RESUMO
This work aimed to investigate whether global radiomic features (GRFs) from mammograms can predict difficult-to-interpret normal cases (NCs). Assessments from 537 readers interpreting 239 normal mammograms were used to categorise cases as 120 difficult-to-interpret and 119 easy-to-interpret based on cases having the highest and lowest difficulty scores, respectively. Using lattice- and squared-based approaches, 34 handcrafted GRFs per image were extracted and normalised. Three classifiers were constructed: (i) CC and (ii) MLO using the GRFs from corresponding craniocaudal and mediolateral oblique images only, based on the random forest technique for distinguishing difficult- from easy-to-interpret NCs, and (iii) CC + MLO using the median predictive scores from both CC and MLO models. Useful GRFs for the CC and MLO models were recognised using a scree test. The CC and MLO models were trained and validated using the leave-one-out-cross-validation. The models' performances were assessed by the AUC and compared using the DeLong test. A Kruskal-Wallis test was used to examine if the 34 GRFs differed between difficult- and easy-to-interpret NCs and if difficulty level based on the traditional breast density (BD) categories differed among 115 low-BD and 124 high-BD NCs. The CC + MLO model achieved higher performance (0.71 AUC) than the individual CC and MLO model alone (0.66 each), but statistically non-significant difference was found (all p > 0.05). Six GRFs were identified to be valuable in describing difficult-to-interpret NCs. Twenty features, when compared between difficult- and easy-to-interpret NCs, differed significantly (p < 0.05). No statistically significant difference was observed in difficulty between low- and high-BD NCs (p = 0.709). GRF mammographic analysis can predict difficult-to-interpret NCs.
Assuntos
Neoplasias da Mama , Mamografia , Humanos , Feminino , Mamografia/métodos , Densidade da Mama , Algoritmo Florestas Aleatórias , Neoplasias da Mama/diagnóstico por imagemRESUMO
BACKGROUND: The low subject contrast between cancerous and fibroglandular tissue could obscure breast abnormalities. PURPOSE: To investigate radiologists' performance for detection of breast cancer in low and high mammographic density (MD) when cases are digitally acquired. MATERIAL AND METHODS: A test set of 60 digital mammography cases, of which 20 were cancerous, were examined by 17 radiologists. Mammograms were categorized as low (≤50%) or high (>50%) MD and rated for suspicion of malignancy using the Royal Australian and New Zealand College of Radiology (RANZCR) classification system. Radiologist demographics including cases read per year, age, subspecialty, and years of reporting were recorded. Radiologist performance was analyzed by the following metrics: sensitivity; specificity; area under the receiver operating characteristic (ROC) curve (AUC), location sensitivity, and jackknife free-response ROC (JAFROC) figure of merit (FOM). RESULTS: Comparing high to low MD cases, radiologists showed a significantly higher sensitivity (P = 0.015), AUC (P = 0.003), location sensitivity (P = 0.002), and JAFROC FOM (P = 0.001). In high compared to low MD cases, radiologists with <1000 annual reads and radiologists with no mammographic subspecialty had significantly higher AUC, location sensitivity, and JAFROC FOM. Radiologists with ≥1000 annual reads and radiologists with mammography subspecialty demonstrated a significant increase in location sensitivity in high compared to low MD cases. CONCLUSION: In this experimental situation, radiologists' performance was higher when reading cases with high compared to low MD. Experienced radiologists were able to precisely localize lesions in breasts with higher MD. Further studies in unselected screening materials are needed to verify the results.
Assuntos
Densidade da Mama , Neoplasias da Mama/diagnóstico por imagem , Competência Clínica/estatística & dados numéricos , Mamografia/métodos , Mamografia/estatística & dados numéricos , Adulto , Idoso , Mama/diagnóstico por imagem , Humanos , Jordânia , Pessoa de Meia-Idade , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
OBJECTIVES: To evaluate and compare the image quality of propagation-based phase-contrast computed tomography (PB-CT) using synchrotron radiation and conventional cone-beam breast computed tomography (CBBCT) based on various radiological image quality criteria. METHODS: Eight excised breast tissue samples of various sizes and containing different lesion types were scanned using PB-CT at a synchrotron facility and using CBBCT at a university-affiliated breast imaging centre. PB-CT scans were performed at two different mean glandular dose (MGD) levels: standard (5.8 mGy) and low (1.5 mGy), for comparison with CBBCT scans at the standard MGD (5.8 mGy). Image quality assessment was carried out using six quality criteria and six independent medical imaging experts in a reading room with mammography workstations. The interobserver agreement between readers was evaluated using intraclass correlation coefficient (ICC), and image quality was compared between the two breast imaging modalities using the area under the visual grading characteristic curve (AUCVGC). RESULTS: Interobserver agreement between the readers showed moderate reliability for five image criteria (ICC: ranging from 0.488 to 0.633) and low reliability for one criterion (image noise) (ICC 0.307). For five image quality criteria (overall quality, perceptible contrast, lesion sharpness, normal tissue interfaces, and calcification visibility), both standard-dose PB-CT images (AUCVGC 0.958 to 1, p ≤ .05) and low dose PB-CT images (AUCVGC 0.785 to 0.834, p ≤ .05) were of significantly higher image quality than standard-dose CBBCT images. CONCLUSIONS: Synchrotron-based PB-CT can achieve a significantly higher radiological image quality at a substantially lower radiation dose compared with conventional CBBCT. KEY POINTS: ⢠PB-CT using synchrotron radiation results in higher image quality than conventional CBBCT for breast imaging. ⢠PB-CT using synchrotron radiation requires a lower radiation dose than conventional CBBCT for breast imaging. ⢠PB-CT can help clinicians diagnose patients with breast cancer.
Assuntos
Doenças Mamárias/diagnóstico , Mama/diagnóstico por imagem , Tomografia Computadorizada de Feixe Cônico/métodos , Mamografia/métodos , Síncrotrons , Feminino , Humanos , Doses de Radiação , Reprodutibilidade dos TestesRESUMO
Visual search behaviour and the interpretation of mammograms have been studied for errors in breast cancer detection. We aim to ascertain whether machine-learning models can learn about radiologists' attentional level and the interpretation of mammograms. We seek to determine whether these models are practical and feasible for use in training and teaching programmes. Eight radiologists of varying experience levels in reading mammograms reviewed 120 two-view digital mammography cases (59 cancers). Their search behaviour and decisions were captured using a head-mounted eye-tracking device and software allowing them to record their decisions. This information from radiologists was used to build an ensembled machine-learning model using top-down hierarchical deep convolution neural network. Separately, a model to determine type of missed cancer (search, perception or decision-making) was also built. Analysis and comparison of variants of these models using different convolution networks with and without transfer learning were also performed. Our ensembled deep-learning network architecture can be trained to learn about radiologists' attentional level and decisions. High accuracy (95%, p value â 0 [better than dumb/random model]) and high agreement between true and predicted values (kappa = 0.83) in such modelling can be achieved. Transfer learning techniques improve by < 10% with the performance of this model. We also show that spatial convolution neural networks are insufficient in determining the type of missed cancers. Ensembled hierarchical deep convolution machine-learning models are plausible in modelling radiologists' attentional level and their interpretation of mammograms. However, deep convolution networks fail to characterise the type of false-negative decisions.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Aprendizado Profundo , Mamografia/métodos , Reconhecimento Visual de Modelos , Interpretação de Imagem Radiográfica Assistida por Computador/métodos , Radiologistas , Mama/diagnóstico por imagem , Feminino , Humanos , Aprendizado de Máquina , Redes Neurais de ComputaçãoRESUMO
Inter-pathologist agreement for nuclear atypia scoring of breast cancer is poor. To address this problem, previous studies suggested some criteria for describing the variations appearance of tumor cells relative to normal cells. However, these criteria were still assessed subjectively by pathologists. Previous studies used quantitative computer-extracted features for scoring. However, application of these tools is limited as further improvement in their accuracy is required. This study proposes COMPASS (COMputer-assisted analysis combined with Pathologist's ASSessment) for reproducible nuclear atypia scoring. COMPASS relies on both cytological criteria assessed subjectively by pathologists as well as computer-extracted textural features. Using machine learning, COMPASS combines these two sets of features and output nuclear atypia score. COMPASS's performance was evaluated using 300 images for which expert-consensus derived reference nuclear pleomorphism scores were available, and they were scanned by two scanners from different vendors. A personalized model was built for three pathologists who gave scores to six atypia-related criteria for each image. Leave-one-out cross validation (LOOCV) was used. COMPASS was trained and tested for each pathologist separately. Percentage agreement between COMPASS and the reference nuclear scores was 93.8%, 92.9%, and 93.1% for three pathologists. COMPASS's performance in nuclear grading was almost identical for both scanners, with Cohen's kappa ranging from 0.80 to 0.86 for different pathologists and different scanners. Independently, the images were also assessed by two experienced senior pathologists. Cohen's kappa of COMPASS was comparable to the Cohen's kappa for two senior pathologists (0.79 and 0.68).
Assuntos
Neoplasias da Mama/patologia , Biópsia , Mama/patologia , Feminino , Humanos , Gradação de Tumores , Variações Dependentes do Observador , Patologistas , Reprodutibilidade dos Testes , Estudos RetrospectivosRESUMO
Purpose To compare three mammographic density measures defined by different pixel intensity thresholds as predictors of breast cancer risk for two different digital mammographic systems. Materials and Methods The Korean Breast Cancer Study included 398 women with invasive breast cancer and 737 control participants matched for age at mammography (±1 year), examination date, mammographic system, and menopausal status. Mammographic density was measured by using the automated Laboratory for Individualized Breast Radiodensity Assessment (LIBRA) software and the semiautomated Cumulus software at the conventional threshold (Cumulus) and at increasingly higher thresholds (Altocumulus and Cirrocumulus, respectively). Measures were Box-Cox-transformed and adjusted for age, body mass index, and menopausal status. Conditional logistic regression was used to estimate risk associations. For calculation of measures of predictive value, the change in odds per standard deviation (OPERA) and the area under the receiver operating characteristic curve (AUC) were used. Results For dense area, with use of the direct conversion system the OPERAs were 1.72 (95% confidence interval [CI]: 1.38, 2.15) for LIBRA, 1.58 (95% CI: 1.27, 1.97) for Cumulus, 2.04 (95% CI: 1.60, 2.59) for Altocumulus, and 3.48 (95% CI: 2.45, 4.47) for Cirrocumulus (P < .001). The corresponding AUCs were 0.70, 0.69, 0.76, and 0.89, respectively. With use of the indirect conversion system, the corresponding OPERAs were 1.50 (95% CI: 1.28, 1.76), 1.36 (95% CI: 1.16, 1.59), 1.40 (95% CI: 1.19, 1.64), and 1.47 (95% CI: 1.25, 1.73) (P < .001) and the AUCs were 0.64, 0.60, 0.61, and 0.63, respectively. Conclusion It is possible that mammographic density defined by higher pixel thresholds could capture more risk-predicting information with use of a direct conversion mammographic system; the mammographically bright, rather than white, regions are etiologically important. © RSNA, 2017.
Assuntos
Densidade da Mama , Neoplasias da Mama/patologia , Área Sob a Curva , Neoplasias da Mama/diagnóstico por imagem , Estudos de Casos e Controles , Detecção Precoce de Câncer , Feminino , Humanos , Mamografia/métodos , Pessoa de Meia-Idade , Fatores de RiscoRESUMO
OBJECTIVE: The purpose of this article is to review different x-ray phase-contrast breast imaging techniques and their potential application in clinical settings. CONCLUSION: Phase-contrast imaging depicts not only the absorption contrast but also the refraction contrast of the transmitted x-ray beam. Early data suggest that this new modality may overcome some of the diagnostic limitations associated with current clinically available mammography systems and that it has potential for improving breast cancer detection.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Mamografia/métodos , Intensificação de Imagem Radiográfica/métodos , Detecção Precoce de Câncer , Feminino , Humanos , Programas de Rastreamento , Sensibilidade e Especificidade , Raios XRESUMO
Breast cancer is the most common neoplasm diagnosed amongst women worldwide and is the leading cause of female cancer death. However, breast cancer in China is not comprehensively understood compared with Westernised countries, although the 5-year prevalence statistics indicate that approximately 11 % of worldwide breast cancer occurs in China and that the incidence has increased rapidly in recent decades. This paper reviews the descriptive epidemiology of Chinese breast cancer in terms of incidence, mortality, survival and prevalence, and explores relevant factors such as age of manifestation and geographic locations. The statistics are compared with data from the Westernised world with particular emphasis on the United States and Australia. Potential causal agents responsible for differences in breast cancer epidemiology between Chinese and other populations are also explored. The need to minimise variability and discrepancies in methods of data acquisition, analysis and presentation is highlighted.
Assuntos
Neoplasias da Mama/epidemiologia , Idade de Início , China/epidemiologia , Feminino , Humanos , Incidência , Pessoa de Meia-Idade , Prevalência , Sistema de Registros , Taxa de SobrevidaRESUMO
PURPOSE: To investigate opacification of head and neck vasculature during computed tomography (CT) of supraclavicular lymph nodes with a quadruple-phase contrast media and saline dual-injection protocol. MATERIALS AND METHODS: This retrospective study was institutional review board approved. In 180 consecutive patients, routine head and neck CT was performed with one of two protocols: protocol A, craniocaudal scan direction with 100 mL of contrast material injected intravenously as a single bolus; or protocol B, 100 mL of contrast material injected in four phases (phases 1-2, 60 mL of contrast material and saline injected at 2.5 mL/sec; phases 3-4, 40 mL of contrast material and saline injected at 2.5 mL/sec); both protocols had a fixed scan delay of 70 seconds. Attenuation of supraclavicular arteries and veins was measured with arteriovenous contrast ratio (AVCR) and contrast-to-noise ratio (CNR). Effective dose was calculated. Data were compared with the two-sample t test. Receiver operating characteristic (ROC) and visual grading characteristic analyses were performed. RESULTS: Arterial attenuation was up to 20% higher (P < .05) after protocol B (mean ± standard deviation, 234.5 HU ± 33.2) than protocol A (160.0 HU ± 29.5). Venous system attenuation was significantly lower in protocol B (164.0 HU ± 17.0) than in protocol A (664.0 HU ± 12.0), with up to a 75% reduction (P < .0001). Protocol B generated significant (P < .0001) improvements in AVCR at multiple anatomic sites. At all anatomic levels, mean CNR with protocol B (34.4 HU ± 9.0) was significantly higher than that with protocol A (14.5 HU ± 14.0) (P < .0313). Effective dose was significantly reduced with protocol B (2.6 mSv ± 0.4 vs 3.2 mSv ± 0.8 with protocol A; P < .0041). ROC analysis demonstrated significantly higher area under the ROC curve for protocol B (P < .0022), with interreader agreement increasing from poor to excellent in lymph node visualization. CONCLUSION: Significant improvement in lymph node visualization at the cervicothoracic junction is achieved with a quadruple-phase contrast media injection protocol.
Assuntos
Meios de Contraste/administração & dosagem , Neoplasias de Cabeça e Pescoço/diagnóstico por imagem , Metástase Neoplásica/diagnóstico por imagem , Tomografia Computadorizada por Raios X/métodos , Ácidos Tri-Iodobenzoicos/administração & dosagem , Artefatos , Feminino , Neoplasias de Cabeça e Pescoço/patologia , Humanos , Injeções Intravenosas , Masculino , Metástase Neoplásica/patologia , Doses de Radiação , Estudos RetrospectivosRESUMO
OBJECTIVES: To determine the impact of specific reporting tasks on the performance of radiologists when reading chest radiographs. METHODS: Ten experienced radiologists read a set of 40 postero-anterior (PA) chest radiographs: 21 nodule free and 19 with a proven solitary nodule. There were two reporting conditions: an unframed task (UFT) to report any abnormality and a framed task (FT) reporting only lung nodule/s. Jackknife free-response operating characteristic (JAFROC) figure of merit (FOM), specificity, location sensitivity and number of true positive (TP), false positive (FP), true negative (TN) and false negative (FN) decisions were used for analysis. RESULTS: JAFROC FOM for tasks showed a significant reduction in performance for framed tasks (P = 0.006) and an associated decrease in specificity (P = 0.011) but no alteration to the location sensitivity score. There was a significant increase in number of FP decisions made during framed versus unframed tasks for nodule-containing (P = 0.005) and nodule-free (P = 0.011) chest radiographs. No significant differences in TP were recorded. CONCLUSIONS: Radiologists report more FP decisions when given specific reporting instructions to search for nodules on chest radiographs. The relevance of clinical history supplied to radiologists is called into question and may induce a negative effect. KEY POINTS: ⢠Framed reporting tasks increases false positive rates when searching for pulmonary nodules ⢠False positive results were observed in both nodule-containing and nodule-free cases ⢠Radiologist's decision-making may be influenced by clinical history in thoracic imaging.
Assuntos
Competência Clínica , Neoplasias Pulmonares/diagnóstico por imagem , Nódulo Pulmonar Solitário/diagnóstico por imagem , Tomada de Decisão Clínica , Reações Falso-Positivas , Feminino , Humanos , Neoplasias Pulmonares/patologia , Masculino , Variações Dependentes do Observador , Interpretação de Imagem Radiográfica Assistida por Computador , Radiografia Torácica/métodos , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Nódulo Pulmonar Solitário/patologia , Tomografia Computadorizada por Raios X/métodosRESUMO
γ-Aminobutyric acid (GABA) has been implicated in several pain conditions, yet no study has systematically evaluated GABA levels in migraine using (1) H-MRS. The accurate detection, separation and quantification of GABA in individuals with migraine could elucidate the role of this neurotransmitter in migraine pathophysiology. Such information may eventually be useful in the diagnosis and development of more effective treatments for migraine. The aims of this study were therefore to compare the concentration of GABA+ in individuals with migraine with that in asymptomatic individuals, and to determine the diagnostic potential of GABA+ in the classification of those with or without migraine. In this case-control study, GABA+ levels in the brain were determined in 19 participants with migraine and 19 matched controls by (1) H-MRS using Mescher-Garwood point-resolved spectroscopy (MEGA-PRESS) sequence. The diagnostic accuracy of GABA+ for the detection of migraine and the optimal cut-off value were determined by receiver operating characteristic analysis. GABA+ levels were significantly higher (p = 0.002) in those with migraine [median, 1.41 institutional units (IU); interquartile range, 1.31-1.50 IU] than in controls (median, 1.18 IU; interquartile range, 1.12-1.35 IU). The GABA+ concentration appears to have good accuracy for the classification of individuals with or without migraine [area under the curve (95% confidence interval), 0.837 (0.71-0.96); p < 0.001]. The optimal GABA+ cut-off value for migraine was 1.30 IU, with a sensitivity of 84.2%, specificity of 68.4% and positive likelihood ratio of +2.67. The outcomes of this study suggest altered GABA metabolism in migraine. These results add to the scarce evidence on the putative role of GABA in migraine and provide a basis to further explore the causal relationship between GABA+ and the pathophysiology of migraine. This study also demonstrates that GABA+ concentration has good diagnostic accuracy for migraine. These findings offer new research and practice directions for migraine diagnosis.
Assuntos
Encéfalo/metabolismo , Transtornos de Enxaqueca/diagnóstico , Transtornos de Enxaqueca/metabolismo , Espectroscopia de Prótons por Ressonância Magnética/métodos , Ácido gama-Aminobutírico/metabolismo , Adulto , Biomarcadores/metabolismo , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Neurotransmissores/metabolismo , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Distribuição Tecidual , Regulação para Cima , Adulto JovemRESUMO
OBJECTIVES: The objectives are To to compare the diagnostic performance of combined digital breast tomosynthesis (DBT) and digital mammography (DM) with that of DM alone, as a function of radiologists' experience with DBT. METHODS: Ethical committee approval was obtained. Fifty cases (27 cancer, 23 normal), each containing both digital mammography (DM) and digital breast tomosynthesis (DBT) images, were reviewed by 26 radiologists, divided into three groups according to level of experience with DBT (none, workshop experience, and clinical experience). The radiologists' diagnostic performance using DM was compared with that using DM + DBT, and evaluated by area under receiver-operating characteristic curve (AUC), jackknife free-response receiver-operator characteristics figure of metric (JAFROC FOM), sensitivity, location sensitivity, and specificity. RESULTS: For all readers combined, performance using DM + DBT was significantly higher than for DM alone by both AUC (0.788 vs 0.681, p < 0.001) and JAFROC FOM (0.745 vs 0.621, p < 0.001). Similar results were obtained for readers with no DBT experience (AUC 0.775 vs 0.682, p = 0.004; JAFROC FOM 0.695 vs 0.603, p = 0.016) and with clinical DBT experience (AUC 0.789 vs 0.681, p = 0.042; and JAFROC FOM 0.764 vs 0.632, p = 0.031). CONCLUSIONS: Addition of DBT to DM significantly improves radiologists' diagnostic performance whether or not they have prior DBT experience. KEY POINTS: ⢠Adding DBT to DM increased the number of detected cancers ⢠DBT + DM led to more accurate localization of breast cancers than DM ⢠Addition of DBT improved radiologists' performance regardless of prior DBT experience ⢠High-volume radiologists with different DBT experience levels performed similarly on DM + DBT.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Competência Clínica , Mamografia/métodos , Intensificação de Imagem Radiográfica/métodos , Radiologia , Tomografia por Raios X/métodos , Feminino , Humanos , Curva ROC , Estudos Retrospectivos , Recursos HumanosRESUMO
The aim of this work is to investigate how radiologist expertise and image appearance may have an impact on inter-reader variability of mammographic density (MD) identification. Seventeen radiologists, divided into three expertise groups, were asked to manually segment the areas they consider to be MD in 40 clinical images. The variation in identification of MD for each image was quantified by finding the range of segmentation areas. The impact of radiologist expertise and image appearance on this variation was explored. The range of areas chosen by participating radiologists varied from 7 to 73% across the 40 images, with a mean range of 35 ± 13%. Participants with high expertise were more likely to choose similar areas to one another, compared to participants with medium and low expertise levels (mean range were 19 ± 10%, 29 ± 13% and 25 ± 14 %, respectively, p < 0.0001). There was a significantly higher average grey level for the area segmented by all radiologists as MD compared to the area of variation, with mean grey level value for 8-bit images being 146 ± 19 vs. 99 ± 14, respectively. MD segmentation borders were consistent in areas where there was a sharp intensity change within a short distance. In conclusion, radiologists with high expertise tend to have a higher agreement when identifying MD. Tissues which have a lower contrast and a less visually sharp gradient change at the interface between high density tissue and adipose background lead to inter-reader variation in choosing mammographic density.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Competência Clínica/estatística & dados numéricos , Glândulas Mamárias Humanas/anormalidades , Mamografia/estatística & dados numéricos , Densidade da Mama , Feminino , Humanos , Variações Dependentes do Observador , Reprodutibilidade dos TestesRESUMO
Current literature shows that radiologist experience does not affect detection tasks when the object does not require medical training to detect. However, the research was never sufficiently detailed to examine if the contrast detection threshold is also the same for radiologists versus nonradiologists. Previously, contrast threshold research was performed predominantly on nonradiologists. Therefore, any differences could lead to over- or under-estimation of the performance capabilities of radiologists. Fourteen readers, evenly divided between radiologists and nonradiologists, read a set of 150 mammogram-like images. The study was performed with the location of the objects known and unknown, requiring two separate readings. No difference in the contrast detection threshold between reader groups for either the location-unknown (4.9 just noticeable differences) or location-known (3.3 just noticeable differences) images was seen. The standard deviation for the location-unknown condition had no difference (p 0.91). But for the location-known condition, a significant difference (p 0.0009) was seen between radiologists and nonradiologists. No difference in contrast detection based on reader experience was observed, but decreased variance was seen with radiologists in the location-known condition.
Assuntos
Competência Clínica , Mamografia , Radiologia , Humanos , Variações Dependentes do Observador , Curva ROCRESUMO
OBJECTIVE: Radiologists can detect the gist of abnormal based on their rapid initial impression on a mammogram (ie, global gist signal [GGS]). This study explores (1) whether global radiomic (ie, computer-extracted) features can predict the GGS; and if so, (ii) what features are the most important drivers of the signals. METHODS: The GGS of cases in two extreme conditions was considered: when observers detect a very strong gist (high-gist) and when the gist of abnormal was not/poorly perceived (low-gist). Gist signals/scores from 13 observers reading 4191 craniocaudal mammograms were collected. As gist is a noisy signal, the gist scores from all observers were averaged and assigned to each image. The high-gist and low-gist categories contained all images in the fourth and first quartiles, respectively. One hundred thirty handcrafted global radiomic features (GRFs) per mammogram were extracted and utilized to construct eight separate machine learning random forest classifiers (All, Normal, Cancer, Prior-1, Prior-2, Missed, Prior-Visible, and Prior-Invisible) for characterizing high-gist from low-gist images. The models were trained and validated using the 10-fold cross-validation approach. The models' performances were evaluated by the area under receiver operating characteristic curve (AUC). Important features for each model were identified through a scree test. RESULTS: The Prior-Visible model achieved the highest AUC of 0.84 followed by the Prior-Invisible (0.83), Normal (0.82), Prior-1 (0.81), All (0.79), Prior-2 (0.77), Missed (0.75), and Cancer model (0.69). Cluster shade, standard deviation, skewness, kurtosis, and range were identified to be the most important features. CONCLUSIONS: Our findings suggest that GRFs can accurately classify high- from low-gist images. ADVANCES IN KNOWLEDGE: Global mammographic radiomic features can accurately predict high- from low-gist images with five features identified to be valuable in describing high-gist images. These are critical in providing better understanding of the mammographic image characteristics that drive the strength of the GGSs which could be exploited to advance breast cancer (BC) screening and risk prediction, enabling early detection and treatment of BC thereby further reducing BC-related deaths.
Assuntos
Neoplasias da Mama , Radiômica , Humanos , Feminino , Mamografia , Computadores , RadiologistasRESUMO
PURPOSE: To establish the extent to which test set reading can represent actual clinical reporting in screening mammography. MATERIALS AND METHODS: Institutional ethics approval was granted, and informed consent was obtained from each participating screen reader. The need for informed consent with respect to the use of patient materials was waived. Two hundred mammographic examinations were selected from examinations reported by 10 individual expert screen readers, resulting in 10 reader-specific test sets. Data generated from actual clinical reports were compared with three test set conditions: clinical test set reading with prior images, laboratory test set reading with prior images, and laboratory test set reading without prior images. A further set of five expert screen readers was asked to interpret a common set of images in two identical test set conditions to establish a baseline for intraobserver variability. Confidence scores (from 1 to 4) were assigned to the respective decisions made by readers. Region-of-interest (ROI) figures of merit (FOMs) and side-specific sensitivity and specificity were described for the actual clinical reporting of each reader-specific test set and were compared with those for the three test set conditions. Agreement between pairs of readings was performed by using the Kendall coefficient of concordance. RESULTS: Moderate or acceptable levels of agreement were evident (W = 0.69-0.73, P < .01) when describing group performance between actual clinical reporting and test set conditions that were reasonably close to the established baseline (W = 0.77, P < .01) and were lowest when prior images were excluded. Higher median values for ROI FOMs were demonstrated for the test set conditions than for the actual clinical reporting values; this was possibly linked to changes in sensitivity. CONCLUSION: Reasonable levels of agreement between actual clinical reporting and test set conditions can be achieved, although inflated sensitivity may be evident with test set conditions.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Mamografia , Competência Profissional , Tomada de Decisões , Diagnóstico Diferencial , Feminino , Humanos , Variações Dependentes do Observador , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
PURPOSE: To explore relationships between reader performance and reader characteristics in mammography for specific radiologist groupings on the basis of annual number of readings. MATERIALS AND METHODS: The institutional review board approved the study and waived the need for patient consent to use all images. Readers gave informed consent. One hundred sixteen radiologists independently reviewed 60 mammographic cases: 20 cases with cancer and 40 cases with normal findings. Readers located any visualized cancer, and levels of confidence were scored from 1 to 5. A jackknifing free response operating characteristic (JAFROC) method was used, and figures of merit along with sensitivity and specificity were correlated with reader characteristics by using Spearman techniques and standard multiple regressions. RESULTS: Reader performance was positively correlated with number of years since qualification as a radiologist (P ≤ .01), number of years reading mammograms (P ≤ .03), and number of readings per year (P ≤ .0001). The number of years since qualification as a radiologist (P ≤ .004) and number of years of reading mammograms (P ≤ .002) were negatively related to JAFROC values for radiologists with annual volumes of less than 1000 mammographic readings. For individuals with more than 5000 mammographic readings per year, JAFROC values were positively related to the number of years that the reader was qualified as a radiologist (P ≤ .01), number of years of reading mammograms (P ≤ .002), and number of hours per week of reading mammograms (P ≤ .003). Number of mammographic readings per year was positively related with JAFROC scores for readers with an annual volume between 1000 and 5000 readings (P ≤ .03). Differences in JAFROC scores appear to be more related to specificity than location sensitivity, with the former demonstrating significant relationships with four of the five characteristics analyzed, whereas no relationships were shown for the latter. CONCLUSION: Radiologists' determinants of performance are associated with annual reading volumes. Ability to recognize normal images is a discriminating factor in individuals with a high volume of mammographic readings.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Neoplasias da Mama/epidemiologia , Interpretação de Imagem Assistida por Computador/métodos , Mamografia/estatística & dados numéricos , Competência Profissional/estatística & dados numéricos , Adulto , Idoso , Feminino , Humanos , Aumento da Imagem/métodos , Pessoa de Meia-Idade , New South Wales/epidemiologia , Variações Dependentes do Observador , Prevalência , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , UltrassonografiaRESUMO
OBJECTIVES: Caudocranial scan direction and contrast injection timing based on measured patient vessel dynamics can significantly improve artery opacification and reduce contrast dose in the assessment of acute aortic syndrome using gated and non-gated thoracic CTA. This study aimed to investigate enhancement of the thoracic aorta using caudocranial scan direction and a patient-specific contrast regimen. METHODS: Electrocardiogram-gated (n = 120) and non-gated (n = 200) thoracic computed tomography angiography was performed on patients with nontraumatic acute aortic syndrome. Patients were assigned to one of 2 acquisition/contrast regimens, namely, regimen A, craniocaudal scan direction with 120 mL contrast, and regimen B, caudocranial scan direction using a patient-specific contrast formula. Opacity of 9 arterial and venous segments was measured, arteriovenous contrast ratio calculated, and values compared using Mann-Whitney U statistics. Receiver operating characteristic analyses and visual grading characteristic assessed diagnostic efficacy and clinical image quality. Interobserver variations were investigated using κ methods. RESULTS: Regimen B when compared to A, for both scanning/contrast techniques, demonstrated higher opacification in the aorta (P < 0.01) and lower opacification in the venous system (P < 0.0001). For protocol B, arteriovenous contrast ratio was significantly increased (P < 0.0001) and mean contrast volume reduced (P < 0.05) during gated [94 (10 mL)] and non-gated [78 (5 mL)] thoracic computed tomography angiography compared to A. Receiver operating characteristic analysis Az scores and interobserver agreement were significantly higher with regimen B than A (P < 0.05). CONCLUSIONS: Caudocranial scan direction and injection timing based on patient-specific vessel dynamics can optimize artery opacification and diagnostic efficacy while reducing contrast volumes.
Assuntos
Angiografia/métodos , Doenças da Aorta/diagnóstico por imagem , Técnicas de Imagem de Sincronização Cardíaca/métodos , Meios de Contraste/administração & dosagem , Posicionamento do Paciente/métodos , Radiografia Torácica/métodos , Tomografia Computadorizada por Raios X/métodos , Doença Aguda , Adolescente , Adulto , Idoso , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Intensificação de Imagem Radiográfica/métodos , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Síndrome , Adulto JovemRESUMO
This study aimed to determine if phantom-based methodologies for optimization of hepatic lesion detection with computed tomography (CT) require randomization of lesion placement and inclusion of normal images. A phantom containing fixed opacities of varying size (diameters, 2.4, 4.8, and 9.5 mm) was scanned at various exposure and slice thickness settings. Two image sets were compared: All images in the first image set contained opacities with known location; the second image set contained images with opacities in random locations. Following Institutional Review Board approval, nine experienced observers scored opacity visualization using a 4-point confidence scale. Comparisons between image sets were performed using Spearman, Kappa, and Wilcoxon techniques. Observer scores demonstrated strong correlation between both approaches when all opacity sizes were combined (r = 0.92, p < 0.0001), for the 9.5 mm opacity (r = 0.96, p < 0.0001) and for the 2.4 mm opacity (r = 0.64, p < 0.05). There was no significant correlation for the 4.8 mm opacity. A significantly higher sensitivity score for the known compared with the unknown location was found for the 9.5 mm opacity and 4.8 mm opacity for a single slice thickness and exposure condition (p < 0.05). Phantom-based optimization of CT hepatic examinations requires randomized lesion location when investigating challenging conditions; however, a standard phantom with fixed lesion location is suitable for the optimization of routine liver protocols. The development of more sophisticated phantoms or methods than those currently available is indicated for the optimization of CT protocols for diagnostic tasks involving the detection of subtle change.
Assuntos
Fígado/diagnóstico por imagem , Imagens de Fantasmas/normas , Interpretação de Imagem Radiográfica Assistida por Computador/métodos , Tomografia Computadorizada por Raios X/métodos , Algoritmos , Estudos de Avaliação como Assunto , Feminino , Humanos , Processamento de Imagem Assistida por Computador , Fígado/patologia , Masculino , Variações Dependentes do Observador , Doses de Radiação , Valores de Referência , Medição de Risco , Sensibilidade e EspecificidadeRESUMO
This study measured reading workstation monitors and the viewing environment currently available within BreastScreen New South Wales (BSNSW) centres to determine levels of adherence to national and international guidelines. Thirteen workstations from four BSNSW service centres were assessed using the American Association of Physicists in Medicine Task Group 18 Quality Control test pattern. Reading workstation monitor performance and ambient light levels when interpreting screening mammographic images were assessed using spectroradiometer CS-2000 and chroma meter CL-200. Overall, radiologic monitors within BSNSW were operating at good acceptable levels. Some non-adherence to published guidelines included the percentage difference in maximum luminance between pairs of primary monitors at individual workstations (61.5 % or 30.8 % of workstations depending on specific guidelines), maximum luminance (23.1 % of workstations), luminance non-uniformity (11.5 % of workstations) and minimum luminance (3.8 % of workstations). A number of ambient light measurements did not comply with the only available evidence-based guideline relevant to the methodology used in this study. Larger ambient light variations across sites are shown when monitors were switched off, suggesting that differences in ambient lighting between sites can be masked when a standard mammogram is displayed for photometric measurements. Overall, BSNSW demonstrated good adherence to available guidelines, although some non-compliance has been shown. Recently updated United Kingdom and Australian guidelines should help reduce confusion generated by the plethora and sometimes dated nature of currently available recommendations.