Free-response Receiver Operating Characteristic Analysis Research Articles

Semisupervised Learning with Report-guided Pseudo Labels for Deep Learning-based Prostate Cancer Detection Using Biparametric MRI.

To evaluate a novel method of semisupervised learning (SSL) guided by automated sparse information from diagnostic reports to leverage additional data for deep learning-based malignancy detection in patients with clinically significant prostate cancer. This retrospective study included 7756 prostate MRI examinations (6380 patients) performed between January 2014 and December 2020 for model development. An SSL method, report-guided SSL (RG-SSL), was developed for detection of clinically significant prostate cancer using biparametric MRI. RG-SSL, supervised learning (SL), and state-of-the-art SSL methods were trained using 100, 300, 1000, or 3050 manually annotated examinations. Performance on detection of clinically significant prostate cancer by RG-SSL, SL, and SSL was compared on 300 unseen examinations from an external center with a histopathologically confirmed reference standard. Performance was evaluated using receiver operating characteristic (ROC) and free-response ROC analysis. P values for performance differences were generated with a permutation test. At 100 manually annotated examinations, mean examination-based diagnostic area under the ROC curve (AUC) values for RG-SSL, SL, and the best SSL were 0.86 ± 0.01 (SD), 0.78 ± 0.03, and 0.81 ± 0.02, respectively. Lesion-based detection partial AUCs were 0.62 ± 0.02, 0.44 ± 0.04, and 0.48 ± 0.09, respectively. Examination-based performance of SL with 3050 examinations was matched by RG-SSL with 169 manually annotated examinations, thus requiring 14 times fewer annotations. Lesion-based performance was matched with 431 manually annotated examinations, requiring six times fewer annotations. RG-SSL outperformed SSL in clinically significant prostate cancer detection and achieved performance similar to SL even at very low annotation budgets.Keywords: Annotation Efficiency, Computer-aided Detection and Diagnosis, MRI, Prostate Cancer, Semisupervised Deep Learning Supplemental material is available for this article. Published under a CC BY 4.0 license.

Breast cancer detection with upstream data fusion, machine learning, and automated registration: initial results.

To develop an artificial intelligence algorithm for the detection of breast cancer by combining upstream data fusion (UDF), machine learning (ML), and automated registration, using digital breast tomosynthesis (DBT) and breast ultrasound (US). Our retrospective study included examinations from 875 women obtained between April 2013 and January 2019. Included patients had a DBT mammogram, breast US, and biopsy proven breast lesion. Images were annotated by a breast imaging radiologist. An AI algorithm was developed based on ML for image candidate detections and UDF for fused detections. After exclusions, images from 150 patients were evaluated. Ninety-five cases were used for training and validation of ML. Fifty-five cases were included in the UDF test set. UDF performance was evaluated with a free-response receiver operating characteristic (FROC) curve. Forty percent of cases evaluated with UDF (22/55) yielded true ML detections in all three images (craniocaudal DBT, mediolateral oblique DBT, and US). Of these, 20/22 (90.9%) produced a UDF fused detection that contained and classified the lesion correctly. FROC analysis for these cases showed 90% sensitivity at 0.3 false positives per case. In contrast, ML yielded an average of 8.0 false alarms per case. An AI algorithm combining UDF, ML, and automated registration was developed and applied to test cases, showing that UDF can yield fused detections and decrease false alarms when applied to breast cancer detection. Improvement of ML detection is needed to realize the full benefit of UDF.

Automated deep-learning system in the assessment of MRI-visible prostate cancer: comparison of advanced zoomed diffusion-weighted imaging and conventional technique

BackgroundDeep-learning-based computer-aided diagnosis (DL-CAD) systems using MRI for prostate cancer (PCa) detection have demonstrated good performance. Nevertheless, DL-CAD systems are vulnerable to high heterogeneities in DWI, which can interfere with DL-CAD assessments and impair performance. This study aims to compare PCa detection of DL-CAD between zoomed-field-of-view echo-planar DWI (z-DWI) and full-field-of-view DWI (f-DWI) and find the risk factors affecting DL-CAD diagnostic efficiency.MethodsThis retrospective study enrolled 354 consecutive participants who underwent MRI including T2WI, f-DWI, and z-DWI because of clinically suspected PCa. A DL-CAD was used to compare the performance of f-DWI and z-DWI both on a patient level and lesion level. We used the area under the curve (AUC) of receiver operating characteristics analysis and alternative free-response receiver operating characteristics analysis to compare the performances of DL-CAD using f- DWI and z-DWI. The risk factors affecting the DL-CAD were analyzed using logistic regression analyses. P values less than 0.05 were considered statistically significant.ResultsDL-CAD with z-DWI had a significantly better overall accuracy than that with f-DWI both on patient level and lesion level (AUCpatient: 0.89 vs. 0.86; AUClesion: 0.86 vs. 0.76; P < .001). The contrast-to-noise ratio (CNR) of lesions in DWI was an independent risk factor of false positives (odds ratio [OR] = 1.12; P < .001). Rectal susceptibility artifacts, lesion diameter, and apparent diffusion coefficients (ADC) were independent risk factors of both false positives (ORrectal susceptibility artifact = 5.46; ORdiameter, = 1.12; ORADC = 0.998; all P < .001) and false negatives (ORrectal susceptibility artifact = 3.31; ORdiameter = 0.82; ORADC = 1.007; all P ≤ .03) of DL-CAD.ConclusionsZ-DWI has potential to improve the detection performance of a prostate MRI based DL-CAD.Trial registrationChiCTR, NO. ChiCTR2100041834. Registered 7 January 2021.

Rib fracture detection system based on deep learning

Rib fracture detection is time-consuming and demanding work for radiologists. This study aimed to introduce a novel rib fracture detection system based on deep learning which can help radiologists to diagnose rib fractures in chest computer tomography (CT) images conveniently and accurately. A total of 1707 patients were included in this study from a single center. We developed a novel rib fracture detection system on chest CT using a three-step algorithm. According to the examination time, 1507, 100 and 100 patients were allocated to the training set, the validation set and the testing set, respectively. Free Response ROC analysis was performed to evaluate the sensitivity and false positivity of the deep learning algorithm. Precision, recall, F1-score, negative predictive value (NPV) and detection and diagnosis were selected as evaluation metrics to compare the diagnostic efficiency of this system with radiologists. The radiologist-only study was used as a benchmark and the radiologist-model collaboration study was evaluated to assess the model’s clinical applicability. A total of 50,170,399 blocks (fracture blocks, 91,574; normal blocks, 50,078,825) were labelled for training. The F1-score of the Rib Fracture Detection System was 0.890 and the precision, recall and NPV values were 0.869, 0.913 and 0.969, respectively. By interacting with this detection system, the F1-score of the junior and the experienced radiologists had improved from 0.796 to 0.925 and 0.889 to 0.970, respectively; the recall scores had increased from 0.693 to 0.920 and 0.853 to 0.972, respectively. On average, the diagnosis time of radiologist assisted with this detection system was reduced by 65.3 s. The constructed Rib Fracture Detection System has a comparable performance with the experienced radiologist and is readily available to automatically detect rib fracture in the clinical setting with high efficacy, which could reduce diagnosis time and radiologists’ workload in the clinical practice.

Image quality in liver CT: low-dose deep learning vs standard-dose model-based iterative reconstructions

To compare the overall image quality and detectability of significant (malignant and pre-malignant) liver lesions of low-dose liver CT (LDCT, 33.3% dose) using deep learning denoising (DLD) to standard-dose CT (SDCT, 100% dose) using model-based iterative reconstruction (MBIR). In this retrospective study, CT images of 80 patients with hepatic focal lesions were included. For noninferiority analysis of overall image quality, a margin of - 0.5 points (scored in a 5-point scale) for the difference between scan protocols was pre-defined. Other quantitative or qualitative image quality assessments were performed. Additionally, detectability of significant liver lesions was compared, with 64 pairs of CT, using the jackknife alternative free-response ROC analysis, with noninferior margin defined by the lower limit of 95% confidence interval (CI) of the difference of figure-of-merit less than - 0.1. The mean overall image quality scores with LDCT and SDCT were 3.77 ± 0.38 and 3.94 ± 0.34, respectively, demonstrating a difference of - 0.17 (95% CI: - 0.21 to - 0.12), which did not cross the predefined noninferiority margin of - 0.5. Furthermore, LDCT showed significantly superior quantitative results of liver lesion contrast to noise ratio (p < 0.05). However, although LDCT scored higher than the average score in qualitative image quality assessments, they were significantly lower than those of SDCT (p < 0.05). Figure-of-merit for lesion detection was 0.859 for LDCT and 0.878 for SDCT, showing noninferiority (difference: - 0.019, 95% CI: - 0.058 to 0.021). LDCT using DLD with 67% radiation dose reduction showed non-inferior overall image quality and lesion detectability, compared to SDCT. • Low-dose liver CT using deep learning denoising (DLD), at 67% dose reduction, provided non-inferior overall image quality compared to standard-dose CT using model-based iterative reconstruction (MBIR). • Low-dose CT using DLD showed significantly less noise and higher CNR lesion to liver than standard-dose CT using MBIR and demonstrated at least average image quality score among all readers, albeit with lower scores than standard-dose CT using MBIR. • Low-dose liver CT showed noninferior detectability for malignant and pre-malignant liver lesions, compared to standard-dose CT.

Deep neural network-based detection and segmentation of intracranial aneurysms on 3D rotational DSA.

Accurate diagnosis and measurement of intracranial aneurysms are challenging. This study aimed to develop a 3D convolutional neural network (CNN) model to detect and segment intracranial aneurysms (IA) on 3D rotational DSA (3D-RA) images. 3D-RA images were collected and annotated by 5 neuroradiologists. The annotated images were then divided into three datasets: training, validation, and test. A 3D Dense-UNet-like CNN (3D-Dense-UNet) segmentation algorithm was constructed and trained using the training dataset. Diagnostic performance to detect aneurysms and segmentation accuracy was assessed for the final model on the test dataset using the free-response receiver operating characteristic (FROC). Finally, the CNN-inferred maximum diameter was compared against expert measurements by Pearson's correlation and Bland-Altman limits of agreement (LOA). A total of 451 patients with 3D-RA images were split into n = 347/41/63 training/validation/test datasets, respectively. For aneurysm detection, observed FROC analysis showed that the model managed to attain a sensitivity of 0.710 at 0.159 false positives (FP)/case, and 0.986 at 1.49 FP/case. The proposed method had good agreement with reference manual aneurysmal maximum diameter measurements (8.3 ± 4.3 mm vs. 7.8 ± 4.8 mm), with a correlation coefficient r = 0.77, small bias of 0.24 mm, and LOA of -6.2 to 5.71 mm. 37.0% and 77% of diameter measurements were within ±1 mm and ±2.5 mm of expert measurements. A 3D-Dense-UNet model can detect and segment aneurysms with relatively high accuracy using 3D-RA images. The automatically measured maximum diameter has potential clinical application value.

Clinical usefulness of temporal subtraction CT in detecting vertebral bone metastases

PurposeThe purpose of this study was to determine whether temporal subtraction (TS) computed tomography (CT) contributes to the detection of vertebral bone metastases. MethodThe calculation of TS CT was composed of bony landmark detection, bone segmentation with a multiatlas-based method, and spatial registration. Temporal increase and decrease of the CT values were visualized in blue and red, respectively. Paired CT images of 20 patients with cancer and newly-developed vertebral metastases were analyzed. Control CT examinations of 20 different patients were also included. The presence of vertebral metastases on the TS CT was evaluated by two board-certified radiologists. Five additional board-certified radiologists and five radiology residents independently interpreted the 40 paired CT images with and without TS CT. ResultsIn the lesion conspicuity evaluation, 96% of vertebral metastases were scored as excellent or good. In the image interpretation examination, according to free-response receiver operating characteristics analysis, the overall figure of merit (FOM) of the board-certified radiologist group was 0.892 and 0.898 with and without TS CT, respectively. The FOM of the resident group improved from 0.849 to 0.902 with viewing TS CT. In the sub-analysis focusing on the location of the lesion, the FOM of the resident group significantly improved from 0.75 to 0.92 in vertebral arch lesions (p = 0.001). ConclusionsThe TS CT may be useful to detect vertebral metastases because almost all the vertebral metastases were shown to be favorable visualization. The TS CT was proven to be especially helpful for radiology residents in detecting vertebral arch metastases.

Detection of suspected brain infarctions on CT can be significantly improved with temporal subtraction images.

To assess whether temporal subtraction (TS) images of brain CT improve the detection of suspected brain infarctions. Study protocols were approved by our institutional review board, and informed consent was waived because of the retrospective nature of this study. Forty-two sets of brain CT images of 41 patients, each consisting of a pair of brain CT images scanned at two time points (previous and current) between January 2011 and November 2016, were collected for an observer performance study. The 42 sets consisted of 23 cases with a total of 77 newly developed brain infarcts or hyperdense artery signs confirmed by two radiologists who referred to additional clinical information and 19 negative control cases. To create TS images, the previous images were registered to the current images by partly using a non-rigid registration algorithm and then subtracted. Fourteen radiologists independently interpreted the images to identify the lesions with and without TS images with an interval of over 4 weeks. A figure of merit (FOM) was calculated along with the jackknife alternative free-response receiver-operating characteristic analysis. Sensitivity, number of false positives per case (FPC) and reading time were analyzed by the Wilcoxon signed-rank test. The mean FOM increased from 0.528 to 0.737 with TS images (p < 0.0001). The mean sensitivity and FPC improved from 26.5% and 0.243 to 56.0% and 0.153 (p < 0.0001 and p = 0.239), respectively. The mean reading time was 173 s without TS and 170 s with TS (p = 0.925). The detectability of suspected brain infarctions was significantly improved with TS CT images. • Although it is established that MRI is superior to CT in the detection of strokes, the first choice of modality for suspected stroke patients is often CT. • An observer performance study with 14 radiologists was performed to evaluate whether temporal subtraction images derived from a non-rigid transformation algorithm can significantly improve the detectability of newly developed brain infarcts on CT. • Temporal subtraction images were shown to significantly improve the detectability of newly developed brain infarcts on CT.

Cascade convolutional neural networks for automatic detection of thyroid nodules in ultrasound images.

It is very important for calculation of clinical indices and diagnosis to detect thyroid nodules from ultrasound images. However, this task is a challenge mainly due to heterogeneous thyroid nodules with distinct components are similar to background in ultrasound images. In this study, we employ cascade deep convolutional neural networks (CNNs) to develop and evaluate a fully automatic detection of thyroid nodules from 2D ultrasound images. Our cascade CNNs are a type of hybrid model, consisting of two different CNNs and a new splitting method. Specifically, it employs a deep CNN to learn the segmentation probability maps from the ground true data. Then, all the segmentation probability maps are split into different connected regions by the splitting method. Finally, another deep CNN is used to automatically detect the thyroid nodules from ultrasound thyroid images. Experiment results illustrate the cascade CNNs are very effective in detection of thyroid nodules. Specially, the value of area under the curve of receiver operating characteristic is 98.51%. The Free-response receiver operating characteristic (FROC) and jackknife alternative FROC (JAFROC) analyses show a significant improvement in the performance of our cascade CNNs compared to that of other methods. The multi-view strategy can improve the performance of cascade CNNs. Moreover, our special splitting method can effectively separate different connected regions so that the second CNN can correctively gain the positive and negative samples according to the automatic labels. The experiment results demonstrate the potential clinical applications of this proposed method. This technique can offer physicians an objective second opinion, and reduce their heavy workload so as to avoid misdiagnosis causes because of excessive fatigue. In addition, it is easy and reproducible for a person without medical expertise to diagnose thyroid nodules.

A JAFROC study of nodule detection performance in CT images of a thorax acquired during PET/CT

PurposeTwo types of CT images (modalities) are acquired in PET/CT: for attenuation correction (AC) and diagnosis. The purpose of the study was to compare nodule detection and localization performance between these two modalities. MethodsCT images, using both modalities, of an anthropomorphic chest phantom containing zero or more simulated spherical nodules of 5, 8, 10 and 12 mm diameters and contrasts −800, −630 and 100 HU were acquired. An observer performance study using nine observers interpreting 45 normal (zero nodules) images and 47 abnormal images (1–3 nodules; average 1.26) was conducted using the free-response receiver operating characteristic (FROC) paradigm. Data were analysed using an R software package implemented jackknife alternative FROC (JAFROC) analysis. Both empirical areas under the equally weighted AFROC curve (wAFROC) and under the highest rating inferred ROC (HR-ROC) curve were used as figures of merit (FOM). To control the probability of Type I error test alpha was set at 0.05. ResultsNodule detection as measured by either FOM was significantly better on the diagnostic quality images (2nd modality), irrespective of the method of analysis, [reader averaged inter-modality wAFROC FOM difference = −0.07 (−0.11,−0.04); reader averaged inter-modality HR-ROC FOM difference = −0.05 (−0.09, −0.01)]. ConclusionNodule detection was statistically worse on images acquired for AC; suggesting that images acquired for AC should not be used to evaluate pulmonary pathology.

“Pine tree” appearance on 18F-FDG PET/CT MIP image in spinal tuberculosis

Two types of CT images (modalities) are acquired in PET/CT: for attenuation correction (AC) and diagnosis. The purpose of the study was to compare nodule detection and localization performance between these two modalities.CT images, using both modalities, of an anthropomorphic chest phantom containing zero or more simulated spherical nodules of 5, 8, 10 and 12 mm diameters and contrasts −800, −630 and 100 HU were acquired. An observer performance study using nine observers interpreting 45 normal (zero nodules) images and 47 abnormal images (1–3 nodules; average 1.26) was conducted using the free-response receiver operating characteristic (FROC) paradigm. Data were analysed using an R software package implemented jackknife alternative FROC (JAFROC) analysis. Both empirical areas under the equally weighted AFROC curve (wAFROC) and under the highest rating inferred ROC (HR-ROC) curve were used as figures of merit (FOM). To control the probability of Type I error test alpha was set at 0.05.Nodule detection as measured by either FOM was significantly better on the diagnostic quality images (2nd modality), irrespective of the method of analysis, [reader averaged inter-modality wAFROC FOM difference = −0.07 (−0.11,−0.04); reader averaged inter-modality HR-ROC FOM difference = −0.05 (−0.09, −0.01)].Nodule detection was statistically worse on images acquired for AC; suggesting that images acquired for AC should not be used to evaluate pulmonary pathology.

Improved cancer detection in automated breast ultrasound by radiologists using Computer Aided Detection

ObjectiveTo investigate the effect of dedicated Computer Aided Detection (CAD) software for automated breast ultrasound (ABUS) on the performance of radiologists screening for breast cancer. Methods90 ABUS views of 90 patients were randomly selected from a multi-institutional archive of cases collected between 2010 and 2013. This dataset included normal cases (n=40) with >1year of follow up, benign (n=30) lesions that were either biopsied or remained stable, and malignant lesions (n=20). Six readers evaluated all cases with and without CAD in two sessions. CAD-software included conventional CAD-marks and an intelligent minimum intensity projection of the breast tissue. Readers reported using a likelihood-of-malignancy scale from 0 to 100. Alternative free-response ROC analysis was used to measure the performance. ResultsWithout CAD, the average area-under-the-curve (AUC) of the readers was 0.77 and significantly improved with CAD to 0.84 (p=0.001). Sensitivity of all readers improved (range 5.2–10.6%) by using CAD but specificity decreased in four out of six readers (range 1.4–5.7%). No significant difference was observed in the AUC between experienced radiologists and residents both with and without CAD. ConclusionsDedicated CAD-software for ABUS has the potential to improve the cancer detection rates of radiologists screening for breast cancer.

Detecting Cardiovascular Disease from Mammograms With Deep Learning.

Coronary artery disease is a major cause of death in women. Breast arterial calcifications (BACs), detected inmammograms, can be useful riskmarkers associated with the disease. We investigate the feasibility of automated and accurate detection ofBACsinmammograms for risk assessment of coronary artery disease. We develop a 12-layer convolutional neural network to discriminate BAC from non-BAC and apply a pixelwise, patch-based procedure for BAC detection. To assess the performance of the system, we conduct a reader study to provide ground-truth information using the consensus of human expert radiologists. We evaluate the performance using a set of 840 full-field digital mammograms from 210 cases, using both free-responsereceiveroperatingcharacteristic (FROC) analysis and calcium mass quantification analysis. The FROC analysis shows that the deep learning approach achieves a level of detection similar to the human experts. The calcium mass quantification analysis shows that the inferred calcium mass is close to the ground truth, with a linear regression between them yielding a coefficient of determination of 96.24%. Taken together, these results suggest that deep learning can be used effectively to develop an automated system for BAC detection inmammograms to help identify and assess patients with cardiovascular risks.

Analysis of computer-aided detection techniques and signal characteristics for clustered microcalcifications on digital mammography and digital breast tomosynthesis

With IRB approval, digital breast tomosynthesis (DBT) images of human subjects were collected using a GE GEN2 DBT prototype system. Corresponding digital mammograms (DMs) of the same subjects were collected retrospectively from patient files. The data set contained a total of 237 views of DBT and equal number of DM views from 120 human subjects, each included 163 views with microcalcification clusters (MCs) and 74 views without MCs. The data set was separated into training and independent test sets. The pre-processing, object prescreening and segmentation, false positive reduction and clustering strategies for MC detection by three computer-aided detection (CADe) systems designed for DM, DBT, and a planar projection image generated from DBT were analyzed. Receiver operating characteristic (ROC) curves based on features extracted from microcalcifications and free-response ROC (FROC) curves based on scores from MCs were used to quantify the performance of the systems. Jackknife FROC (JAFROC) and non-parametric analysis methods were used to determine the statistical difference between the FROC curves. The difference between the CADDM and CADDBT systems when the false positive rate was estimated from cases without MCs did not reach statistical significance. The study indicates that the large search space in DBT may not be a limiting factor for CADe to achieve similar performance as that observed in DM.

Improving the accuracy in detection of clustered microcalcifications with a context-sensitive classification model.

In computer-aided detection of microcalcifications (MCs), the detection accuracy is often compromised by frequent occurrence of false positives (FPs), which can be attributed to a number of factors, including imaging noise, inhomogeneity in tissue background, linear structures, and artifacts in mammograms. In this study, the authors investigated a unified classification approach for combating the adverse effects of these heterogeneous factors for accurate MC detection. To accommodate FPs caused by different factors in a mammogram image, the authors developed a classification model to which the input features were adapted according to the image context at a detection location. For this purpose, the input features were defined in two groups, of which one group was derived from the image intensity pattern in a local neighborhood of a detection location, and the other group was used to characterize how a MC is different from its structural background. Owing to the distinctive effect of linear structures in the detector response, the authors introduced a dummy variable into the unified classifier model, which allowed the input features to be adapted according to the image context at a detection location (i.e., presence or absence of linear structures). To suppress the effect of inhomogeneity in tissue background, the input features were extracted from different domains aimed for enhancing MCs in a mammogram image. To demonstrate the flexibility of the proposed approach, the authors implemented the unified classifier model by two widely used machine learning algorithms, namely, a support vector machine (SVM) classifier and an Adaboost classifier. In the experiment, the proposed approach was tested for two representative MC detectors in the literature [difference-of-Gaussians (DoG) detector and SVM detector]. The detection performance was assessed using free-response receiver operating characteristic (FROC) analysis on a set of 141 screen-film mammogram (SFM) images (66 cases) and a set of 188 full-field digital mammogram (FFDM) images (95 cases). The FROC analysis results show that the proposed unified classification approach can significantly improve the detection accuracy of two MC detectors on both SFM and FFDM images. Despite the difference in performance between the two detectors, the unified classifiers can reduce their FP rate to a similar level in the output of the two detectors. In particular, with true-positive rate at 85%, the FP rate on SFM images for the DoG detector was reduced from 1.16 to 0.33 clusters/image (unified SVM) and 0.36 clusters/image (unified Adaboost), respectively; similarly, for the SVM detector, the FP rate was reduced from 0.45 clusters/image to 0.30 clusters/image (unified SVM) and 0.25 clusters/image (unified Adaboost), respectively. Similar FP reduction results were also achieved on FFDM images for the two MC detectors. The proposed unified classification approach can be effective for discriminating MCs from FPs caused by different factors (such as MC-like noise patterns and linear structures) in MC detection. The framework is general and can be applicable for further improving the detection accuracy of existing MC detectors.

Detection of masses in digital breast tomosynthesis using complementary information of simulated projection.

The purpose of this study is to develop a computer-aided detection system that combines the detection results in 3D digital breast tomosynthesis (DBT) volume and 2D simulated projection (synthesized image which is not provided by the vendor but generated from DBT volume in this study) to improve the accuracy of mass detection in DBT. The 3D DBT volume has a problem of blurring in the out-of-focus plane because it is reconstructed from a limited number of projection view images acquired over a limited angular range. To solve the problem, the simulated projection is generated by measuring the blurriness of voxels in the DBT volume and adopting conspicuity voxels. A contour-based detection algorithm is applied to detecting masses in the simulated projection. The DBT volume is analyzed by using an unsupervised mass detection algorithm, which results in mass candidates in the DBT volume. The mass likelihood scores estimated for mass candidates on the DBT volume and the simulated projection are merged in a probabilistic manner through a Bayesian network model to differentiate masses and false positives (FPs). Experiments were conducted on a clinical data set of 320 DBT volumes. In 90 volumes, at least one biopsy-proven malignant mass was presented. The longest diameter of masses ranged from 7.0 to 56.4 mm (mean = 25.4 mm). The sizes of masses in the data set were relatively large compared to the sizes of the masses reported in other detection studies. Three image quality measurements (overall sharpness, sharpness of mass boundary, and contrast) were used to evaluate the image quality of the simulated projection compared to the DBT central slice where the mass was most conspicuous and other projection methods (maximum intensity projection and average projection). A free-response receiver operating characteristic (FROC) analysis was adopted for evaluating the accuracy of mass detection in the DBT volume, the simulated projection, and the combined approach. A jackknife FROC analysis (JAFROC) was used to estimate the statistical significance of the difference between two FROC curves. The overall sharpness and the sharpness of mass boundary in the simulated projection are higher than those in the DBT central slice and other projection methods. The contrast of the simulated projection is lower than the DBT central slice. The mass detection in the DBT volume achieved region-based sensitivities of 80% and 85% with 1.75 and 2.11 FPs per DBT volume. The proposed combined mass detection approach achieved same sensitivities with reduced FPs of 1.33 and 1.93 per DBT volume. The difference of the FROC curves between the combined approach and the mass detection in the DBT volume was statistically significant (p < 0.01) by JAFROC analysis. This study indicates that the combined approach that merges the detection results in the DBT volume and the simulated projection is a promising approach to improve the accuracy of mass detection in DBT.

A Metric for Reducing False Positives in the Computer-Aided Detection of Breast Cancer from Dynamic Contrast-Enhanced Magnetic Resonance Imaging Based Screening Examinations of High-Risk Women.

Magnetic resonance imaging (MRI)-enabled cancer screening has been shown to be a highly sensitive method for the early detection of breast cancer. Computer-aided detection systems have the potential to improve the screening process by standardizing radiologists to a high level of diagnostic accuracy. This retrospective study was approved by the institutional review board of Sunnybrook Health Sciences Centre. This study compares the performance of a proposed method for computer-aided detection (based on the second-order spatial derivative of the relative signal intensity) with the signal enhancement ratio (SER) on MRI-based breast screening examinations. Comparison is performed using receiver operating characteristic (ROC) curve analysis as well as free-response receiver operating characteristic (FROC) curve analysis. A modified computer-aided detection system combining the proposed approach with the SER method is also presented. The proposed method provides improvements in the rates of false positive markings over the SER method in the detection of breast cancer (as assessed by FROC analysis). The modified computer-aided detection system that incorporates both the proposed method and the SER method yields ROC results equal to that produced by SER while simultaneously providing improvements over the SER method in terms of false positives per noncancerous exam. The proposed method for identifying malignancies outperforms the SER method in terms of false positives on a challenging dataset containing many small lesions and may play a useful role in breast cancer screening by MRI as part of a computer-aided detection system.

Computerized detection of noncalcified plaques in coronary CT angiography: evaluation of topological soft gradient prescreening method and luminal analysis.

The buildup of noncalcified plaques (NCPs) that are vulnerable to rupture in coronary arteries is a risk for myocardial infarction. Interpretation of coronary CT angiography (cCTA) to search for NCP is a challenging task for radiologists due to the low CT number of NCP, the large number of coronary arteries, and multiple phase CT acquisition. The authors conducted a preliminary study to develop machine learning method for automated detection of NCPs in cCTA. With IRB approval, a data set of 83 ECG-gated contrast enhanced cCTA scans with 120 NCPs was collected retrospectively from patient files. A multiscale coronary artery response and rolling balloon region growing (MSCAR-RBG) method was applied to each cCTA volume to extract the coronary arterial trees. Each extracted vessel was reformatted to a straightened volume composed of cCTA slices perpendicular to the vessel centerline. A topological soft-gradient (TSG) detection method was developed to prescreen for NCP candidates by analyzing the 2D topological features of the radial gradient field surface along the vessel wall. The NCP candidates were then characterized by a luminal analysis that used 3D geometric features to quantify the shape information and gray-level features to evaluate the density of the NCP candidates. With machine learning techniques, useful features were identified and combined into an NCP score to differentiate true NCPs from false positives (FPs). To evaluate the effectiveness of the image analysis methods, the authors performed tenfold cross-validation with the available data set. Receiver operating characteristic (ROC) analysis was used to assess the classification performance of individual features and the NCP score. The overall detection performance was estimated by free response ROC (FROC) analysis. With our TSG prescreening method, a prescreening sensitivity of 92.5% (111/120) was achieved with a total of 1181 FPs (14.2 FPs/scan). On average, six features were selected during the tenfold cross-validation training. The average area under the ROC curve (AUC) value for training was 0.87 ± 0.01 and the AUC value for validation was 0.85 ± 0.01. Using the NCP score, FROC analysis of the validation set showed that the FP rates were reduced to 3.16, 1.90, and 1.39 FPs/scan at sensitivities of 90%, 80%, and 70%, respectively. The topological soft-gradient prescreening method in combination with the luminal analysis for FP reduction was effective for detection of NCPs in cCTA, including NCPs causing positive or negative vessel remodeling. The accuracy of vessel segmentation, tracking, and centerline identification has a strong impact on NCP detection. Studies are underway to further improve these techniques and reduce the FPs of the CADe system.

Computer-aided detection of clustered microcalcifications in multiscale bilateral filtering regularized reconstructed digital breast tomosynthesis volume.

Develop a computer-aided detection (CADe) system for clustered microcalcifications in digital breast tomosynthesis (DBT) volume enhanced with multiscale bilateral filtering (MSBF) regularization. With Institutional Review Board approval and written informed consent, two-view DBT of 154 breasts, of which 116 had biopsy-proven microcalcification (MC) clusters and 38 were free of MCs, was imaged with a General Electric GEN2 prototype DBT system. The DBT volumes were reconstructed with MSBF-regularized simultaneous algebraic reconstruction technique (SART) that was designed to enhance MCs and reduce background noise while preserving the quality of other tissue structures. The contrast-to-noise ratio (CNR) of MCs was further improved with enhancement-modulated calcification response (EMCR) preprocessing, which combined multiscale Hessian response to enhance MCs by shape and bandpass filtering to remove the low-frequency structured background. MC candidates were then located in the EMCR volume using iterative thresholding and segmented by adaptive region growing. Two sets of potential MC objects, cluster centroid objects and MC seed objects, were generated and the CNR of each object was calculated. The number of candidates in each set was controlled based on the breast volume. Dynamic clustering around the centroid objects grouped the MC candidates to form clusters. Adaptive criteria were designed to reduce false positive (FP) clusters based on the size, CNR values and the number of MCs in the cluster, cluster shape, and cluster based maximum intensity projection. Free-response receiver operating characteristic (FROC) and jackknife alternative FROC (JAFROC) analyses were used to assess the performance and compare with that of a previous study. Unpaired two-tailed t-test showed a significant increase (p < 0.0001) in the ratio of CNRs for MCs with and without MSBF regularization compared to similar ratios for FPs. For view-based detection, a sensitivity of 85% was achieved at an FP rate of 2.16 per DBT volume. For case-based detection, a sensitivity of 85% was achieved at an FP rate of 0.85 per DBT volume. JAFROC analysis showed a significant improvement in the performance of the current CADe system compared to that of our previous system (p = 0.003). MBSF regularized SART reconstruction enhances MCs. The enhancement in the signals, in combination with properly designed adaptive threshold criteria, effective MC feature analysis, and false positive reduction techniques, leads to a significant improvement in the detection of clustered MCs in DBT.

Temporal Subtraction Method for Lung Nodule Detection on Successive Thoracic CT Soft-Copy Images

To assess the effects of a new computed tomographic (CT) temporal subtraction (TS) method on radiologist performance in lung nodule detection on thin-section CT images. The institutional review board approved this study, and the informed consent requirement was waived. Fifty pairs (current and previous CT images) of standard-dose 2-mm thin-section CT images and corresponding CT TS images were used for an observer performance study. Two thoracic radiologists identified 30 nodules ranging in size from 5 to 19 mm, and these nodules served as the reference standard of actionable nodules (noncalcified nodules larger than 4 mm). Eight radiologists (four attending radiologists, four radiology residents) participated in this observer study. Ratings and locations of lesions determined by observers were used to assess the significance of differences between radiologists' performances without and with the CT TS images in jacknife free-response receiver operating characteristics analysis. Average figure of merit values increased significantly for all radiologists (from 0.838 without CT TS images to 0.894 with CT TS images [P = .033]). Average sensitivity for detection of actionable nodules was improved from 73.4% to 83.4%, with a false-positive rate of 0.15 per case, by using CT TS images. The reading time with CT TS images was not significantly different from that without. The novel CT TS method would increase observer performance for lung nodule detection without considerably extending the reading time.