MRI-based Deep Learning Assessment of Amyloid, Tau, and Neurodegeneration Biomarker Status across the Alzheimer Disease Spectrum.

Background PET can be used for amyloid-tau-neurodegeneration (ATN) classification in Alzheimer disease, but incurs considerable cost and exposure to ionizing radiation. MRI currently has limited use in characterizing ATN status. Deep learning techniques can detect complex patterns in MRI data and have potential for noninvasive characterization of ATN status. Purpose To use deep learning to predict PET-determined ATN biomarker status using MRI and readily available diagnostic data. Materials and Methods MRI and PET data were retrospectively collected from the Alzheimer's Disease Imaging Initiative. PET scans were paired with MRI scans acquired within 30 days, from August 2005 to September 2020. Pairs were randomly split into subsets as follows: 70% for training, 10% for validation, and 20% for final testing. A bimodal Gaussian mixture model was used to threshold PET scans into positive and negative labels. MRI data were fed into a convolutional neural network to generate imaging features. These features were combined in a logistic regression model with patient demographics, APOE gene status, cognitive scores, hippocampal volumes, and clinical diagnoses to classify each ATN biomarker component as positive or negative. Area under the receiver operating characteristic curve (AUC) analysis was used for model evaluation. Feature importance was derived from model coefficients and gradients. Results There were 2099 amyloid (mean patient age, 75 years ± 10 [SD]; 1110 male), 557 tau (mean patient age, 75 years ± 7; 280 male), and 2768 FDG PET (mean patient age, 75 years ± 7; 1645 male) and MRI pairs. Model AUCs for the test set were as follows: amyloid, 0.79 (95% CI: 0.74, 0.83); tau, 0.73 (95% CI: 0.58, 0.86); and neurodegeneration, 0.86 (95% CI: 0.83, 0.89). Within the networks, high gradients were present in key temporal, parietal, frontal, and occipital cortical regions. Model coefficients for cognitive scores, hippocampal volumes, and APOE status were highest. Conclusion A deep learning algorithm predicted each component of PET-determined ATN status with acceptable to excellent efficacy using MRI and other available diagnostic data. © RSNA, 2023 Supplemental material is available for this article.

Thyroid Nodules on Ultrasound in Children and Young Adults: Comparison of Diagnostic Performance of Radiologists' Impressions, ACR TI-RADS, and a Deep Learning Algorithm.

BACKGROUND. In current clinical practice, thyroid nodules in children are generally evaluated on the basis of radiologists' overall impressions of ultrasound images. OBJECTIVE. The purpose of this article is to compare the diagnostic performance of radiologists' overall impression, the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS), and a deep learning algorithm in differentiating benign and malignant thyroid nodules on ultrasound in children and young adults. METHODS. This retrospective study included 139 patients (median age 17.5 years; 119 female patients, 20 male patients) evaluated from January 1, 2004, to September 18, 2020, who were 21 years old and younger with a thyroid nodule on ultrasound with definitive pathologic results from fine-needle aspiration and/or surgical excision to serve as the reference standard. A single nodule per patient was selected, and one transverse and one longitudinal image each of the nodules were extracted for further evaluation. Three radiologists independently characterized nodules on the basis of their overall impression (benign vs malignant) and ACR TI-RADS. A previously developed deep learning algorithm determined for each nodule a likelihood of malignancy, which was used to derive a risk level. Sensitivities and specificities for malignancy were calculated. Agreement was assessed using Cohen kappa coefficients. RESULTS. For radiologists' overall impression, sensitivity ranged from 32.1% to 75.0% (mean, 58.3%; 95% CI, 49.2-67.3%), and specificity ranged from 63.8% to 93.9% (mean, 79.9%; 95% CI, 73.8-85.7%). For ACR TI-RADS, sensitivity ranged from 82.1% to 87.5% (mean, 85.1%; 95% CI, 77.3-92.1%), and specificity ranged from 47.0% to 54.2% (mean, 50.6%; 95% CI, 41.4-59.8%). The deep learning algorithm had a sensitivity of 87.5% (95% CI, 78.3-95.5%) and specificity of 36.1% (95% CI, 25.6-46.8%). Interobserver agreement among pairwise combinations of readers, expressed as kappa, for overall impression was 0.227-0.472 and for ACR TI-RADS was 0.597-0.643. CONCLUSION. Both ACR TI-RADS and the deep learning algorithm had higher sensitivity albeit lower specificity compared with overall impressions. The deep learning algorithm had similar sensitivity but lower specificity than ACR TI-RADS. Interobserver agreement was higher for ACR TI-RADS than for overall impressions. CLINICAL IMPACT. ACR TI-RADS and the deep learning algorithm may serve as potential alternative strategies for guiding decisions to perform fine-needle aspiration of thyroid nodules in children.

Deep Learning for Breast MRI Style Transfer with Limited Training Data.

In this work we introduce a novel medical image style transfer method, StyleMapper, that can transfer medical scans to an unseen style with access to limited training data. This is made possible by training our model on unlimited possibilities of simulated random medical imaging styles on the training set, making our work more computationally efficient when compared with other style transfer methods. Moreover, our method enables arbitrary style transfer: transferring images to styles unseen in training. This is useful for medical imaging, where images are acquired using different protocols and different scanner models, resulting in a variety of styles that data may need to be transferred between. Our model disentangles image content from style and can modify an image's style by simply replacing the style encoding with one extracted from a single image of the target style, with no additional optimization required. This also allows the model to distinguish between different styles of images, including among those that were unseen in training. We propose a formal description of the proposed model. Experimental results on breast magnetic resonance images indicate the effectiveness of our method for style transfer. Our style transfer method allows for the alignment of medical images taken with different scanners into a single unified style dataset, allowing for the training of other downstream tasks on such a dataset for tasks such as classification, object detection and others.

Improving Image Classification of Knee Radiographs: An Automated Image Labeling Approach.

Large numbers of radiographic images are available in musculoskeletal radiology practices which could be used for training of deep learning models for diagnosis of knee abnormalities. However, those images do not typically contain readily available labels due to limitations of human annotations. The purpose of our study was to develop an automated labeling approach that improves the image classification model to distinguish normal knee images from those with abnormalities or prior arthroplasty. The automated labeler was trained on a small set of labeled data to automatically label a much larger set of unlabeled data, further improving the image classification performance for knee radiographic diagnosis. We used BioBERT and EfficientNet as the feature extraction backbone of the labeler and imaging model, respectively. We developed our approach using 7382 patients and validated it on a separate set of 637 patients. The final image classification model, trained using both manually labeled and pseudo-labeled data, had the higher weighted average AUC (WA-AUC 0.903) value and higher AUC values among all classes (normal AUC 0.894; abnormal AUC 0.896, arthroplasty AUC 0.990) compared to the baseline model (WA-AUC = 0.857; normal AUC 0.842; abnormal AUC 0.848, arthroplasty AUC 0.987), trained using only manually labeled data. Statistical tests show that the improvement is significant on normal (p value < 0.002), abnormal (p value < 0.001), and WA-AUC (p value = 0.001). Our findings demonstrated that the proposed automated labeling approach significantly improves the performance of image classification for radiographic knee diagnosis, allowing for facilitating patient care and curation of large knee datasets.

Prediction of Upstaging in Ductal Carcinoma in Situ Based on Mammographic Radiomic Features.

Background Improving diagnosis of ductal carcinoma in situ (DCIS) before surgery is important in choosing optimal patient management strategies. However, patients may harbor occult invasive disease not detected until definitive surgery. Purpose To assess the performance and clinical utility of mammographic radiomic features in the prediction of occult invasive cancer among women diagnosed with DCIS on the basis of core biopsy findings. Materials and Methods In this Health Insurance Portability and Accountability Act-compliant retrospective study, digital magnification mammographic images were collected from women who underwent breast core-needle biopsy for calcifications that was performed at a single institution between September 2008 and April 2017 and yielded a diagnosis of DCIS. The database query was directed at asymptomatic women with calcifications without a mass, architectural distortion, asymmetric density, or palpable disease. Logistic regression with regularization was used. Differences across training and internal test set by upstaging rate, age, lesion size, and estrogen and progesterone receptor status were assessed by using the Kruskal-Wallis or χ2 test. Results The study consisted of 700 women with DCIS (age range, 40-89 years; mean age, 59 years ± 10 [standard deviation]), including 114 with lesions (16.3%) upstaged to invasive cancer at subsequent surgery. The sample was split randomly into 400 women for the training set and 300 for the testing set (mean ages: training set, 59 years ± 10; test set, 59 years ± 10; P = .85). A total of 109 radiomic and four clinical features were extracted. The best model on the test set by using all radiomic and clinical features helped predict upstaging with an area under the receiver operating characteristic curve of 0.71 (95% CI: 0.62, 0.79). For a fixed high sensitivity (90%), the model yielded a specificity of 22%, a negative predictive value of 92%, and an odds ratio of 2.4 (95% CI: 1.8, 3.2). High specificity (90%) corresponded to a sensitivity of 37%, positive predictive value of 41%, and odds ratio of 5.0 (95% CI: 2.8, 9.0). Conclusion Machine learning models that use radiomic features applied to mammographic calcifications may help predict upstaging of ductal carcinoma in situ, which can refine clinical decision making and treatment planning. © RSNA, 2022.

Artificial Intelligence (AI) Tools for Thyroid Nodules on Ultrasound, From the AJR Special Series on AI Applications.

Artificial intelligence (AI) methods for evaluating thyroid nodules on ultrasound have been widely described in the literature, with reported performance of AI tools matching or in some instances surpassing radiologists' performance. As these data have accumulated, products for classification and risk stratification of thyroid nodules on ultrasound have become commercially available. This article reviews FDA-approved products currently on the market, with a focus on product features, reported performance, and considerations for implementation. The products perform risk stratification primarily using a Thyroid Imaging Reporting and Data System (TIRADS), though may provide additional prediction tools independent of TIRADS. Key issues in implementation include integration with radiologist interpretation, impact on workflow and efficiency, and performance monitoring. AI applications beyond nodule classification, including report construction and incidental findings follow-up, are also described. Anticipated future directions of research and development in AI tools for thyroid nodules are highlighted.

Multi-label annotation of text reports from computed tomography of the chest, abdomen, and pelvis using deep learning.

BACKGROUND: There is progress to be made in building artificially intelligent systems to detect abnormalities that are not only accurate but can handle the true breadth of findings that radiologists encounter in body (chest, abdomen, and pelvis) computed tomography (CT). Currently, the major bottleneck for developing multi-disease classifiers is a lack of manually annotated data. The purpose of this work was to develop high throughput multi-label annotators for body CT reports that can be applied across a variety of abnormalities, organs, and disease states thereby mitigating the need for human annotation. METHODS: We used a dictionary approach to develop rule-based algorithms (RBA) for extraction of disease labels from radiology text reports. We targeted three organ systems (lungs/pleura, liver/gallbladder, kidneys/ureters) with four diseases per system based on their prevalence in our dataset. To expand the algorithms beyond pre-defined keywords, attention-guided recurrent neural networks (RNN) were trained using the RBA-extracted labels to classify reports as being positive for one or more diseases or normal for each organ system. Alternative effects on disease classification performance were evaluated using random initialization or pre-trained embedding as well as different sizes of training datasets. The RBA was tested on a subset of 2158 manually labeled reports and performance was reported as accuracy and F-score. The RNN was tested against a test set of 48,758 reports labeled by RBA and performance was reported as area under the receiver operating characteristic curve (AUC), with 95% CIs calculated using the DeLong method. RESULTS: Manual validation of the RBA confirmed 91-99% accuracy across the 15 different labels. Our models extracted disease labels from 261,229 radiology reports of 112,501 unique subjects. Pre-trained models outperformed random initialization across all diseases. As the training dataset size was reduced, performance was robust except for a few diseases with a relatively small number of cases. Pre-trained classification AUCs reached > 0.95 for all four disease outcomes and normality across all three organ systems. CONCLUSIONS: Our label-extracting pipeline was able to encompass a variety of cases and diseases in body CT reports by generalizing beyond strict rules with exceptional accuracy. The method described can be easily adapted to enable automated labeling of hospital-scale medical data sets for training image-based disease classifiers.

Multivariate machine learning models for prediction of pathologic response to neoadjuvant therapy in breast cancer using MRI features: a study using an independent validation set.

PURPOSE: To determine whether a multivariate machine learning-based model using computer-extracted features of pre-treatment dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict pathologic complete response (pCR) to neoadjuvant therapy (NAT) in breast cancer patients. METHODS: Institutional review board approval was obtained for this retrospective study of 288 breast cancer patients at our institution who received NAT and had a pre-treatment breast MRI. A comprehensive set of 529 radiomic features was extracted from each patient's pre-treatment MRI. The patients were divided into equal groups to form a training set and an independent test set. Two multivariate machine learning models (logistic regression and a support vector machine) based on imaging features were trained to predict pCR in (a) all patients with NAT, (b) patients with neoadjuvant chemotherapy (NACT), and (c) triple-negative or human epidermal growth factor receptor 2-positive (TN/HER2+) patients who had NAT. The multivariate models were tested using the independent test set, and the area under the receiver operating characteristics (ROC) curve (AUC) was calculated. RESULTS: Out of the 288 patients, 64 achieved pCR. The AUC values for predicting pCR in TN/HER+ patients who received NAT were significant (0.707, 95% CI 0.582-0.833, p < 0.002). CONCLUSIONS: The multivariate models based on pre-treatment MRI features were able to predict pCR in TN/HER2+ patients.

Management of Thyroid Nodules Seen on US Images: Deep Learning May Match Performance of Radiologists.

BackgroundManagement of thyroid nodules may be inconsistent between different observers and time consuming for radiologists. An artificial intelligence system that uses deep learning may improve radiology workflow for management of thyroid nodules.PurposeTo develop a deep learning algorithm that uses thyroid US images to decide whether a thyroid nodule should undergo a biopsy and to compare the performance of the algorithm with the performance of radiologists who adhere to American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS).Materials and MethodsIn this retrospective analysis, studies in patients referred for US with subsequent fine-needle aspiration or with surgical histologic analysis used as the standard were evaluated. The study period was from August 2006 to May 2010. A multitask deep convolutional neural network was trained to provide biopsy recommendations for thyroid nodules on the basis of two orthogonal US images as the input. In the training phase, the deep learning algorithm was first evaluated by using 10-fold cross-validation. Internal validation was then performed on an independent set of 99 consecutive nodules. The sensitivity and specificity of the algorithm were compared with a consensus of three ACR TI-RADS committee experts and nine other radiologists, all of whom interpreted thyroid US images in clinical practice.ResultsIncluded were 1377 thyroid nodules in 1230 patients with complete imaging data and conclusive cytologic or histologic diagnoses. For the 99 test nodules, the proposed deep learning algorithm achieved 13 of 15 (87%: 95% confidence interval [CI]: 67%, 100%) sensitivity, the same as expert consensus (P > .99) and higher than five of nine radiologists. The specificity of the deep learning algorithm was 44 of 84 (52%; 95% CI: 42%, 62%), which was similar to expert consensus (43 of 84; 51%; 95% CI: 41%, 62%; P = .91) and higher than seven of nine other radiologists. The mean sensitivity and specificity for the nine radiologists was 83% (95% CI: 64%, 98%) and 48% (95% CI: 37%, 59%), respectively.ConclusionSensitivity and specificity of a deep learning algorithm for thyroid nodule biopsy recommendations was similar to that of expert radiologists who used American College of Radiology Thyroid Imaging and Reporting Data System guidelines.© RSNA, 2019Online supplemental material is available for this article.

Using Artificial Intelligence to Revise ACR TI-RADS Risk Stratification of Thyroid Nodules: Diagnostic Accuracy and Utility.

Background Risk stratification systems for thyroid nodules are often complicated and affected by low specificity. Continual improvement of these systems is necessary to reduce the number of unnecessary thyroid biopsies. Purpose To use artificial intelligence (AI) to optimize the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Materials and Methods A total of 1425 biopsy-proven thyroid nodules from 1264 consecutive patients (1026 women; mean age, 52.9 years [range, 18-93 years]) were evaluated retrospectively. Expert readers assigned points based on five ACR TI-RADS categories (composition, echogenicity, shape, margin, echogenic foci), and a genetic AI algorithm was applied to a training set (1325 nodules). Point and pathologic data were used to create an optimized scoring system (hereafter, AI TI-RADS). Performance of the systems was compared by using a test set of the final 100 nodules with interpretations from the expert reader, eight nonexpert readers, and an expert panel. Initial performance of AI TI-RADS was calculated by using a test for differences between binomial proportions. Additional comparisons across readers were conducted by using bootstrapping; diagnostic performance was assessed by using area under the receiver operating curve. Results AI TI-RADS assigned new point values for eight ACR TI-RADS features. Six features were assigned zero points, which simplified categorization. By using expert reader data, the diagnostic performance of ACR TI-RADS and AI TI-RADS was area under the receiver operating curve of 0.91 and 0.93, respectively. For the same expert, specificity of AI TI-RADS (65%, 55 of 85) was higher (P < .001) than that of ACR TI-RADS (47%, 40 of 85). For the eight nonexpert radiologists, mean specificity for AI TI-RADS (55%) was also higher (P < .001) than that of ACR TI-RADS (48%). An interactive AI TI-RADS calculator can be viewed at http://deckard.duhs.duke.edu/∼ai-ti-rads . Conclusion An artificial intelligence-optimized Thyroid Imaging Reporting and Data System (TI-RADS) validates the American College of Radiology TI-RADS while slightly improving specificity and maintaining sensitivity. Additionally, it simplifies feature assignments, which may improve ease of use. © RSNA, 2019 Online supplemental material is available for this article.

Deep learning in radiology: An overview of the concepts and a survey of the state of the art with focus on MRI.

Deep learning is a branch of artificial intelligence where networks of simple interconnected units are used to extract patterns from data in order to solve complex problems. Deep-learning algorithms have shown groundbreaking performance in a variety of sophisticated tasks, especially those related to images. They have often matched or exceeded human performance. Since the medical field of radiology mainly relies on extracting useful information from images, it is a very natural application area for deep learning, and research in this area has rapidly grown in recent years. In this article, we discuss the general context of radiology and opportunities for application of deep-learning algorithms. We also introduce basic concepts of deep learning, including convolutional neural networks. Then, we present a survey of the research in deep learning applied to radiology. We organize the studies by the types of specific tasks that they attempt to solve and review a broad range of deep-learning algorithms being utilized. Finally, we briefly discuss opportunities and challenges for incorporating deep learning in the radiology practice of the future. Level of Evidence: 3 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;49:939-954.

Machine learning-based prediction of future breast cancer using algorithmically measured background parenchymal enhancement on high-risk screening MRI.

BACKGROUND: Preliminary work has demonstrated that background parenchymal enhancement (BPE) assessed by radiologists is predictive of future breast cancer in women undergoing high-risk screening MRI. Algorithmically assessed measures of BPE offer a more precise and reproducible means of measuring BPE than human readers and thus might improve the predictive performance of future cancer development. PURPOSE: To determine if algorithmically extracted imaging features of BPE on screening breast MRI in high-risk women are associated with subsequent development of cancer. STUDY TYPE: Case-control study. POPULATION: In all, 133 women at high risk for developing breast cancer; 46 of these patients developed breast cancer subsequently over a follow-up period of 2 years. FIELD STRENGTH/SEQUENCE: 5 T or 3.0 T T1 -weighted precontrast fat-saturated and nonfat-saturated sequences and postcontrast nonfat-saturated sequences. ASSESSMENT: Automatic features of BPE were extracted with a computer algorithm. Subjective BPE scores from five breast radiologists (blinded to clinical outcomes) were also available. STATISTICAL TESTS: Leave-one-out crossvalidation for a multivariate logistic regression model developed using the automatic features and receiver operating characteristic (ROC) analysis were performed to calculate the area under the curve (AUC). Comparison of automatic features and subjective features was performed using a generalized regression model and the P-value was obtained. Odds ratios for automatic and subjective features were compared. RESULTS: The multivariate model discriminated patients who developed cancer from the patients who did not, with an AUC of 0.70 (95% confidence interval: 0.60-0.79, P < 0.001). The imaging features remained independently predictive of subsequent development of cancer (P < 0.003) when compared with the subjective BPE assessment of the readers. DATA CONCLUSION: Automatically extracted BPE measurements may potentially be used to further stratify risk in patients undergoing high-risk screening MRI. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 5 J. Magn. Reson. Imaging 2019;50:456-464.

Association of distant recurrence-free survival with algorithmically extracted MRI characteristics in breast cancer.

BACKGROUND: While important in diagnosis of breast cancer, the scientific assessment of the role of imaging in prognosis of outcomes and treatment planning is limited. PURPOSE: To evaluate the potential of using quantitative imaging variables for stratifying risk of distant recurrence in breast cancer patients. STUDY TYPE: Retrospective. POPULATION: In all, 892 female invasive breast cancer patients. SEQUENCE: Dynamic contrast-enhanced MRI with field strength 1.5 T and 3 T. ASSESSMENT: Computer vision algorithms were applied to extract a comprehensive set of 529 imaging features quantifying size, shape, enhancement patterns, and heterogeneity of the tumors and the surrounding tissue. Using a development set with 446 cases, we selected 20 imaging features with high prognostic value. STATISTICAL TESTS: We evaluated the imaging features using an independent test set with 446 cases. The principal statistical measure was a concordance index between individual imaging features and patient distant recurrence-free survival (DRFS). RESULTS: The strongest association with DRFS that persisted after controlling for known prognostic clinical and pathology variables was found for signal enhancement ratio (SER) partial tumor volume (concordance index [C] = 0.768, 95% confidence interval [CI]: 0.679-0.856), tumor major axis length (C = 0.742, 95% CI: 0.650-0.834), kurtosis of the SER map within tumor (C = 0.640, 95% CI: 0.521-0.760), tumor cluster shade (C = 0.313, 95% CI: 0.216-0.410), and washin rate information measure of correlation (C = 0.702, 95% CI: 0.601-0.803). DATA CONCLUSION: Quantitative assessment of breast cancer features seen in a routine breast MRI might be able to be used for assessment of risk of distant recurrence. LEVEL OF EVIDENCE: 4 Technical Efficacy: Stage 6 J. Magn. Reson. Imaging 2019.

How accurate and precise are CT based measurements of iodine concentration? A comparison of the minimum detectable concentration difference among single source and dual source dual energy CT in a phantom study.

OBJECTIVES: To assess the impact of scan- and patient-related factors on the error and the minimum detectable difference in iodine concentration among different generations of single-source (SS) fast kV-switching and dual-source (DS) dual-energy CT (DECT). METHODS: Lesions having eight different iodine concentrations (0.2-4 mgI/mL) were emulated in a 3D-printed phantom of medium and large size. Each combination of concentration and size was scanned in dual-energy mode on four different SS and DS DECTs. Radiation doses were 7 and 10 mGy (medium size) and 10, 13, and 16 mGy (large size). Iodine maps were reconstructed with filtered back projection (FBP) and vendor-specific iterative reconstruction algorithms (IRs). Absolute error of iodine quantification (E) was measured. Multivariate regression models determined the influence of CT scanner, iodine concentration, phantom size, radiation dose, and reconstruction algorithm on E. The minimum detectable difference in iodine concentration (ICmin) under the same imaging conditions (intra-conditional) and among different imaging conditions (inter-conditional) was calculated. RESULTS: The error was significantly lower in current than in previous DECT generations (p < 0.001). For all CT scanner conditions, the error was significantly higher with increasing phantom size and decreasing radiation dose (p < 0.001). Iodine concentration only significantly affected the error for SS DECT (p < 0.001). ICmin depended on patient- and scan-related factors and ranged from 0.4 to 1.5 mgI/mL. CONCLUSIONS: Patient- and scan-related factors have a significant impact on the error and minimum detectable difference in iodine concentration within and among SS fast kV-switching and DS DECT. KEY POINTS: • Patient- and scan-related factors have a significant impact on the error and minimum detectable difference in dual-energy CT-based iodine quantification. • Third-generation DECTs outperformed second-generation scanners for both single-source and dual-source dual-energy CT. • The minimum intra- and inter-conditional detectable difference in iodine concentration ranged from 0.4 to 1.5 mg iodine/mL.

Can Texture Analysis Be Used to Distinguish Benign From Malignant Adrenal Nodules on Unenhanced CT, Contrast-Enhanced CT, or In-Phase and Opposed-Phase MRI?

OBJECTIVE: The purpose of this study is to determine whether second-order texture analysis can be used to distinguish lipid-poor adenomas from malignant adrenal nodules on unenhanced CT, contrast-enhanced CT (CECT), and chemical-shift MRI. MATERIALS AND METHODS: In this retrospective study, 23 adrenal nodules (15 lipid-poor adenomas and eight adrenal malignancies) in 20 patients (nine female patients and 11 male patients; mean age, 59 years [range, 15-80 years]) were assessed. All patients underwent unenhanced CT, CECT, and chemical-shift MRI. Twenty-one second-order texture features from the gray-level cooccurrence matrix and gray-level run-length matrix were calculated in 3D. The mean values for 21 texture features and four imaging features (lesion size, unenhanced CT attenuation, CECT attenuation, and signal intensity index) were compared using a t test. The diagnostic performance of texture analysis versus imaging features was also compared using AUC values. Multivariate logistic regression models to predict malignancy were constructed for texture analysis and imaging features. RESULTS: Lesion size, unenhanced CT attenuation, and the signal intensity index showed significant differences between benign and malignant adrenal nodules. No significant difference was seen for CECT attenuation. Eighteen of 21 CECT texture features and nine of 21 unenhanced CT texture features revealed significant differences between benign and malignant adrenal nodules. CECT texture features (mean AUC value, 0.80) performed better than CECT attenuation (mean AUC value, 0.60). Multivariate logistic regression models showed that CECT texture features, chemical-shift MRI texture features, and imaging features were predictive of malignancy. CONCLUSION: Texture analysis has a potential role in distinguishing benign from malignant adrenal nodules on CECT and may decrease the need for additional imaging studies in the workup of incidentally discovered adrenal nodules.

A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE-MRI features.

BACKGROUND: Recent studies showed preliminary data on associations of MRI-based imaging phenotypes of breast tumours with breast cancer molecular, genomic, and related characteristics. In this study, we present a comprehensive analysis of this relationship. METHODS: We analysed a set of 922 patients with invasive breast cancer and pre-operative MRI. The MRIs were analysed by a computer algorithm to extract 529 features of the tumour and the surrounding tissue. Machine-learning-based models based on the imaging features were trained using a portion of the data (461 patients) to predict the following molecular, genomic, and proliferation characteristics: tumour surrogate molecular subtype, oestrogen receptor, progesterone receptor and human epidermal growth factor status, as well as a tumour proliferation marker (Ki-67). Trained models were evaluated on the set of the remaining 461 patients. RESULTS: Multivariate models were predictive of Luminal A subtype with AUC = 0.697 (95% CI: 0.647-0.746, p < .0001), triple negative breast cancer with AUC = 0.654 (95% CI: 0.589-0.727, p < .0001), ER status with AUC = 0.649 (95% CI: 0.591-0.705, p < .001), and PR status with AUC = 0.622 (95% CI: 0.569-0.674, p < .0001). Associations between individual features and subtypes we also found. CONCLUSIONS: There is a moderate association between tumour molecular biomarkers and algorithmically assessed imaging features.

Intra-tumor molecular heterogeneity in breast cancer: definitions of measures and association with distant recurrence-free survival.

PURPOSE: The purpose of the study was to define quantitative measures of intra-tumor heterogeneity in breast cancer based on histopathology data gathered from multiple samples on individual patients and determine their association with distant recurrence-free survival (DRFS). METHODS: We collected data from 971 invasive breast cancers, from 1st January 2000 to 23rd March 2014, that underwent repeat tumor sampling at our institution. We defined and calculated 31 measures of intra-tumor heterogeneity including ER, PR, and HER2 immunohistochemistry (IHC), proliferation, EGFR IHC, grade, and histology. For each heterogeneity measure, Cox proportional hazards models were used to determine whether patients with heterogeneous disease had different distant recurrence-free survival (DRFS) than those with homogeneous disease. RESULTS: The presence of heterogeneity in ER percentage staining was prognostic of reduced DRFS with a hazard ratio of 4.26 (95% CI 2.22-8.18, p < 0.00002). It remained significant after controlling for the ER status itself (p < 0.00062) and for patients that had chemotherapy (p < 0.00032). Most of the heterogeneity measures did not show any association with DRFS despite the considerable sample size. CONCLUSIONS: Intra-tumor heterogeneity of ER receptor status may be a predictor of patient DRFS. Histopathologic data from multiple tissue samples may offer a view of tumor heterogeneity and assess recurrence risk.

Comparison of Visualization Rates of LI-RADS Version 2014 Major Features With IV Gadobenate Dimeglumine or Gadoxetate Disodium in Patients at Risk for Hepatocellular Carcinoma.

OBJECTIVE: The purpose of this study is to compare visualization rates of the major features covered by Liver Imaging Reporting and Data System (LI-RADS) version 2014 in patients at risk for hepatocellular carcinoma using either gadobenate dimeglumine or gadoxetate disodium IV contrast agent. MATERIALS AND METHODS: This retrospective study included liver MRI examinations performed with either gadobenate dimeglumine or gadoxetate disodium contrast enhancement. Using age, sex, underlying liver disease, and presence of cirrhosis, patients were placed into matched cohorts. All hepatic nodules 1 cm or larger (up to five per subject) were included, resulting in 63 subjects with 130 nodules (median nodule size, 1.9 cm) imaged with gadobenate and 64 subjects with 117 nodules (median nodule size, 2.0 cm) imaged with gadoxetate. Three radiologists reviewed the studies for LI-RADS major features independently. Bootstrap resampling with 10,000 repetitions was used to compare feature detection rates. RESULTS: Arterial phase hyperenhancement was seen in a similar number of nodules with gadobenate dimeglumine (mean, 91.5% [119/130]) and gadoxetate disodium (mean, 88.0% [103/117]) (p = 0.173). Dynamic phase washout was more commonly seen with gadobenate dimeglumine (mean, 60.2% [78.3/130]) than with gadoxetate disodium (mean, 45.3% [53/117]) (p = 0.006). The capsule feature was more often visualized with gadobenate dimeglumine (mean, 50.2% [65.3/130]) than with gadoxetate disodium (mean, 33.3% [39/117]) (p < 0.001). Interreader agreement for arterial phase enhancement and dynamic phase washout was almost perfect for both contrast agents (κ > 0.83). Agreement for the capsule feature was moderate for gadobenate dimeglumine (κ = 0.52) and substantial for gadoxetate disodium (κ = 0.67). CONCLUSION: The rates of visualization of arterial phase hyperenhancement are similar in studies performed with gadobenate dimeglumine and gadoxetate disodium, but dynamic phase washout and capsule appearance are more commonly visualized with gadobenate dimeglumine.

Algorithms for prediction of the Oncotype DX recurrence score using clinicopathologic data: a review and comparison using an independent dataset.

PURPOSE: Given the potential savings in cost and resource utilization, several algorithms have been proposed to predict Oncotype DX recurrence score (ODX RS) using commonly acquired histopathologic variables. Although it is promising, additional independent validation of these surrogate markers is needed prior to guide the patient management. METHODS: In this retrospective study, we analyzed 305 patients with invasive breast cancer at our institution who had ODX RS available. We selected five equations that provide a surrogate measure of ODX as previously published by Klein et al. (Magee equations 1-3), Gage et al., and Tang et al. All equations used estrogen receptor status and progesterone receptor status along with different combinations of grade, proliferation indices (Ki-67, mitotic rate), HER2 status, and tumor size. RESULTS: Of all surrogate scores tested, the Magee equation 2 provided the highest correlation with ODX both with regard to raw score (Pearson's correlation coefficient = 0.66 95% CI 0.59-0.72) and categorical correlation (Cohen's kappa = 0.43, 95% CI 0.33-0.53). Although Magee equation 2 provided a way to reliably identify high-risk disease by assigning 95% of the patients with high ODX RS to either the intermediate- or high-risk group, it was unable to reliably identify the potential for patients to have intermediate- or high-risk disease by ODX (66% of such patients identified). CONCLUSIONS: Although commonly available surrogates for ODX appear to predict high-risk ODX RS, they are unable to reliably rule out the presence of patients with intermediate-risk disease by ODX. Given the potential benefit of adjuvant chemotherapy in women with intermediate-risk disease by ODX, current surrogates are unable to safely substitute for ODX. Characterizing the true recurrence risk in patients with intermediate-risk disease by ODX is critical to the clinical adoption of current surrogate markers and is an area of ongoing clinical trials.

Can algorithmically assessed MRI features predict which patients with a preoperative diagnosis of ductal carcinoma in situ are upstaged to invasive breast cancer?

PURPOSE: To assess the ability of algorithmically assessed magnetic resonance imaging (MRI) features to predict the likelihood of upstaging to invasive cancer in newly diagnosed ductal carcinoma in situ (DCIS). MATERIALS AND METHODS: We identified 131 patients at our institution from 2000-2014 with a core needle biopsy-confirmed diagnosis of pure DCIS, a 1.5 or 3T preoperative bilateral breast MRI with nonfat-saturated T1 -weighted MRI sequences, no preoperative therapy before breast MRI, and no prior history of breast cancer. A fellowship-trained radiologist identified the lesion on each breast MRI using a bounding box. Twenty-nine imaging features were then computed automatically using computer algorithms based on the radiologist's annotation. RESULTS: The rate of upstaging of DCIS to invasive cancer in our study was 26.7% (35/131). Out of all imaging variables tested, the information measure of correlation 1, which quantifies spatial dependency in neighboring voxels of the tumor, showed the highest predictive value of upstaging with an area under the curve (AUC) = 0.719 (95% confidence interval [CI]: 0.609-0.829). This feature was statistically significant after adjusting for tumor size (P < 0.001). CONCLUSION: Automatically assessed MRI features may have a role in triaging which patients with a preoperative diagnosis of DCIS are at highest risk for occult invasive disease. LEVEL OF EVIDENCE: 4 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1332-1340.