disclosure Informationhttps://doi.org/10.7326/AFED201610040 SectionsAboutVisual Abstract ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinkedInRedditEmail Clinical Practice PointsVariation in Mammographic Breast Density Assessments Among Radiologists in Clinical Practice: Findings From a Multicenter Observational StudyAbout half of U.S. states currently have legislation requiring radiology facilities to disclose information on mammographic breast density to women, and many include language recommending discussion of supplemental screening options. However, breast density assessment, as routinely practiced in the clinical setting, is subjective. The authors performed cross-sectional and longitudinal analyses of prospectively collected observational data to examine variation in breast density assessment across radiologists in clinical practice.Use this study to:Ask your learners why breast density is important. What is the relationship between breast density and risk for breast cancer? What affect does breast density have on the performance of mammography? The authors discuss these in the paper's introduction.Arrange for a radiologist to show your team mammograms from patients with a range of breast tissue densities. Ask the radiologist to discuss the difficulties faced in evaluating breast density. On what bases do they quantify density?Does your state require notification of patients regarding the density of their breasts on mammography? How do your learners think such information should be communicated to patients? How would they explain its importance? How does the variation in radiologists' reading demonstrated in this study affect how they might counsel patients?Ask your learners how they would counsel a patient with dense breast tissue regarding cancer screening. Would their approach to test ordering be different? Use a recent guideline from the U.S. Preventive Services Task Force to help inform the discussion.Prediction Models of Mortality in Acute Pancreatitis in Adults. A Systematic ReviewThis systematic review of 94 studies evaluates the accuracy and utility of 18 severity scoring systems for predicting mortality in adults with acute pancreatitis.Use this study to:Start a teaching session with a multiple-choice question. We've provided one below!Ask your learners to generate a differential diagnosis of acute pancreatitis.What variables do they consider in evaluating the severity of the patients' conditions? How do they decide whether a patient needs hospital admission? Intensive care? Use the accompanying editorial to help inform your discussion.How do prognostic tools help us? What performance parameters should be considered before using a prognostic tool? Do your learners use any to help explain prognosis to their patients?How should patients with acute pancreatitis be managed? What interventions are helpful? How should they be monitored? What complications may occur, and how are they managed? Use the information in DynaMed Plus: Acute Pancreatitis (a benefit of your ACP membership) to help prepare a teaching session.DepressionWorldwide, unipolar depression is a leading cause of disability and loss of years of productivity. In the United States, depression estimates range from 5% to 10% but can be as high as 40% depending on the population seen in primary care or specialty settings. Only about half of depressed persons receive adequate treatment despite the existence of high-quality, evidence-based therapies.Use this review to:Ask who should be screened for depression. How should one screen for it?What are the diagnostic criteria?Who should be referred for psychotherapy? Behavioral cognitive therapy?Who should start pharmacotherapy, and with which agents? Use the table to compare their indications and possible side effects.How long should therapy continue? How should it be monitored?Download the teaching slides to help prepare for a teaching session. Use the multiple-choice questions provided to help break up a session, and be sure to log in to enter the answers and earn CME credit for yourself!Are physicians at increased risk for depression? What resources are available locally to help (e.g., crisis lines, employee health, counseling)? How might being labeled as “depressed” affect a physician?Humanities and ProfessionalismAnnals Graphic Medicine - Today's Doctor–Patient RelationshipThis artwork begs the question, “Where exactly do we practice medicine?”Use this artwork to:Show it to your learners and ask them for their reactions.What point is the author making? Do your learners agree?Do they feel they spend enough time with their patients? Why or why not? Is this within their control? If not, are there things they may do to improve the situation?Teach at the bedside! Show this picture to some of the patients on your team. Ask what the patients think!On Being a Patient: I Had to Get Cancer to Become a More Empathetic DoctorDr. Norden describes how his understanding of patients' anxiety, the burden of treatment side effects, and problematic health care communications changed as a result of his own experience with cancer.Use this essay to:Listen to an audio recording, read by On Being a Doctor/Patient editor, Dr. Michael LaCombe.Have your learners heard some or all of these lessons before?Do your learners think they do enough to be empathic and avoid the problems Dr. Norden describes in caring for their patients? Why do we seem to require reminders—over and over—of these essential messages? Are there ways to remind ourselves of these essential lessons regularly, and often enough? Do we all need to have been patients to become as compassionate as possible as physicians?Ask your learners to think about their own experiences. Have they ever experienced impersonal care, or care that was not as compassionate as it should have been? Perhaps they recall the experiences of family members. You might not want to ask them to answer aloud, as learners may not wish to share their own stories or reveal medical histories, but you can ask them to think about it.MKSAP 17 QuestionA 35-year-old man is evaluated in the emergency department for a 6-hour history of epigastric abdominal pain that radiates to the back. He also has nausea and occasional bilious vomiting. He has consumed between six and twelve beers daily for 10 to 15 years.On physical examination, temperature is 37.2 °C (99.0 °F), blood pressure is 110/65 mm Hg, pulse rate is 105/min, and respiration rate is 22/min. Abdominal examination discloses epigastric tenderness without guarding or rebound. Bowel sounds are present but hypoactive, and there is mild abdominal distention. No jaundice is noted.Laboratory studies reveal a leukocyte count of 14,000/μL (14 × 109/L), a blood urea nitrogen level of 25 mg/dL (8.9 mmol/L), and a serum lipase level of 952 U/L.Abdominal ultrasound shows a normal-appearing gallbladder and no biliary dilation. The patient is admitted to the hospital. Over the next 48 hours, he has ongoing abdominal pain, nausea, and poor appetite despite supportive therapy consisting of pain medication and aggressive intravenous fluid replacement. Subsequent contrast-enhanced CT of the abdomen shows nonenhancing areas of the head and body of the pancreas (consistent with necrosis) and several peripancreatic fluid collections.Which of the following is the most appropriate management?A. Drainage of the fluid collectionsB. Endoscopic retrograde cholangiopancreatographyC. Enteral nutrition by nasojejunal tubeD. Total parenteral nutritionCorrect AnswerC. Enteral nutrition by nasojejunal tubeEducational ObjectiveTreat acute pancreatitis with enteral nutrition.CritiqueEnteral feeding is the most appropriate management. Enteral feeding has been shown to reduce infectious complications, multiple organ failure, operative interventions, and mortality compared with feeding by total parenteral nutrition in patients with severe acute pancreatitis. This patient has moderately severe acute pancreatitis based on evidence of pancreatic necrosis and peripancreatic fluid collections. He also has several risk factors for severe disease based on the presence of three of four Systemic Inflammatory Response Syndrome (SIRS) criteria (pulse rate >90/min, leukocyte count >12,000/µL [12 × 109/L], and respiration rate >20/min) and a blood urea nitrogen level greater than 23 mg/dL (8.2 mmol/L). Nasogastric and nasojejunal feeding appear to be comparable in safety and efficacy. The optimal time to start enteral nutrition remains under investigation, but it should commence no later than 72 hours after presentation. In mild acute pancreatitis, oral feeding may start when nausea and vomiting resolve.Drainage of acute peripancreatic fluid collections (APFCs) is not appropriate at this time because most collections resolve without intervention. Asymptomatic APFCs require no treatment. Symptomatic APFCs can be treated medically with bowel rest, jejunal feeding, pancreatic enzymes, octreotide, and rarely pancreatic duct stenting. Rarely, APFCs persist beyond 4 weeks, when they become encapsulated and are labeled a pancreatic pseudocyst. Pseudocysts are amenable to drainage if clinically indicated based on persistent pain despite medical therapy, infected pseudocyst, or obstruction of the gastric outlet or biliary tract.Endoscopic retrograde cholangiopancreatography in acute pancreatitis should be used only in the following clinical scenarios: (1) in a patient with ascending cholangitis (fever, right upper quadrant pain, and jaundice) concomitant with acute pancreatitis, or (2) in a patient with gallstone pancreatitis who is not improving clinically and has worsening liver chemistry test results. Patients with gallstone pancreatitis and no complications should have a cholecystectomy prior to discharge.Key PointEnteral feeding has been shown to reduce infectious complications, multiple organ failure, operative interventions, and mortality compared with feeding by total parenteral nutrition in patients with severe acute pancreatitis.BibliographyTenner S, Baillie J, DeWitt J, Vege SS; American College of Gastroenterology. American College of Gastroenterology guideline: management of acute pancreatitis. Am J Gastroenterol. 2013;108:1400-15; 1416. doi:10.1038/ajg.2013.218Do you like reading Annals for Educators? Receive it direct to your inbox. Sign up for the Annals for Educators alert today. Comments0 CommentsSign In to Submit A Comment Author, Article, and Disclosure InformationAffiliations: From the Editors of Annals of Internal Medicine and Education Guest Editor, Gretchen Diemer, MD, FACP, Associate Dean of Graduate Medical Education and Affiliations, Thomas Jefferson University. PreviousarticleNextarticle Advertisement FiguresReferencesRelatedDetails Metrics 4 October 2016Volume 165, Issue 7Page: ED7KeywordsAbdominal painAcute pancreatitisMortalityNauseaNutritionPatient advocacyRadiologistsRespirationTotal parenteral nutritionVomiting ePublished: 4 October 2016 Issue Published: 4 October 2016 Copyright & PermissionsCopyright © 2016 by American College of Physicians. All Rights Reserved.Loading ...