Implementation of a Computerized Decision Support System for Computed Tomography Scan Requests for Nontraumatic Headache in the Emergency Department.

BACKGROUND: Nontraumatic headache is a frequent complaint in the emergency department (ED). Cranial computed tomography (CT) is a widely available test for the diagnostic work-up, despite the risk of exposure to ionizing radiation. OBJECTIVES: We sought to develop and evaluate a cranial CT request computerized decision support system (CDSS) for adults with their first presentation of unusual severe nontraumatic headache in the ED. METHODS: Electronic database searches identified clinical decision and prediction rules and studies delineating risk factors in nontraumatic headache. A long list of risk factors extracted from these articles was reduced by a 30-member multidisciplinary expert panel (radiologists, emergency physicians, methodologists), using a 90% agreement threshold. This shortlist was used to develop the algorithm for the cranial CT request CDSS, which was implemented in March 2016. Impact evaluation compared CT scan frequency and diagnostic yield of pathologic findings before (March-August 2015) and after (March-August 2016) implementation. RESULTS: From the 10 selected studies, 10 risk factors were shortlisted to activate a request for cranial CT. Before implementation, 377 cranial CTs were ordered (15.3% of 2469 CT scans) compared with 244 after (9.5% of 2561 CT scans; pre-post difference 5.74%; 95% confidence interval [CI] 3.92-7.56%; p < 0.001), corresponding to a 37.6% relative reduction in the test ordering rate (95% CI 25.7-49.5%; p < 0.001). Despite the reduction in cranial CT scans, we did not observe an increase in pathological findings after introducing the decision support system (70 cases before [18.5%] vs. 35 cases after [14.3%]; pre-post difference -4.0% [95% CI -10.0 to 1.6%]; p = 0.170). CONCLUSION: In nontraumatic headache among adults seen in the ED, CDSS decreased the cranial CT request rate but the diagnostic yield did not improve.

The validity of ICD codes coupled with imaging procedure codes for identifying acute venous thromboembolism using administrative data.

The purpose of this study was to evaluate the accuracy of using a combination of International Classification of Diseases (ICD) diagnostic codes and imaging procedure codes for identifying deep vein thrombosis (DVT) and pulmonary embolism (PE) within administrative databases. Information from the Alberta Health (AH) inpatients and ambulatory care administrative databases in Alberta, Canada was obtained for subjects with a documented imaging study result performed at a large teaching hospital in Alberta to exclude venous thromboembolism (VTE) between 2000 and 2010. In 1361 randomly-selected patients, the proportion of patients correctly classified by AH administrative data, using both ICD diagnostic codes and procedure codes, was determined for DVT and PE using diagnoses documented in patient charts as the gold standard. Of the 1361 patients, 712 had suspected PE and 649 had suspected DVT. The sensitivities for identifying patients with PE or DVT using administrative data were 74.83% (95% confidence interval [CI]: 67.01-81.62) and 75.24% (95% CI: 65.86-83.14), respectively. The specificities for PE or DVT were 91.86% (95% CI: 89.29-93.98) and 95.77% (95% CI: 93.72-97.30), respectively. In conclusion, when coupled with relevant imaging codes, VTE diagnostic codes obtained from administrative data provide a relatively sensitive and very specific method to ascertain acute VTE.

Classification of COVID-19 Chest CT Images Based on Ensemble Deep Learning.

Novel coronavirus pneumonia (NCP) has become a global pandemic disease, and computed tomography-based (CT) image analysis and recognition are one of the important tools for clinical diagnosis. In order to assist medical personnel to achieve an efficient and fast diagnosis of patients with new coronavirus pneumonia, this paper proposes an assisted diagnosis algorithm based on ensemble deep learning. The method combines the Stacked Generalization ensemble learning with the VGG16 deep learning to form a cascade classifier, and the information constituting the cascade classifier comes from multiple subsets of the training set, each of which is used to collect deviant information about the generalization behavior of the data set, such that this deviant information fills the cascade classifier. The algorithm was experimentally validated for classifying patients with novel coronavirus pneumonia, patients with common pneumonia (CP), and normal controls, and the algorithm achieved a prediction accuracy of 93.57%, sensitivity of 94.21%, specificity of 93.93%, precision of 89.40%, and F1-score of 91.74% for the three categories. The results show that the method proposed in this paper has good classification performance and can significantly improve the performance of deep neural networks for multicategory prediction tasks.

Convolutional Sparse Support Estimator-Based COVID-19 Recognition From X-Ray Images.

Coronavirus disease (COVID-19) has been the main agenda of the whole world ever since it came into sight. X-ray imaging is a common and easily accessible tool that has great potential for COVID-19 diagnosis and prognosis. Deep learning techniques can generally provide state-of-the-art performance in many classification tasks when trained properly over large data sets. However, data scarcity can be a crucial obstacle when using them for COVID-19 detection. Alternative approaches such as representation-based classification [collaborative or sparse representation (SR)] might provide satisfactory performance with limited size data sets, but they generally fall short in performance or speed compared to the neural network (NN)-based methods. To address this deficiency, convolution support estimation network (CSEN) has recently been proposed as a bridge between representation-based and NN approaches by providing a noniterative real-time mapping from query sample to ideally SR coefficient support, which is critical information for class decision in representation-based techniques. The main premises of this study can be summarized as follows: 1) A benchmark X-ray data set, namely QaTa-Cov19, containing over 6200 X-ray images is created. The data set covering 462 X-ray images from COVID-19 patients along with three other classes; bacterial pneumonia, viral pneumonia, and normal. 2) The proposed CSEN-based classification scheme equipped with feature extraction from state-of-the-art deep NN solution for X-ray images, CheXNet, achieves over 98% sensitivity and over 95% specificity for COVID-19 recognition directly from raw X-ray images when the average performance of 5-fold cross validation over QaTa-Cov19 data set is calculated. 3) Having such an elegant COVID-19 assistive diagnosis performance, this study further provides evidence that COVID-19 induces a unique pattern in X-rays that can be discriminated with high accuracy.

AANet: Adaptive Attention Network for COVID-19 Detection From Chest X-Ray Images.

Accurate and rapid diagnosis of COVID-19 using chest X-ray (CXR) plays an important role in large-scale screening and epidemic prevention. Unfortunately, identifying COVID-19 from the CXR images is challenging as its radiographic features have a variety of complex appearances, such as widespread ground-glass opacities and diffuse reticular-nodular opacities. To solve this problem, we propose an adaptive attention network (AANet), which can adaptively extract the characteristic radiographic findings of COVID-19 from the infected regions with various scales and appearances. It contains two main components: an adaptive deformable ResNet and an attention-based encoder. First, the adaptive deformable ResNet, which adaptively adjusts the receptive fields to learn feature representations according to the shape and scale of infected regions, is designed to handle the diversity of COVID-19 radiographic features. Then, the attention-based encoder is developed to model nonlocal interactions by self-attention mechanism, which learns rich context information to detect the lesion regions with complex shapes. Extensive experiments on several public datasets show that the proposed AANet outperforms state-of-the-art methods.

Radiation dose levels in chest computed tomography scans of coronavirus disease 2019 pneumonia: A survey of 2119 patients in Chongqing, southwest China.

ABSTRACT: To investigate computed tomography (CT) diagnostic reference levels for coronavirus disease 2019 (COVID-19) pneumonia by collecting radiation exposure parameters of the most performed chest CT examinations and emphasize the necessity of low-dose CT in COVID-19 and its significance in radioprotection.The survey collected RIS data from 2119 chest CT examinations for 550 COVID-19 patients performed in 92 hospitals from January 23, 2020 to May 1, 2020. Dose data such as volume computed tomography dose index, dose-length product, and effective dose (ED) were recorded and analyzed. The radiation dose levels in different hospitals have been compared, and average ED and cumulative ED have been studied.The median dose-length product, volume computed tomography dose index, and ED measurements were 325.2 mGy cm with a range of 6.79 to 1098 mGy cm, 9.68 mGy with a range of 0.62 to 33.80 mGy, and 4.55 mSv with a range of 0.11 to 15.37 mSv for COVID-19 CT scanning protocols in Chongqing, China. The distribution of all observed EDs of radiation received by per patient undergoing CT protocols during hospitalization yielded a median cumulative ED of 17.34 mSv (range, 2.05-53.39 mSv) in the detection and management of COVID-19 patients. The average number of CT scan times for each patient was 4.0 ±â€Š2.0, and the average time interval between 2 CT scans was 7.0 ±â€Š5.0 days. The average cumulative ED of chest CT examinations for COVID-19 patients in Chongqing, China greatly exceeded public limit and the annual dose limit of occupational exposure in a short period.For patients with known or suspected COVID-19, a chest CT should be performed on the principle of rapid-scan, low-dose, single-phase protocol instead of routine chest CT protocol to minimize radiation doses and motion artifacts.

Updated Australian diagnostic reference levels for adult CT.

INTRODUCTION: In 2018, ARPANSA published updated national DRLs for adult CT, which were first published in 2012, and augmented the national DRL categories. This paper presents the updated national DRLs and describes the process by which they were produced. METHODS: Examine patient survey data submitted to the Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) National Diagnostic Reference Level Service (NDRLS). Determine the quartiles of the distributions of median survey dose metrics with categorisation by procedure type. Engage a liaison panel representing the radiology professions to review procedure categories and recommend revised national DRLs. The revised NDRL procedure categories are: head (non-contrast brain (trauma/headache)), cervical spine (Non-contrast (trauma)), soft-tissue neck (post-contrast (oncology)), chest (post-contrast (oncology)), abdomen-pelvis (post-contrast (oncology)), kidney-ureter-bladder (non-contrast (suspected renal colic)), chest-abdomen-pelvis (post-contrast (oncology)) and lumbar spine (non-contrast (degenerative pain)). RESULTS: The existing six procedure categories were revised and refined. Updated Australian national diagnostic reference levels for adult CT were recommended and endorsed for eight procedure categories: head (52 mGy/880 mGycm), cervical spine (23 mGy/470 mGycm),soft-tissue neck (17 mGy/450 mGycm), chest (10 mGy/390 mGycm), abdomen-pelvis (13 mGy/600 mGycm), kidney-ureter-bladder (13 mGy/600 mGycm), chest-abdomen-pelvis (11 mGy/940 mGycm) and lumbar spine (26 mGy/670 mGycm). The updated national DRLs are between 12 and 26% lower than the previous DRLs for dose-length product and between 13 and 63% lower for volume computed tomography dose index. CONCLUSIONS: Australian national DRLs for adult CT have been reviewed and revised. The revised national DRLs are lower, better reflecting current practice among imaging facilities in Australia. The revised Australian national DRLs are similar to those in other developed countries.

Systematic review of artificial intelligence techniques in the detection and classification of COVID-19 medical images in terms of evaluation and benchmarking: Taxonomy analysis, challenges, future solutions and methodological aspects.

This study presents a systematic review of artificial intelligence (AI) techniques used in the detection and classification of coronavirus disease 2019 (COVID-19) medical images in terms of evaluation and benchmarking. Five reliable databases, namely, IEEE Xplore, Web of Science, PubMed, ScienceDirect and Scopus were used to obtain relevant studies of the given topic. Several filtering and scanning stages were performed according to the inclusion/exclusion criteria to screen the 36 studies obtained; however, only 11 studies met the criteria. Taxonomy was performed, and the 11 studies were classified on the basis of two categories, namely, review and research studies. Then, a deep analysis and critical review were performed to highlight the challenges and critical gaps outlined in the academic literature of the given subject. Results showed that no relevant study evaluated and benchmarked AI techniques utilised in classification tasks (i.e. binary, multi-class, multi-labelled and hierarchical classifications) of COVID-19 medical images. In case evaluation and benchmarking will be conducted, three future challenges will be encountered, namely, multiple evaluation criteria within each classification task, trade-off amongst criteria and importance of these criteria. According to the discussed future challenges, the process of evaluation and benchmarking AI techniques used in the classification of COVID-19 medical images considered multi-complex attribute problems. Thus, adopting multi-criteria decision analysis (MCDA) is an essential and effective approach to tackle the problem complexity. Moreover, this study proposes a detailed methodology for the evaluation and benchmarking of AI techniques used in all classification tasks of COVID-19 medical images as future directions; such methodology is presented on the basis of three sequential phases. Firstly, the identification procedure for the construction of four decision matrices, namely, binary, multi-class, multi-labelled and hierarchical, is presented on the basis of the intersection of evaluation criteria of each classification task and AI classification techniques. Secondly, the development of the MCDA approach for benchmarking AI classification techniques is provided on the basis of the integrated analytic hierarchy process and VlseKriterijumska Optimizacija I Kompromisno Resenje methods. Lastly, objective and subjective validation procedures are described to validate the proposed benchmarking solutions.

Interobserver Agreement for the Computed Tomography Severity Grading Scales for Acute Traumatic Brain Injury.

The purpose of this study was to determine the interobserver variability among providers of different specialties and levels of experience across five established computed tomography (CT) scoring systems for acute traumatic brain injury (TBI). One hundred cases were selected at random from a retrospective population of adult patients transported to our emergency department and subjected to a non-contrast head CT due to suspicion of TBI. Eight neuroradiologists and neurosurgeons in trainee (residents and fellows) and attending roles independently scored each non-contrast head CT scan on the Marshall, Rotterdam, Helsinki, Stockholm, and NeuroImaging Radiological Interpretation System (NIRIS) head CT scales. Interobserver variability of scale scores-overall and by specialty and level of training-was quantified using the intraclass correlation coefficient (ICC), and agreement with respect to National Institutes of Health Common Data Elements (NIH CDEs) was assessed using Cohen's kappa. All CT severity scoring systems showed high interobserver agreement as evidenced by high ICCs, ranging from 0.75-0.89. For all scoring systems, neuroradiologists (ICC range from 0.81-0.94) tended to have higher interobserver agreement than neurosurgeons (ICC range from 0.63-0.76). For all scoring systems, attendings (ICC range from 0.76-0.89) had similar interobserver agreement to trainees (ICC range from 0.73-0.89). Agreement with respect to NIH CDEs was high for ascertaining presence/absence of hemorrhage, skull fracture, and mass effect, with estimated kappa statistics of least 0.89. Acute TBI CT scoring systems demonstrate high interobserver agreement. These results provide scientific rigor for future use of these systems for the classification of acute TBI.

A Novel Classification: Anomalous Routes of the Facial Nerve in Relation to Inner Ear Malformations.

OBJECTIVES/HYPOTHESIS: The objective of this study was to classify anomalous facial nerve (FN) routes and to determine their association with inner ear malformations (IEMs). STUDY DESIGN: Retrospective cross sectional study. METHODS: The computed tomography images of 519 patients (796 ears) with IEMs were retrospectively evaluated, and the abnormal routes of the FN were classified as: Meatal segment: type 1, normal internal auditory canal (IAC); type 2, narrow IAC; type 3, facial canal (FC) only; type 4: separate FC/duplicated IAC. Labyrinthine segment (LS): type 1, normal; type 2a/b/c, mild/moderate/severe anterior displacement; type 3, superior displacement; type 4: straight LS. Tympanic segment (TS): type 1, normal; type 2, superiorly displaced TS; type 3, TS at the oval window; type 4: TS inferior to the oval window; type 5: unclassified. Mastoid segment: type 1, normal facial recess (FR)/normal mastoid segment; type 2: narrow FR; type 3, unclassified. RESULTS: In meatal segment classification, a narrow IAC was common in ears with cochlear hypoplasia (CH) (76.1%), and only FC was common in ears with severe IEMs (62.7%) such as Michel deformity, common cavity, and cochlear aplasia. Incomplete partition-III has its unique superiorly displaced LS (100%). CH-IV also has its unique mild anterosuperior displacement. Ears with a superiorly displaced TS usually (93.1%) had aplastic or hypoplastic semicircular canals. The FR is likely to be narrow in CH and severe IEMs. CONCLUSIONS: The FN route is affected in IEMs, which must be kept in mind when operating on ears with IEMs. Especially in CH cases, all segments of the FN can be abnormal. LEVEL OF EVIDENCE: 4 Laryngoscope, 130:E696-E703, 2020.

Day-of-Injury Computed Tomography and Longitudinal Rehabilitation Outcomes: A Comparison of the Marshall and Rotterdam Computed Tomography Scoring Methods.

OBJECTIVE: The aim of the study was to compare the relative predictive value of Marshall Classification System and Rotterdam scores on long-term rehabilitation outcomes. This study hypothesized that Rotterdam would outperform Marshall Classification System. DESIGN: The study used an observational cohort design with a consecutive sample of 88 participants (25 females, mean age = 42.0 [SD = 21.3]) with moderate to severe traumatic brain injury who were admitted to trauma service with subsequent transfer to the rehabilitation unit between February 2009 and July 2011 and who had clearly readable computed tomography scans. Twenty-three participants did not return for the 9-mo postdischarge follow-up. Day-of-injury computed tomography images were scored using both Marshall Classification System and Rotterdam criteria by two independent raters, blind to outcomes. Functional outcomes were measured by length of stay in rehabilitation and the cognitive and motor subscales of the Functional Independence Measure at rehabilitation discharge and 9-mo postdischarge follow-up. RESULTS: Neither Marshall Classification System nor Rotterdam scales as a whole significantly predicted Functional Independence Measure motor or cognitive outcomes at discharge or 9-mo follow-up. Both scales, however, predicted length of stay in rehabilitation. Specific Marshall scores (3 and 6) and Rotterdam scores (5 and 6) significantly predicted subacute outcomes such as Functional Independence Measure cognitive at discharge from rehabilitation and length of stay. CONCLUSIONS: Marshall Classification System and Rotterdam scales may have limited utility in predicting long-term functional outcome, but specific Marshall and Rotterdam scores, primarily linked to increased severity and intracranial pressure, may predict subacute outcomes.

New Classification of Cochlear Hypoplasia Type Malformation: Relevance in Cochlear Implantation.

OBJECTIVES: This paper attempts to create a new classification type of cochlear hypoplasia (CH)-type malformation taking into consideration of vestibular section and internal auditory canal (IAC). MATERIALS AND METHODS: Preoperative computed-tomography (CT) scans of cochlear implant (CI) candidates (N=31) from various clinics across the world with CH type malformation were taken for analysis. CT dataset were loaded into 3D-slicer freeware for three-dimensional (3D) segmentation of the inner-ear by capturing complete inner-ear structures from the entire dataset. Cochlear size in terms of diameter of available cochlear basal turn and length of cochlear lumen was measured from the dataset. In addition, structural connection between IAC and cochlear portions was scrutinized, which is highly relevant to the proposed CH classification in this study. RESULTS: CH group-I has the normal presence of IAC leading to cochlear and vestibular portions, whereas CH group-II is like CH group-I but with some degree of disruption in vestibular portion. In CH group-III, a disconnection between IAC and the cochlear portion irrespective of other features. Within all these three CH groups, the basal turn diameter varied between 3.1 mm and 9.6 mm, and the corresponding cochlear lumen length varied between 3 mm and 21 mm for the CI electrode array placement. CONCLUSION: A new classification of CH mainly based on the IAC connecting the cochlear and vestibular portions is presented in this study. CI electrode array length could be selected based on the length of the cochlear lumen, which can be observed from the 3D image.

COVID-19 S: A new proposal for diagnosis and structured reporting of COVID-19 on computed tomography imaging.

PURPOSE: Because of the widespread use of CT in the diagnosis of COVID 19, indeterminate presentations such as single, few or unilateral lesions amount to a considerable number. We aimed to develop a new classification and structured reporting system on CT imaging (COVID-19 S) that would facilitate the diagnosis of COVID-19 in the most accurate way. METHODS: Our retrospective cohort included 803 patients with a chest CT scan upon suspicion of COVID 19. The patients' history, physical examination, CT findings, RT PCR, and other laboratory test results were reviewed, and a final diagnosis was made as COVID 19 or non-COVID 19. Chest CT scans were classified according to the COVID 19 S CT diagnosis criteria. Cohen's kappa analysis was used. RESULTS: Final clinical diagnosis was COVID-19 in 98 patients (12%). According to the COVID-19 S CT diagnosis criteria, the number of patients in the normal, compatible with COVID 19, indeterminate and alternative diagnosis groups were 581 (72.3%), 97 (12.1%), 16 (2.0%) and 109 (13.6%). When the indeterminate group was combined with the group compatible with COVID 19, the sensitivity and specificity of COVID-19 S were 99.0% and 87.1%, with 85.8% positive predictive value (PPV) and 99.1% negative predictive value (NPV). When the indeterminate group was combined with the alternative diagnosis group, the sensitivity and specificity of COVID-19 S were 93.9% and 96.0%, with 94.8% PPV and 95.2% NPV. CONCLUSION: COVID-19 S CT classification system may meet the needs of radiologists in distinguishing COVID-19 from pneumonia of other etiologies and help optimize patient management and disease control in this pandemic by the use of structured reporting.

Mapping of HIE CT terms to LOINC®: analysis of content-dependent coverage and coverage improvement through new term creation.

Objective: We describe and evaluate the mapping of computerized tomography (CT) terms from 40 hospitals participating in a health information exchange (HIE) to a standard terminology. Methods: Proprietary CT exam terms and corresponding exam frequency data were obtained from 40 participant HIE sites that transmitted radiology data to the HIE from January 2013 through October 2015. These terms were mapped to the Logical Observations Identifiers Names and Codes (LOINC®) terminology using the Regenstrief LOINC mapping assistant (RELMA) beginning in January 2016. Terms without initial LOINC match were submitted to LOINC as new term requests on an ongoing basis. After new LOINC terms were created, proprietary terms without an initial match were reviewed and mapped to these new LOINC terms where appropriate. Content type and token coverage were calculated for the LOINC version at the time of initial mapping (v2.54) and for the most recently released version at the time of our analysis (v2.63). Descriptive analysis was performed to assess for significant differences in content-dependent coverage between the 2 versions. Results: LOINC's content type and token coverages of HIE CT exam terms for version 2.54 were 83% and 95%, respectively. Two-hundred-fifteen new LOINC CT terms were created in the interval between the releases of version 2.54 and 2.63, and content type and token coverages, respectively, increased to 93% and 99% (P < .001). Conclusion: LOINC's content type coverage of proprietary CT terms across 40 HIE sites was 83% but improved significantly to 93% following new term creation.

Comparison of Deep Learning-Based and Patch-Based Methods for Pseudo-CT Generation in MRI-Based Prostate Dose Planning.

PURPOSE: Deep learning methods (DLMs) have recently been proposed to generate pseudo-CT (pCT) for magnetic resonance imaging (MRI) based dose planning. This study aims to evaluate and compare DLMs (U-Net and generative adversarial network [GAN]) using various loss functions (L2, single-scale perceptual loss [PL], multiscale PL, weighted multiscale PL) and a patch-based method (PBM). METHODS AND MATERIALS: Thirty-nine patients received a volumetric modulated arc therapy for prostate cancer (78 Gy). T2-weighted MRIs were acquired in addition to planning CTs. The pCTs were generated from the MRIs using 7 configurations: 4 GANs (L2, single-scale PL, multiscale PL, weighted multiscale PL), 2 U-Net (L2 and single-scale PL), and the PBM. The imaging endpoints were mean absolute error and mean error, in Hounsfield units, between the reference CT (CTref) and the pCT. Dose uncertainties were quantified as mean absolute differences between the dose volume histograms (DVHs) calculated from the CTref and pCT obtained by each method. Three-dimensional gamma indexes were analyzed. RESULTS: Considering the image uncertainties in the whole pelvis, GAN L2 and U-Net L2 showed the lowest mean absolute error (≤34.4 Hounsfield units). The mean errors were not different than 0 (P ≤ .05). The PBM provided the highest uncertainties. Very few DVH points differed when comparing GAN L2 or U-Net L2 DVHs and CTref DVHs (P ≤ .05). Their dose uncertainties were ≤0.6% for the prostate planning target Volume V95%, ≤0.5% for the rectum V70Gy, and ≤0.1% for the bladder V50Gy. The PBM, U-Net PL, and GAN PL presented the highest systematic dose uncertainties. The gamma pass rates were >99% for all DLMs. The mean calculation time to generate 1 pCT was 15 s for the DLMs and 62 min for the PBM. CONCLUSIONS: Generating pCT for MRI dose planning with DLMs and PBM provided low-dose uncertainties. In particular, the GAN L2 and U-Net L2 provided the lowest dose uncertainties together with a low computation time.

Nonhypervascular hypoattenuating nodules depicted on either portal or equilibrium phase multiphasic CT images in the cirrhotic liver.

OBJECTIVE: The objective of this study was to investigate the outcome and clinical implications of nonhypervascular hypoattenuating nodules observed on portal or equilibrium phase CT images of cirrhotic livers. MATERIALS AND METHODS: One hundred one cirrhotic patients (male:female = 69:32) with hypoattenuating nodules observed on initial portal or equilibrium phase CT images were retrospectively evaluated by follow-up CT performed 6-66 months after the initial CT examination. Depending on the background nodularity, patients were separated into macronodular (n = 33, 288 nodules) and micronodular (n = 68, 346 nodules) cirrhotic groups. Each nodule was categorized as category I (enlarged) or category II (stable). Nodule categories were correlated with the initial lesion size and the pattern of background cirrhosis. RESULTS: The frequency of category I nodules was higher in patients with micronodular cirrhosis (40%) than in those with macronodular cirrhosis (27%) (p = 0.001). Category I nodules were significantly larger than category II nodules in patients with micronodular cirrhosis (p < 0.001). The doubling times of category I nodules had no statistical difference between patients with micronodular or macronodular cirrhosis (p = 0.954). Of the category I nodules in patients with micronodular cirrhosis, 8.6% showed malignant changes. CONCLUSION: More careful attention should be paid to large nodules in patients with micronodular cirrhosis because of the potentially greater risk of malignancy, and small hypoattenuating nodules should be more often followed up in shorter intervals than large nodules.

Clarifying the TNM staging of rectal cancer in the context of modern imaging and neo-adjuvant treatment: 'y''u' and 'p' need 'mr' and 'ct'.

Recent developments in cross-sectional imaging, particularly computerized tomography and magnetic resonance imaging have provided increasingly accurate noninvasive preoperative staging, especially for rectal cancer. Image-based TNM staging, by definition a pathological system, has entered both the literature and everyday usage with no universal agreement as to the exact terminology. Clarification of the current terminology and suggestions to reflect recent developments are outlined to facilitate multidisciplinary team decision processes and objective stratification for entry to, and monitoring of, future clinical trials in rectal cancer.

Indexing thoracic CT reports using a preliminary version of a standardized radiological lexicon (RadLex).

INTRODUCTION: To validate a preliminary version of a radiological lexicon (RadLex) against terms found in thoracic CT reports and to index report content in RadLex term categories. MATERIAL AND METHODS: Terms from a random sample of 200 thoracic CT reports were extracted using a text processor and matched against RadLex. Report content was manually indexed by two radiologists in consensus in term categories of Anatomic Location, Finding, Modifier, Relationship, Image Quality, and Uncertainty. Descriptive statistics were used and differences between age groups and report types were tested for significance using Kruskal-Wallis and Mann-Whitney Test (significance level <0.05). RESULTS: From 363 terms extracted, 304 (84%) were found and 59 (16%) were not found in RadLex. Report indexing showed a mean of 16.2 encoded items per report and 3.2 Finding per report. Term categories most frequently encoded were Modifier (1,030 of 3,244, 31.8%), Anatomic Location (813, 25.1%), Relationship (702, 21.6%) and Finding (638, 19.7%). Frequency of indexed items per report was higher in older age groups, but no significant difference was found between first study and follow up study reports. Frequency of distinct findings per report increased with patient age (p < 0.05). CONCLUSION: RadLex already covers most terms present in thoracic CT reports based on a small sample analysis from one institution. Applications for report encoding need to be developed to validate the lexicon against a larger sample of reports and address the issue of automatic relationship encoding.

Automatic classification of radiological reports for clinical care.

Radiological reporting generates a large amount of free-text clinical narratives, a potentially valuable source of information for improving clinical care and supporting research. The use of automatic techniques to analyze such reports is necessary to make their content effectively available to radiologists in an aggregated form. In this paper we focus on the classification of chest computed tomography reports according to a classification schema proposed for this task by radiologists of the Italian hospital ASST Spedali Civili di Brescia. The proposed system is built exploiting a training data set containing reports annotated by radiologists. Each report is classified according to the schema developed by radiologists and textual evidences are marked in the report. The annotations are then used to train different machine learning based classifiers. We present in this paper a method based on a cascade of classifiers which make use of a set of syntactic and semantic features. The resulting system is a novel hierarchical classification system for the given task, that we have experimentally evaluated.

How Common Is the Ponticulus Posticus?: A Computed Tomography Based Analysis of 2917 Patients.

STUDY DESIGN: A computed tomography (CT) study of the morphology of the C1 vertebra. OBJECTIVE: Is to determine the prevalence of ponticulus posticus (PP) by analyzing CT scans performed on a large, diverse population in the northeast United States. This study also proposes a CT-based classification system both to aid in unifying the description of PP, and to aid in future research. SUMMARY OF BACKGROUND DATA: The prevalence of PP varies from 5% to 68% in published studies. There may be geographic variation in the prevalence of PP. Our objective was to establish the prevalence of PP in the general population, and to develop a comprehensive classification system to describe PP. METHODS: We evaluated cervical spine CT scans performed on patients in the emergency room of a level I trauma center over a 1-year period (January 1, 2014-December 31, 2014). The CT images were evaluated for the presence of a PP, and if present the following demographic data were collected: age, sex, race/ethnicity, and body mass index (BMI). We propose a novel classification system to standardize the description of PP identified on CT scan. RESULTS: Two thousand, nine hundred and seventeen cervical spine CT scans were reviewed in this study. The prevalence of PP was 22.5%. Men had a higher prevalence of PP than women (53.5% male vs. 46.5% female P ≤ 0.01). When compared with the overall population, African-Americans were more likely to have a PP (P ≤ 0.01), while Caucasian patients were less likely (P ≤ 0.01). The novel classification consisted of a two letter designation for each patient, with the first letter denoting the right sided arch and the second letter the left sided arch. Each side of the arch described as an A, B, or C type. The A type had no presence of a PP, B type had in incomplete PP, and C type had a complete PP. The most common type of a PP was CC encompassing 25% of the patients. The presence of a PP was more common in the left sided arch than the right side (B and C type Left 89.2% vs. B and C type Right 84.7%, P = 0.02). CONCLUSION: We found a 22.5% prevalence of PP in 2917 patients undergoing a cervical spine CT. This is the largest study to evaluating the prevalence of PP. LEVEL OF EVIDENCE: 4.