Clinical MRI morphological analysis of functional seizures compared to seizure-naïve and psychiatric controls.

PURPOSE: Functional seizures (FS), also known as psychogenic nonepileptic seizures (PNES), are physical manifestations of acute or chronic psychological distress. Functional and structural neuroimaging have identified objective signs of this disorder. We evaluated whether magnetic resonance imaging (MRI) morphometry differed between patients with FS and clinically relevant comparison populations. METHODS: Quality-screened clinical-grade MRIs were acquired from 666 patients from 2006 to 2020. Morphometric features were quantified with FreeSurfer v6. Mixed-effects linear regression compared the volume, thickness, and surface area within 201 regions-of-interest for 90 patients with FS, compared to seizure-naïve patients with depression (n = 243), anxiety (n = 68), and obsessive-compulsive disorder (OCD, n = 41), respectively, and to other seizure-naïve controls with similar quality MRIs, accounting for the influence of multiple confounds including depression and anxiety based on chart review. These comparison populations were obtained through review of clinical records plus research studies obtained on similar scanners. RESULTS: After Bonferroni-Holm correction, patients with FS compared with seizure-naïve controls exhibited thinner bilateral superior temporal cortex (left 0.053 mm, p = 0.014; right 0.071 mm, p = 0.00006), thicker left lateral occipital cortex (0.052 mm, p = 0.0035), and greater left cerebellar white-matter volume (1085 mm3, p = 0.0065). These findings were not accounted for by lower MRI quality in patients with FS. CONCLUSIONS: These results reinforce prior indications of structural neuroimaging correlates of FS and, in particular, distinguish brain morphology in FS from that in depression, anxiety, and OCD. Future work may entail comparisons with other psychiatric disorders including bipolar and schizophrenia, as well as exploration of brain structural heterogeneity within FS.

Deep Learning in Medical Image Analysis.

Deep learning is the state-of-the-art machine learning approach. The success of deep learning in many pattern recognition applications has brought excitement and high expectations that deep learning, or artificial intelligence (AI), can bring revolutionary changes in health care. Early studies of deep learning applied to lesion detection or classification have reported superior performance compared to those by conventional techniques or even better than radiologists in some tasks. The potential of applying deep-learning-based medical image analysis to computer-aided diagnosis (CAD), thus providing decision support to clinicians and improving the accuracy and efficiency of various diagnostic and treatment processes, has spurred new research and development efforts in CAD. Despite the optimism in this new era of machine learning, the development and implementation of CAD or AI tools in clinical practice face many challenges. In this chapter, we will discuss some of these issues and efforts needed to develop robust deep-learning-based CAD tools and integrate these tools into the clinical workflow, thereby advancing towards the goal of providing reliable intelligent aids for patient care.

Variabilities in Reference Standard by Radiologists and Performance Assessment in Detection of Pulmonary Embolism in CT Pulmonary Angiography.

Annotating lesion locations by radiologists' manual marking is a key step to provide reference standard for the training and testing of a computer-aided detection system by supervised machine learning. Inter-reader variability is not uncommon in readings even by expert radiologists. This study evaluated the variability of the radiologist-identified pulmonary emboli (PEs) to demonstrate the importance of improving the reliability of the reference standard by a multi-step process for performance evaluation. In an initial reading of 40 CTPA PE cases, two experienced thoracic radiologists independently marked the PE locations. For markings from the two radiologists that did not agree, each radiologist re-read the cases independently to assess the discordant markings. Finally, for markings that still disagreed after the second reading, the two radiologists read together to reach a consensus. The variability of radiologists was evaluated by analyzing the agreement between two radiologists. For the 40 cases, 475 and 514 PEs were identified by radiologists R1 and R2 in the initial independent readings, respectively. For a total of 545 marks by the two radiologists, 81.5% (444/545) of the marks agreed but 101 marks in 36 cases differed. After consensus, 65 (64.4%) and 36 (35.6%) of the 101 marks were determined to be true PEs and false positives (FPs), respectively. Of these, 48 and 17 were false negatives (FNs) and 14 and 22 were FPs by R1 and R2, respectively. Our study demonstrated that there is substantial variability in reference standards provided by radiologists, which impacts the performance assessment of a lesion detection system. Combination of multiple radiologists' readings and consensus is needed to improve the reliability of a reference standard.

Use of Screening Mammography to Detect Occult Malignancy in Autologous Breast Reconstructions: A 15-year Experience.

Purpose To examine how often screening mammography depicts clinically occult malignancy in breast reconstruction with autologous myocutaneous flaps (AMFs). Materials and Methods Between January 1, 2000, and July 15, 2015, the authors retrospectively identified 515 women who had undergone mammography of 618 AMFs and who had at least 1 year of clinical follow-up. Of the 618 AMFs, 485 (78.5%) were performed after mastectomy for cancer and 133 (21.5%) were performed after prophylactic mastectomy. Medical records were used to determine the frequency, histopathologic characteristics, presentation, time to recurrence, and detection modality of malignancy. Cancer detection rate (CDR), sensitivity, specificity, positive predictive value, and false-positive biopsy rate were calculated. Results An average of 6.7 screening mammograms (range, 1-16) were obtained over 15.5 years. The frequency of local-regional recurrence (LRR) was 3.9% (20 of 515 women; 95% confidence interval [CI]: 2.2%, 5.6%); all LRRs were invasive, and none were detected in the breast mound after prophylactic mastectomy. Of the 20 women with LRR, 13 (65%) were screened annually before the diagnosis. Seven of those 13 women (54%) had clinically occult LRR, and mammography depicted five. Five of the six clinically evident recurrences (83%) were interval cancers. The median time between reconstruction and first recurrence was 4.4 years (range, 0.8-16.2 years). The CDR per AMF was 1.5 per 1000 screening mammograms (five of 3358; 95% CI: 0.18, 2.8) after mastectomy for cancer and 0 of 1000 examinations (0 of 805 mammograms; 95% CI: 0, 5) after prophylactic mastectomy. Sensitivity, specificity, positive predictive value, and false-positive biopsy rate were 42% (five of 12), 99.4% (4125 of 4151), 16% (five of 31), and 0.6% (26 of 4151), respectively. Conclusion The CDR of screening mammography (1.5 per 1000 screening mammograms) of the AMF after mastectomy for cancer is comparable to that for one native breast of an age-matched woman. Screening mammography adds little value after prophylactic mastectomy. © RSNA, 2018.

Quantitative Analysis of MR Imaging to Assess Treatment Response for Patients with Multiple Myeloma by Using Dynamic Intensity Entropy Transformation: A Preliminary Study.

PURPOSE: To develop a quantitative measure of bone marrow changes in magnetic resonance (MR) images and investigate its capability for assessment of treatment response for patients with multiple myeloma (MM). MATERIALS AND METHODS: This study was retrospective, institutional review board approved, and HIPAA compliant. Informed consent was waived. Patients (n = 64; mean age, 58.8 years [age range, 27-75 years]) who were diagnosed with MM and underwent autologous bone marrow stem cell transplantation (BMT) were evaluated. A pair of spinal MR examinations performed before and after BMT was collected from each patient's records. A three-dimensional dynamic intensity entropy transformation (DIET) method was developed to transform MR T1-weighted signal voxel by voxel to a quantitative entropy enhancement value (qEEV), from which predictor variables were derived to train a linear discriminant analysis classifier by using a leave-one-out method. The output of the linear discriminant analysis provided a qEEV-based response index for quantitative assessment of treatment response. The performance of quantitative response index for the discrimination of responder and nonresponder patients was evaluated by receiver operating characteristic curve analysis. RESULTS: Among the 46 and 18 clinically diagnosed responder and nonresponder patients, the quantitative response index at a chosen decision threshold correctly identified 42 responder and 17 nonresponder patients. The agreement between the DIET method and the clinical outcome reached 0.922 (59 of 64; κ = 0.816; area under the receiver operating characteristic curve, 0.886 ± 0.042). CONCLUSION: This study demonstrated the feasibility of quantitative response index to differentiate responder and nonresponder patients and had substantial agreement with clinical outcomes, which indicated that this quantitative measure has the potential to be an image biomarker to assess MM treatment response.

Detection of Severe Lung Infection on Chest Radiographs of COVID-19 Patients: Robustness of AI Models across Multi-Institutional Data.

The diagnosis of severe COVID-19 lung infection is important because it carries a higher risk for the patient and requires prompt treatment with oxygen therapy and hospitalization while those with less severe lung infection often stay on observation. Also, severe infections are more likely to have long-standing residual changes in their lungs and may need follow-up imaging. We have developed deep learning neural network models for classifying severe vs. non-severe lung infections in COVID-19 patients on chest radiographs (CXR). A deep learning U-Net model was developed to segment the lungs. Inception-v1 and Inception-v4 models were trained for the classification of severe vs. non-severe COVID-19 infection. Four CXR datasets from multi-country and multi-institutional sources were used to develop and evaluate the models. The combined dataset consisted of 5748 cases and 6193 CXR images with physicians' severity ratings as reference standard. The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance. We studied the reproducibility of classification performance using the different combinations of training and validation data sets. We also evaluated the generalizability of the trained deep learning models using both independent internal and external test sets. The Inception-v1 based models achieved AUC ranging between 0.81 ± 0.02 and 0.84 ± 0.0, while the Inception-v4 models achieved AUC in the range of 0.85 ± 0.06 and 0.89 ± 0.01, on the independent test sets, respectively. These results demonstrate the promise of using deep learning models in differentiating COVID-19 patients with severe from non-severe lung infection on chest radiographs.

Enhancing Early Lung Cancer Diagnosis: Predicting Lung Nodule Progression in Follow-Up Low-Dose CT Scan with Deep Generative Model.

Early diagnosis of lung cancer can significantly improve patient outcomes. We developed a Growth Predictive model based on the Wasserstein Generative Adversarial Network framework (GP-WGAN) to predict the nodule growth patterns in the follow-up LDCT scans. The GP-WGAN was trained with a training set (N = 776) containing 1121 pairs of nodule images with about 1-year intervals and deployed to an independent test set of 450 nodules on baseline LDCT scans to predict nodule images (GP-nodules) in their 1-year follow-up scans. The 450 GP-nodules were finally classified as malignant or benign by a lung cancer risk prediction (LCRP) model, achieving a test AUC of 0.827 ± 0.028, which was comparable to the AUC of 0.862 ± 0.028 achieved by the same LCRP model classifying real follow-up nodule images (p = 0.071). The net reclassification index yielded consistent outcomes (NRI = 0.04; p = 0.62). Other baseline methods, including Lung-RADS and the Brock model, achieved significantly lower performance (p < 0.05). The results demonstrated that the GP-nodules predicted by our GP-WGAN model achieved comparable performance with the nodules in the real follow-up scans for lung cancer diagnosis, indicating the potential to detect lung cancer earlier when coupled with accelerated clinical management versus the current approach of waiting until the next screening exam.

Leveraging Serial Low-Dose CT Scans in Radiomics-based Reinforcement Learning to Improve Early Diagnosis of Lung Cancer at Baseline Screening.

Purpose To evaluate the feasibility of leveraging serial low-dose CT (LDCT) scans to develop a radiomics-based reinforcement learning (RRL) model for improving early diagnosis of lung cancer at baseline screening. Materials and Methods In this retrospective study, 1951 participants (female patients, 822; median age, 61 years [range, 55-74 years]) (male patients, 1129; median age, 62 years [range, 55-74 years]) were randomly selected from the National Lung Screening Trial between August 2002 and April 2004. An RRL model using serial LDCT scans (S-RRL) was trained and validated using data from 1404 participants (372 with lung cancer) containing 2525 available serial LDCT scans up to 3 years. A baseline RRL (B-RRL) model was trained with only LDCT scans acquired at baseline screening for comparison. The 547 held-out individuals (150 with lung cancer) were used as an independent test set for performance evaluation. The area under the receiver operating characteristic curve (AUC) and the net reclassification index (NRI) were used to assess the performances of the models in the classification of screen-detected nodules. Results Deployment to the held-out baseline scans showed that the S-RRL model achieved a significantly higher test AUC (0.88 [95% CI: 0.85, 0.91]) than both the Brock model (AUC, 0.84 [95% CI: 0.81, 0.88]; P = .02) and the B-RRL model (AUC, 0.86 [95% CI: 0.83, 0.90]; P = .02). Lung cancer risk stratification was significantly improved by the S-RRL model as compared with Lung CT Screening Reporting and Data System (NRI, 0.29; P < .001) and the Brock model (NRI, 0.12; P = .008). Conclusion The S-RRL model demonstrated the potential to improve early diagnosis and risk stratification for lung cancer at baseline screening as compared with the B-RRL model and clinical models. Keywords: Radiomics-based Reinforcement Learning, Lung Cancer Screening, Low-Dose CT, Machine Learning © RSNA, 2024 Supplemental material is available for this article.

Breast mass characterization using 3-dimensional automated ultrasound as an adjunct to digital breast tomosynthesis: a pilot study.

OBJECTIVES: The purpose of this study was to retrospectively evaluate the effect of 3-dimensional automated ultrasound (3D-AUS) as an adjunct to digital breast tomosynthesis (DBT) on radiologists' performance and confidence in discriminating malignant and benign breast masses. METHODS: Two-view DBT (craniocaudal and mediolateral oblique or lateral) and single-view 3D-AUS images were acquired from 51 patients with subsequently biopsy-proven masses (13 malignant and 38 benign). Six experienced radiologists rated, on a 13-point scale, the likelihood of malignancy of an identified mass, first by reading the DBT images alone, followed immediately by reading the DBT images with automatically coregistered 3D-AUS images. The diagnostic performance of each method was measured using receiver operating characteristic (ROC) curve analysis and changes in sensitivity and specificity with the McNemar test. After each reading, radiologists took a survey to rate their confidence level in using DBT alone versus combined DBT/3D-AUS as potential screening modalities. RESULTS: The 6 radiologists had an average area under the ROC curve of 0.92 for both modalities (range, 0.89-0.97 for DBT and 0.90-0.94 for DBT/3D-AUS). With a Breast Imaging Reporting and Data System rating of 4 as the threshold for biopsy recommendation, the average sensitivity of the radiologists increased from 96% to 100% (P > .08) with 3D-AUS, whereas the specificity decreased from 33% to 25% (P > .28). Survey responses indicated increased confidence in potentially using DBT for screening when 3D-AUS was added (P < .05 for each reader). CONCLUSIONS: In this initial reader study, no significant difference in ROC performance was found with the addition of 3D-AUS to DBT. However, a trend to improved discrimination of malignancy was observed when adding 3D-AUS. Radiologists' confidence also improved with DBT/3DAUS compared to DBT alone.

Improved Repeatability of Mouse Tibia Volume Segmentation in Murine Myelofibrosis Model Using Deep Learning.

A murine model of myelofibrosis in tibia was used in a co-clinical trial to evaluate segmentation methods for application of image-based biomarkers to assess disease status. The dataset (32 mice with 157 3D MRI scans including 49 test-retest pairs scanned on consecutive days) was split into approximately 70% training, 10% validation, and 20% test subsets. Two expert annotators (EA1 and EA2) performed manual segmentations of the mouse tibia (EA1: all data; EA2: test and validation). Attention U-net (A-U-net) model performance was assessed for accuracy with respect to EA1 reference using the average Jaccard index (AJI), volume intersection ratio (AVI), volume error (AVE), and Hausdorff distance (AHD) for four training scenarios: full training, two half-splits, and a single-mouse subsets. The repeatability of computer versus expert segmentations for tibia volume of test-retest pairs was assessed by within-subject coefficient of variance (%wCV). A-U-net models trained on full and half-split training sets achieved similar average accuracy (with respect to EA1 annotations) for test set: AJI = 83-84%, AVI = 89-90%, AVE = 2-3%, and AHD = 0.5 mm-0.7 mm, exceeding EA2 accuracy: AJ = 81%, AVI = 83%, AVE = 14%, and AHD = 0.3 mm. The A-U-net model repeatability wCV [95% CI]: 3 [2, 5]% was notably better than that of expert annotators EA1: 5 [4, 9]% and EA2: 8 [6, 13]%. The developed deep learning model effectively automates murine bone marrow segmentation with accuracy comparable to human annotators and substantially improved repeatability.

Computer-aided detection of clustered microcalcifications in digital breast tomosynthesis: a 3D approach.

PURPOSE: To design a computer-aided detection (CADe) system for clustered microcalcifications in reconstructed digital breast tomosynthesis (DBT) volumes and to perform a preliminary evaluation of the CADe system. METHODS: IRB approval and informed consent were obtained in this study. A data set of two-view DBT of 72 breasts containing microcalcification clusters was collected from 72 subjects who were scheduled to undergo breast biopsy. Based on tissue sampling results, 17 cases had breast cancer and 55 were benign. A separate data set of two-view DBT of 38 breasts free of clustered microcalcifications from 38 subjects was collected to independently estimate the number of false-positives (FPs) generated by the CADe system. A radiologist experienced in breast imaging marked the biopsied cluster of microcalcifications with a 3D bounding box using all available clinical and imaging information. A CADe system was designed to detect microcalcification clusters in the reconstructed volume. The system consisted of prescreening, clustering, and false-positive reduction stages. In the prescreening stage, the conspicuity of microcalcification-like objects was increased by an enhancement-modulated 3D calcification response function. An iterative thresholding and 3D object growing method was used to detect cluster seed objects, which were used as potential centers of microcalcification clusters. In the cluster detection stage, microcalcification candidates were identified using a second iterative thresholding procedure, which was applied to the signal-to-noise ratio (SNR) enhanced image voxels with a positive calcification response. Starting with each cluster seed object as the initial cluster center, a dynamic clustering algorithm formed a cluster candidate by including microcalcification candidates within a 3D neighborhood of the cluster seed object that satisfied the clustering criteria. The number, size, and SNR of the microcalcifications in a cluster candidate and the cluster shape were used to reduce the number of FPs. RESULTS: The prescreening stage detected a cluster seed object in 94% of the biopsied microcalcification clusters at a threshold of 100 cluster seed objects per DBT volume. After clustering, the detection sensitivity was 90% at 15 marks per DBT volume. After FP reduction, at 85% sensitivity, the average number of FPs estimated using the data set containing microcalcification clusters was 3.8 per DBT volume, and that estimated using the data set free of microcalcification clusters was 3.4. The detection performance for malignant microcalcification clusters was superior to that for benign clusters. CONCLUSIONS: Our study indicates the feasibility of the 3D approach to the detection of clustered microcalcifications in DBT and that the newly designed enhancement-modulated 3D calcification response function is promising for prescreening. Further work is needed to assess the generalizability of our approach and to improve its performance.

Recursive Training Strategy for a Deep Learning Network for Segmentation of Pathology Nuclei With Incomplete Annotation.

This study developed a recursive training strategy to train a deep learning model for nuclei detection and segmentation using incomplete annotation. A dataset of 141 H&E stained breast cancer pathologic images with incomplete annotation was randomly split into training/validation set and test set of 89 and 52 images, respectively. The positive training samples were extracted at each annotated cell and augmented with affine translation. The negative training samples were selected from the non-cellular regions free of nuclei using a histogram-based semi-automatic method. A U-Net model was initially trained by minimizing a custom loss function. After the first stage of training, the trained U-Net model was applied to the images in the training set in an inference mode. The U-Net segmented objects with high quality were selected by a semi-automated method. Combining the newly selected high quality objects with the annotated nuclei and the previously generated negative samples, the U-Net model was retrained recursively until the stopping criteria were satisfied. For the 52 test images, the U-Net trained with and without using our recursive training method achieved a sensitivity of 90.3% and 85.3% for nuclei detection, respectively. For nuclei segmentation, the average Dice coefficient and average Jaccard index were 0.831±0.213 and 0.750±0.217, 0.780±0.270 and 0.697±0.264, for U-Net with and without recursive training, respectively. The improvement achieved by our proposed method was statistically significant (P < 0.05). In conclusion, our recursive training method effectively enlarged the set of annotated objects for training the deep learning model and further improved the detection and segmentation performance.

Hybrid U-Net-based deep learning model for volume segmentation of lung nodules in CT images.

OBJECTIVE: Accurate segmentation of the lung nodule in computed tomography images is a critical component of a computer-assisted lung cancer detection/diagnosis system. However, lung nodule segmentation is a challenging task due to the heterogeneity of nodules. This study is to develop a hybrid deep learning (H-DL) model for the segmentation of lung nodules with a wide variety of sizes, shapes, margins, and opacities. MATERIALS AND METHODS: A dataset collected from Lung Image Database Consortium image collection containing 847 cases with lung nodules manually annotated by at least two radiologists with nodule diameters greater than 7 mm and less than 45 mm was randomly split into 683 training/validation and 164 independent test cases. The 50% consensus consolidation of radiologists' annotation was used as the reference standard for each nodule. We designed a new H-DL model combining two deep convolutional neural networks (DCNNs) with different structures as encoders to increase the learning capabilities for the segmentation of complex lung nodules. Leveraging the basic symmetric U-shaped architecture of U-Net, we redesigned two new U-shaped deep learning (U-DL) models that were expanded to six levels of convolutional layers. One U-DL model used a shallow DCNN structure containing 16 convolutional layers adapted from the VGG-19 as the encoder, and the other used a deep DCNN structure containing 200 layers adapted from DenseNet-201 as the encoder, while the same decoder with only one convolutional layer at each level was used in both U-DL models, and we referred to them as the shallow and deep U-DL models. Finally, an ensemble layer was used to combine the two U-DL models into the H-DL model. We compared the effectiveness of the H-DL, the shallow U-DL and the deep U-DL models by deploying them separately to the test set. The accuracy of volume segmentation for each nodule was evaluated by the 3D Dice coefficient and Jaccard index (JI) relative to the reference standard. For comparison, we calculated the median and minimum of the 3D Dice and JI over the individual radiologists who segmented each nodule, referred to as M-Dice, min-Dice, M-JI, and min-JI. RESULTS: For the 164 test cases with 327 nodules, our H-DL model achieved an average 3D Dice coefficient of 0.750 ± 0.135 and an average JI of 0.617 ± 0.159. The radiologists' average M-Dice was 0.778 ± 0.102, and the average M-JI was 0.651 ± 0.127; both were significantly higher than those achieved by the H-DL model (p < 0.05). The radiologists' average min-Dice (0.685 ± 0.139) and the average min-JI (0.537 ± 0.153) were significantly lower than those achieved by the H-DL model (p < 0.05). The results indicated that the H-DL model approached the average performance of radiologists and was superior to the radiologist whose manual segmentation had the min-Dice and min-JI. Moreover, the average Dice and average JI achieved by the H-DL model were significantly higher than those achieved by the individual shallow U-DL model (Dice of 0.745 ± 0.139, JI of 0.611 ± 0.161; p < 0.05) or the individual deep U-DL model alone (Dice of 0.739 ± 0.145, JI of 0.604 ± 0.163; p < 0.05). CONCLUSION: Our newly developed H-DL model outperformed the individual shallow or deep U-DL models. The H-DL method combining multilevel features learned by both the shallow and deep DCNNs could achieve segmentation accuracy comparable to radiologists' segmentation for nodules with wide ranges of image characteristics.

Multi-class deep learning segmentation and automated measurements in periodontal sonograms of a porcine model.

OBJECTIVES: Ultrasound emerges as a complement to cone-beam computed tomography in dentistry, but struggles with artifacts like reverberation and shadowing. This study seeks to help novice users recognize soft tissue, bone, and crown of a dental sonogram, and automate soft tissue height (STH) measurement using deep learning. METHODS: In this retrospective study, 627 frames from 111 independent cine loops of mandibular and maxillary premolar and incisors collected from our porcine model (N = 8) were labeled by a reader. 274 premolar sonograms, including data augmentation, were used to train a multi class segmentation model. The model was evaluated against several test sets, including premolar of the same breed (n = 74, Yucatan) and premolar of a different breed (n = 120, Sinclair). We further proposed a rule-based algorithm to automate STH measurements using predicted segmentation masks. RESULTS: The model reached a Dice similarity coefficient of 90.7±4.39%, 89.4±4.63%, and 83.7±10.5% for soft tissue, bone, and crown segmentation, respectively on the first test set (n = 74), and 90.0±7.16%, 78.6±13.2%, and 62.6±17.7% on the second test set (n = 120). The automated STH measurements have a mean difference (95% confidence interval) of -0.22 mm (-1.4, 0.95), a limit of agreement of 1.2 mm, and a minimum ICC of 0.915 (0.857, 0.948) when compared to expert annotation. CONCLUSION: This work demonstrates the potential use of deep learning in identifying periodontal structures on sonograms and obtaining diagnostic periodontal dimensions.

Effect of Dose Level on Radiologists' Detection of Microcalcifications in Digital Breast Tomosynthesis: An Observer Study with Breast Phantoms.

OBJECTIVES: To compare radiologists' sensitivity, confidence level, and reading efficiency of detecting microcalcifications in digital breast tomosynthesis (DBT) at two clinically relevant dose levels. MATERIALS AND METHODS: Six 5-cm-thick heterogeneous breast phantoms embedded with a total of 144 simulated microcalcification clusters of four speck sizes were imaged at two dose modes by a clinical DBT system. The DBT volumes at the two dose levels were read independently by six MQSA radiologists and one fellow with 1-33 years (median 12 years) of experience in a fully-crossed counter-balanced manner. The radiologist located each potential cluster and rated its conspicuity and his/her confidence that the marked location contained a cluster. The differences in the results between the two dose modes were analyzed by two-tailed paired t-test. RESULTS: Compared to the lower-dose mode, the average glandular dose in the higher-dose mode for the 5-cm phantoms increased from 1.34 to 2.07 mGy. The detection sensitivity increased for all speck sizes and significantly for the two smaller sizes (p <0.05). An average of 13.8% fewer false positive clusters was marked. The average conspicuity rating and the radiologists' confidence level were higher for all speck sizes and reached significance (p <0.05) for the three larger sizes. The average reading time per detected cluster reduced significantly (p <0.05) by an average of 13.2%. CONCLUSION: For a 5-cm-thick breast, an increase in average glandular dose from 1.34 to 2.07 mGy for DBT imaging increased the conspicuity of microcalcifications, improved the detection sensitivity by radiologists, increased their confidence levels, reduced false positive detections, and increased the reading efficiency.

Association of computerized mammographic parenchymal pattern measure with breast cancer risk: a pilot case-control study.

PURPOSE: To develop a computerized mammographic parenchymal pattern (MPP) measure and investigate its association with breast cancer risk. MATERIALS AND METHODS: A pilot case-control study was conducted by collecting mammograms from 382 subjects retrospectively. The study was institutional review board approved and HIPAA compliant. Informed consent was waived. The cases included the contralateral mammograms of cancer patients (n = 136) obtained at least 1 year before diagnosis. The controls included mammograms of healthy subjects (n = 246) who had cancer-free follow-up for at least 5 years. The data set was historically divided into a training set and an independent test set. An MPP measure was designed to analyze the texture patterns of fibroglandular tissue in the retroareolar region. Odds ratios (ORs) were used to assess the association between breast cancer risk and MPP. To test the trend in ORs, we divided the MPP measure into three categories (C1, C2, and C3) on the basis of its values from low to high, with C1 as the baseline. The confounding factors in this study included patient age, body mass index, first-degree relatives with history of breast cancer, number of previous breast biopsies, and percentage density (PD). RESULTS: Among all of the subjects from the training and test data sets, the Pearson product-moment correlation coefficient between MPP and PD was 0.13. With logistic regression to adjust the confounding, the adjusted ORs for C2 and C3 relative to C1 in the test set were 2.82 (P = .041) and 13.89 (P < .001), respectively. CONCLUSION: The proposed MPP measure demonstrated a strong association with breast cancer risk and has the potential to serve as an independent factor for risk prediction.

Computer-aided detection of breast masses: four-view strategy for screening mammography.

PURPOSE: To improve the performance of a computer-aided detection (CAD) system for mass detection by using four-view information in screening mammography. METHODS: The authors developed a four-view CAD system that emulates radiologists' reading by using the craniocaudal and mediolateral oblique views of the ipsilateral breast to reduce false positives (FPs) and the corresponding views of the contralateral breast to detect asymmetry. The CAD system consists of four major components: (1) Initial detection of breast masses on individual views, (2) information fusion of the ipsilateral views of the breast (referred to as two-view analysis), (3) information fusion of the corresponding views of the contralateral breast (referred to as bilateral analysis), and (4) fusion of the four-view information with a decision tree. The authors collected two data sets for training and testing of the CAD system: A mass set containing 389 patients with 389 biopsy-proven masses and a normal set containing 200 normal subjects. All cases had four-view mammograms. The true locations of the masses on the mammograms were identified by an experienced MQSA radiologist. The authors randomly divided the mass set into two independent sets for cross validation training and testing. The overall test performance was assessed by averaging the free response receiver operating characteristic (FROC) curves of the two test subsets. The FP rates during the FROC analysis were estimated by using the normal set only. The jackknife free-response ROC (JAFROC) method was used to estimate the statistical significance of the difference between the test FROC curves obtained with the single-view and the four-view CAD systems. RESULTS: Using the single-view CAD system, the breast-based test sensitivities were 58% and 77% at the FP rates of 0.5 and 1.0 per image, respectively. With the four-view CAD system, the breast-based test sensitivities were improved to 76% and 87% at the corresponding FP rates, respectively. The improvement was found to be statistically significant (p < 0.0001) by JAFROC analysis. CONCLUSIONS: The four-view information fusion approach that emulates radiologists' reading strategy significantly improves the performance of breast mass detection of the CAD system in comparison with the single-view approach.

A minority of patients with functional seizures have abnormalities on neuroimaging.

OBJECTIVE: Functional seizures often are managed incorrectly as a diagnosis of exclusion. However, a significant minority of patients with functional seizures may have abnormalities on neuroimaging that typically are associated with epilepsy, leading to diagnostic confusion. We evaluated the rate of epilepsy-associated findings on MRI, FDG-PET, and CT in patients with functional seizures. METHODS: We studied radiologists' reports from neuroimages at our comprehensive epilepsy center from a consecutive series of patients diagnosed with functional seizures without comorbid epilepsy from 2006 to 2019. We summarized the MRI, FDG-PET, and CT results as follows: within normal limits, incidental findings, unrelated findings, non-specific abnormalities, post-operative study, epilepsy risk factors (ERF), borderline epilepsy-associated findings (EAF), and definitive EAF. RESULTS: Of the 256 MRIs, 23% demonstrated ERF (5%), borderline EAF (8%), or definitive EAF (10%). The most common EAF was hippocampal sclerosis, with the majority of borderline EAF comprising hippocampal atrophy without T2 hyperintensity or vice versa. Of the 87 FDG-PETs, 26% demonstrated borderline EAF (17%) or definitive EAF (8%). Epilepsy-associated findings primarily included focal hypometabolism, especially of the temporal lobes, with borderline findings including subtle or questionable hypometabolism. Of the 51 CTs, only 2% had definitive EAF. SIGNIFICANCE: This large case series provides further evidence that, while uncommon, EAF are seen in patients with functional seizures. A significant portion of these abnormal findings are borderline. The moderately high rate of these abnormalities may represent framing bias from the indication of the study being "seizures," the relative subtlety of EAF, or effects of antiseizure medications.

Selective-diffusion regularization for enhancement of microcalcifications in digital breast tomosynthesis reconstruction.

PURPOSE: Digital breast tomosynthesis (DBT) has been shown to improve mass detection. Detection of microcalcifications is more challenging because of the large breast volume to be searched for subtle signals. The simultaneous algebraic reconstruction technique (SART) was found to provide good image quality for DBT, but the image noise is amplified with an increasing number of iterations. In this study, the authors developed a selective-diffusion (SD) method for noise regularization with SART to improve the contrast-to-noise ratio (CNR) of microcalcifications in the DBT slices for human or machine detection. METHODS: The SD method regularizes SART reconstruction during updating with each projection view. Potential microcalcifications are differentiated from the noisy background by estimating the local gradient information. Different degrees of regularization are applied to the signal or noise classes, such that the microcalcifications will be enhanced while the noise is suppressed. The new SD method was compared to several current methods, including the quadratic Laplacian (QL) method, the total variation (TV) method, and the nonconvex total p-variation (TpV) method for noise regularization with SART. A GE GEN2 prototype DBT system with a stationary digital detector was used for the acquisition of DBT scans at 21 angles in 3 degrees increments over a +/-30 degrees range. The reconstruction image quality without regularization and that with the different regularization methods were compared using the DBT scans of an American College of Radiology phantom and a human subject. The CNR and the full width at half maximum (FWHM) of the line profiles of microcalcifications within the in-focus DBT slices were used as image quality measures. RESULTS: For the comparison of large microcalcifications in the DBT data of the subject, the SD method resulted in comparable CNR to the nonconvex TpV method. Both of them performed better than the other two methods. For subtle microcalcifications, the SD method was superior to other methods in terms of CNR. In both the subject and phantom DBT data, for large microcalcifications, the FWHM of the SD method was comparable to that without regularization, which was wider than that of the TV type methods. For subtle microcalcifications, the SD method had comparable FWHM values to the TV type methods. All three regularization methods were superior to the QL method in terms of FWHM. CONCLUSIONS: The SART regularized by the selective-diffusion method enhanced the CNR and preserved the sharpness of microcalcifications. In comparison with three existing regularization methods, the selective-diffusion regularization was superior to the other methods for subtle microcalcifications.

Effect of finite sample size on feature selection and classification: a simulation study.

PURPOSE: The small number of samples available for training and testing is often the limiting factor in finding the most effective features and designing an optimal computer-aided diagnosis (CAD) system. Training on a limited set of samples introduces bias and variance in the performance of a CAD system relative to that trained with an infinite sample size. In this work, the authors conducted a simulation study to evaluate the performances of various combinations of classifiers and feature selection techniques and their dependence on the class distribution, dimensionality, and the training sample size. The understanding of these relationships will facilitate development of effective CAD systems under the constraint of limited available samples. METHODS: Three feature selection techniques, the stepwise feature selection (SFS), sequential floating forward search (SFFS), and principal component analysis (PCA), and two commonly used classifiers, Fisher's linear discriminant analysis (LDA) and support vector machine (SVM), were investigated. Samples were drawn from multidimensional feature spaces of multivariate Gaussian distributions with equal or unequal covariance matrices and unequal means, and with equal covariance matrices and unequal means estimated from a clinical data set. Classifier performance was quantified by the area under the receiver operating characteristic curve Az. The mean Az values obtained by resubstitution and hold-out methods were evaluated for training sample sizes ranging from 15 to 100 per class. The number of simulated features available for selection was chosen to be 50, 100, and 200. RESULTS: It was found that the relative performance of the different combinations of classifier and feature selection method depends on the feature space distributions, the dimensionality, and the available training sample sizes. The LDA and SVM with radial kernel performed similarly for most of the conditions evaluated in this study, although the SVM classifier showed a slightly higher hold-out performance than LDA for some conditions and vice versa for other conditions. PCA was comparable to or better than SFS and SFFS for LDA at small samples sizes, but inferior for SVM with polynomial kernel. For the class distributions simulated from clinical data, PCA did not show advantages over the other two feature selection methods. Under this condition, the SVM with radial kernel performed better than the LDA when few training samples were available, while LDA performed better when a large number of training samples were available. CONCLUSIONS: None of the investigated feature selection-classifier combinations provided consistently superior performance under the studied conditions for different sample sizes and feature space distributions. In general, the SFFS method was comparable to the SFS method while PCA may have an advantage for Gaussian feature spaces with unequal covariance matrices. The performance of the SVM with radial kernel was better than, or comparable to, that of the SVM with polynomial kernel under most conditions studied.