Quantifying effects of blood pressure control on neuroimaging utilization in a large multi-institutional healthcare population.

OBJECTIVES: Essential hypertension is a common chronic condition that can exacerbate or complicate various neurological diseases that may necessitate neuroimaging. Given growing medical imaging costs and the need to understand relationships between population blood pressure control and neuroimaging utilization, we seek to quantify the relationship between maximum blood pressure recorded in a given year and same-year utilization of neuroimaging CT or MR in a large healthcare population. METHODS: A retrospective population-based cohort study was performed by extracting aggregate data from a multi-institutional dataset of patient encounters from 2016, 2018, and 2020 using an informatics platform (Cosmos) consisting of de-duplicated data from over 140 academic and non-academic health systems, comprising over 137 million unique patients. A population-based sample of all patients with recorded blood pressures of at least 50 mmHg DBP or 90 mmHg SBP were included. Cohorts were identified based on maximum annual SBP and DBP meeting or exceeding pre-defined thresholds. For each cohort, we assessed neuroimaging CT and MR utilization, defined as the percentage of patients undergoing ≥1 neuroimaging exam of interest in the same calendar year. RESULTS: The multi-institutional population consisted of >38 million patients for the most recent calendar year analyzed, with overall utilization of 3.8-5.1% for CT and 1.5-2.0% for MR across the study period. Neuroimaging utilization increased substantially with increasing annual maximum BP. Even a modest BP increase to 140 mmHg systolic or 90 mmHg diastolic is associated with 3-4-fold increases in MR and 5-7-fold increases in CT same-year imaging compared to BP values below 120 mmHg / 80 mmHg. CONCLUSION: Higher annual maximum recorded blood pressure is associated with higher same-year neuroimaging CT and MR utilization rates. These observations are relevant to public health efforts on hypertension management to mitigate costs associated with growing imaging utilization.

Artificial Intelligence in Neuroradiology: A Review of Current Topics and Competition Challenges.

There is an expanding body of literature that describes the application of deep learning and other machine learning and artificial intelligence methods with potential relevance to neuroradiology practice. In this article, we performed a literature review to identify recent developments on the topics of artificial intelligence in neuroradiology, with particular emphasis on large datasets and large-scale algorithm assessments, such as those used in imaging AI competition challenges. Numerous applications relevant to ischemic stroke, intracranial hemorrhage, brain tumors, demyelinating disease, and neurodegenerative/neurocognitive disorders were discussed. The potential applications of these methods to spinal fractures, scoliosis grading, head and neck oncology, and vascular imaging were also reviewed. The AI applications examined perform a variety of tasks, including localization, segmentation, longitudinal monitoring, diagnostic classification, and prognostication. While research on this topic is ongoing, several applications have been cleared for clinical use and have the potential to augment the accuracy or efficiency of neuroradiologists.

Topics most predictive of favorable overall assessment in outpatient radiology.

BACKGROUND: Patients' subjective experiences during clinical interactions may affect their engagement in healthcare, and better understanding of the issues patients consider most important may help improve service quality and patient-staff relationships. While diagnostic imaging is a growing component of healthcare utilization, few studies have quantitatively and systematically assessed what patients deem most relevant in radiology settings. To elucidate factors driving patient satisfaction in outpatient radiology, we derived quantitative models to identify items most predictive of patients' overall assessment of radiology encounters. METHODS: Press-Ganey survey data (N = 69,319) collected over a 9-year period at a single institution were retrospectively analyzed, with each item response dichotomized as "favorable" or "unfavorable." Multiple logistic regression analyses were performed on 18 binarized Likert items to compute odds ratios (OR) for those question items significantly predicting Overall Rating of Care or Likelihood of Recommending. In a secondary analysis to identify topics more relevant to radiology than other encounter types, items significantly more predictive of concordant ratings in radiology compared to non-radiology visits were also identified. RESULTS: Among radiology survey respondents, top predictors of Overall Rating and Likelihood of Recommending were items addressing patient concerns or complaints (OR 6.8 and 4.9, respectively) and sensitivity to patient needs (OR 4.7 and 4.5, respectively). When comparing radiology and non-radiology visits, the top items more predictive for radiology included unfavorable responses to helpfulness of registration desk personnel (OR 1.4-1.6), comfort of waiting areas (OR 1.4), and ease of obtaining an appointment at the desired time (OR 1.4). CONCLUSIONS: Items related to patient-centered empathic communication were the most predictive of favorable overall ratings among radiology outpatients, while underperformance in logistical issues related to registration, scheduling, and waiting areas may have greater adverse impact on radiology than non-radiology encounters. Findings may offer potential targets for future quality improvement efforts.

Prediction of model generalizability for unseen data: Methodology and case study in brain metastases detection in T1-Weighted contrast-enhanced 3D MRI.

BACKGROUND AND PURPOSE: A medical AI system's generalizability describes the continuity of its performance acquired from varying geographic, historical, and methodologic settings. Previous literature on this topic has mostly focused on "how" to achieve high generalizability (e.g., via larger datasets, transfer learning, data augmentation, model regularization schemes), with limited success. Instead, we aim to understand "when" the generalizability is achieved: Our study presents a medical AI system that could estimate its generalizability status for unseen data on-the-fly. MATERIALS AND METHODS: We introduce a latent space mapping (LSM) approach utilizing Fréchet distance loss to force the underlying training data distribution into a multivariate normal distribution. During the deployment, a given test data's LSM distribution is processed to detect its deviation from the forced distribution; hence, the AI system could predict its generalizability status for any previously unseen data set. If low model generalizability is detected, then the user is informed by a warning message integrated into a sample deployment workflow. While the approach is applicable for most classification deep neural networks (DNNs), we demonstrate its application to a brain metastases (BM) detector for T1-weighted contrast-enhanced (T1c) 3D MRI. The BM detection model was trained using 175 T1c studies acquired internally (from the authors' institution) and tested using (1) 42 internally acquired exams and (2) 72 externally acquired exams from the publicly distributed Brain Mets dataset provided by the Stanford University School of Medicine. Generalizability scores, false positive (FP) rates, and sensitivities of the BM detector were computed for the test datasets. RESULTS AND CONCLUSION: The model predicted its generalizability to be low for 31% of the testing data (i.e., two of the internally and 33 of the externally acquired exams), where it produced (1) ∼13.5 false positives (FPs) at 76.1% BM detection sensitivity for the low and (2) ∼10.5 FPs at 89.2% BM detection sensitivity for the high generalizability groups respectively. These results suggest that the proposed formulation enables a model to predict its generalizability for unseen data.

Advancing Brain Metastases Detection in T1-Weighted Contrast-Enhanced 3D MRI Using Noisy Student-Based Training.

The detection of brain metastases (BM) in their early stages could have a positive impact on the outcome of cancer patients. The authors previously developed a framework for detecting small BM (with diameters of <15 mm) in T1-weighted contrast-enhanced 3D magnetic resonance images (T1c). This study aimed to advance the framework with a noisy-student-based self-training strategy to use a large corpus of unlabeled T1c data. Accordingly, a sensitivity-based noisy-student learning approach was formulated to provide high BM detection sensitivity with a reduced count of false positives. This paper (1) proposes student/teacher convolutional neural network architectures, (2) presents data and model noising mechanisms, and (3) introduces a novel pseudo-labeling strategy factoring in the sensitivity constraint. The evaluation was performed using 217 labeled and 1247 unlabeled exams via two-fold cross-validation. The framework utilizing only the labeled exams produced 9.23 false positives for 90% BM detection sensitivity, whereas the one using the introduced learning strategy led to ~9% reduction in false detections (i.e., 8.44). Significant reductions in false positives (>10%) were also observed in reduced labeled data scenarios (using 50% and 75% of labeled data). The results suggest that the introduced strategy could be utilized in existing medical detection applications with access to unlabeled datasets to elevate their performances.

Mortality Prediction Analysis among COVID-19 Inpatients Using Clinical Variables and Deep Learning Chest Radiography Imaging Features.

The emergence of the COVID-19 pandemic over a relatively brief interval illustrates the need for rapid data-driven approaches to facilitate clinical decision making. We examined a machine learning process to predict inpatient mortality among COVID-19 patients using clinical and chest radiographic data. Modeling was performed with a de-identified dataset of encounters prior to widespread vaccine availability. Non-imaging predictors included demographics, pre-admission clinical history, and past medical history variables. Imaging features were extracted from chest radiographs by applying a deep convolutional neural network with transfer learning. A multi-layer perceptron combining 64 deep learning features from chest radiographs with 98 patient clinical features was trained to predict mortality. The Local Interpretable Model-Agnostic Explanations (LIME) method was used to explain model predictions. Non-imaging data alone predicted mortality with an ROC-AUC of 0.87 ± 0.03 (mean ± SD), while the addition of imaging data improved prediction slightly (ROC-AUC: 0.91 ± 0.02). The application of LIME to the combined imaging and clinical model found HbA1c values to contribute the most to model prediction (17.1 ± 1.7%), while imaging contributed 8.8 ± 2.8%. Age, gender, and BMI contributed 8.7%, 8.2%, and 7.1%, respectively. Our findings demonstrate a viable explainable AI approach to quantify the contributions of imaging and clinical data to COVID mortality predictions.

Trends and Predictors of Imaging Utilization by Modality within an Academic Health System's Active Patient Population.

RATIONALE AND OBJECTIVES: Evaluate trends and demographic predictors of imaging utilization at a university-affiliated health system. MATERIALS AND METHODS: In this single-institution retrospective study, per capita estimates of imaging utilization among patients active in the health system were computed by cross-referencing all clinical encounters (2004-2016) for 1,628,980 unique patients with a listing of 6,157,303 diagnostic radiology encounters. Time trends in imaging utilization and effects of gender, race/ethnicity, and age were assessed, with subgroup analyses performed by imaging modality. Utilization was analyzed as both a continuous and binary outcome variable. RESULTS: Over 13 years, total diagnostic exams rose 6.8% a year (285,947-622,196 exams per annum), while the active population size grew 7.0% a year (244,238-543,290 active patients per annum). Per capita utilization peaked in 2007 at 1.33 studies/patient/year before dropping to 1.06 from 2011 to 2015. Latest per capita utilization was 0.22 for computed tomography, 0.10 for MR, 0.20 for US, 0.03 for NM, 0.51 for radiography, and 0.07 for mammography. Over the study period, ultrasound utilization doubled, whereas NM and radiography utilization decreased. computed tomography, MR, and mammography showed no significant net change. Univariate analysis of utilization as a continuous variable showed statistically significant effects of gender, race/ethnicity, and age (P < 0.0001), with utilization higher in males and Blacks and lower in Asian/Pacific Islanders and Hispanics. Utilization increased with age, except for a decline after age 75. Many of the effects of age, gender, and race/ethnicity were also found when analyzing the binarized utilization variable. CONCLUSIONS: Although absolute counts of imaging studies more than doubled, the net change in per capita utilization over the study period was minimal. Variations in utilization across age, gender, and race/ethnicity may reflect differential health needs and/or access disparities, warranting future studies.

Augmented networks for faster brain metastases detection in T1-weighted contrast-enhanced 3D MRI.

Early detection of brain metastases (BM) is one of the determining factors for the successful treatment of patients with cancer; however, the accurate detection of small BM lesions (< 15 mm) remains a challenging task. We previously described a framework for the detection of small BM in single-sequence gadolinium-enhanced T1-weighted 3D MRI datasets. It combined classical image processing (IP) with a dedicated convolutional neural network, taking approximately 30 s to process each dataset due to computation-intensive IP stages. To overcome the speed limitation, this study aims to reformulate the framework via an augmented pair of CNNs (eliminating the IP) to reduce the processing times while preserving the BM detection performance. Our previous implementation of the BM detection algorithm utilized Laplacian of Gaussians (LoG) for the candidate selection portion of the solution. In this study, we introduce a novel BM candidate detection CNN (cdCNN) to replace this classical IP stage. The network is formulated to have (1) a similar receptive field as the LoG method, and (2) a bias for the detection of BM lesion loci. The proposed CNN is later augmented with a classification CNN to perform the BM detection task. The cdCNN achieved 97.4% BM detection sensitivity when producing 60 K candidates per 3D MRI dataset, while the LoG achieved 96.5% detection sensitivity with 73 K candidates. The augmented BM detection framework generated on average 9.20 false-positive BM detections per patient for 90% sensitivity, which is comparable with our previous results. However, it processes each 3D data in 1.9 s, presenting a 93.5% reduction in the computation time.

Magnetic Resonance Elastography of Intervertebral Discs: Spin-Echo Echo-Planar Imaging Sequence Validation.

BACKGROUND: Magnetic resonance elastography (MRE) is an imaging technique that can noninvasively assess the shear properties of the intervertebral disc (IVD). Unlike the standard gradient recalled echo (GRE) MRE technique, a spin-echo echo-planar imaging (SE-EPI) sequence has the potential to improve imaging efficiency and patient compliance. PURPOSE: To validate the use of an SE-EPI sequence for MRE of the IVD compared against the standard GRE sequence. STUDY TYPE: Cross-over. SUBJECTS: Twenty-eight healthy volunteers (15 males and 13 females, age range: 19-55). FIELD STRENGTH/SEQUENCE: 3 T; GRE, SE-EPI with breath holds (SE-EPI-BH) and SE-EPI with free breathing (SE-EPI-FB) MRE sequences. ASSESSMENT: MRE-derived shear stiffnesses were calculated via principal frequency analysis. SE-EPI derived shear stiffness and octahedral shear strain signal-to-noise ratios (OSS-SNR) were compared against those derived using the GRE sequence. The reproducibility and repeatability of SE-EPI stiffness measurements were determined. Shear stiffness was evaluated in the nucleus pulposus (NP) and annulus fibrosus (AF) regions of the disc. Scan times between sequences were compared. STATISTICAL TESTS: Linear mixed models, Bland-Altman plots, and Lin's concordance correlation coefficients (CCCs) were used with P < 0.05 considered statistically significant. RESULTS: Good correlation was observed between shear stiffnesses derived from the SE-EPI sequences with those derived from the GRE sequence with CCC values greater than 0.73 and 0.78 for the NP and AF regions, respectively. OSS-SNR was not significantly different between GRE and SE-EPI sequences (P > 0.05). SE-EPI sequences generated highly reproducible and repeatable stiffness measurements with CCC values greater than 0.97 in the NP and AF regions and reduced scan time by at least 51% compared to GRE. SE-EPI-BH and SE-EPI-FB stiffness measurements were similar with CCC values greater than 0.98 for both regions. DATA CONCLUSION: SE-EPI-based MRE-derived stiffnesses were highly reproducible and repeatable and correlated with current standard GRE MRE-derived stiffness estimates while reducing scan times. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY STAGE: 1.

Does Patient Satisfaction Drive Volumes in Outpatient Magnetic Resonance Imaging?

OBJECTIVE: To help quantify the potential microeconomic impact of patient satisfaction in radiology, we tested the hypothesis that patient volume trends reflect patient satisfaction trends in outpatient magnetic resonance imaging (MRI). METHODS: Patient visits (N = 39,595) at distinct outpatient MRI sites within a university-affiliated hospital system during a 1-year period were retrospectively analyzed. Individual sites were grouped as having "decreasing," "stable," or "increasing" volume using an average quarterly volume change threshold of 5%. Based on Press Ganey outpatient services surveys, changes in satisfaction scores from the baseline quarter were calculated. Mood's median tests were applied to assess statistical significance of differences in satisfaction score improvements among the three volume trend designations during the 3 post-baseline fiscal quarters. RESULTS: Quarterly volume was stable at 6 sites, increased at 1 site (by 18%), and decreased at 2 sites (by 20%-24%). There was a statistically significant association between volume trend and net change in satisfaction scores for all 5 domains assessed on the Press Ganey survey: Overall assessment (P < 0.0001), Facilities (P = 0.026), Personal issues (P = 0.013), Registration (P = 0.0004), and Test or treatment (P < 0.0001), with median score changes generally higher at facilities with higher volume trends. DISCUSSION: It can be inferred that patient satisfaction drives volume in this scenario, whereas the converse relationship of volume adversely affecting satisfaction is not observed. Patient satisfaction and volume at MRI sites are interrelated, and patient experiences or perceptions of quality may influence decisions regarding what imaging sites are preferentially utilized.

Training Strategies for Radiology Deep Learning Models in Data-limited Scenarios.

Data-driven approaches have great potential to shape future practices in radiology. The most straightforward strategy to obtain clinically accurate models is to use large, well-curated and annotated datasets. However, patient privacy constraints, tedious annotation processes, and the limited availability of radiologists pose challenges to building such datasets. This review details model training strategies in scenarios with limited data, insufficiently labeled data, and/or limited expert resources. This review discusses strategies to enlarge the data sample, decrease the time burden of manual supervised labeling, adjust the neural network architecture to improve model performance, apply semisupervised approaches, and leverage efficiencies from pretrained models. Keywords: Computer-aided Detection/Diagnosis, Transfer Learning, Limited Annotated Data, Augmentation, Synthetic Data, Semisupervised Learning, Federated Learning, Few-Shot Learning, Class Imbalance.

Patient Satisfaction in Outpatient Radiology: Effects of Modality and Patient Demographic Characteristics.

Objective: To characterize predictors of patient satisfaction in outpatient radiology, we examined whether patient satisfaction differs across radiology modalities and demographic groups. Methods: A random sampling of Press-Ganey outpatient services surveys for radiology and non-radiology visits from September 2008 to September 2017 were retrospectively analyzed. Composite scores averaged across all Likert items were analyzed as both a continuous variable and a dichotomous variable of dissatisfaction (defined as ≤3 on the 5-point scale). Results: Among 9983 radiology surveys, mammography had higher composite scores than MRI, CT, radiography, US, and NM/PET (p < 0.001) and lower dissatisfaction (3.9%) than CT (6.7%), MRI (7.3%), and radiography (8.2%). Low-scoring responses were most common in the Facilities domain (7.8%) and least common in Overall Assessment (3.8%). Satisfaction metrics were lowest for ages 20-29 and highest for ages 70-79. Lower dissatisfaction rates were seen among Hispanics (3%) and whites (6%), compared to blacks (10%) and Asians (18%). Conclusion: Significant differences in patient satisfaction were found across imaging modalities and demographic variables. Further investigations to identify contributing factors may help improve patient experiences.

Review of Artificial Intelligence Training Tools and Courses for Radiologists.

Artificial intelligence (AI) systems play an increasingly important role in all parts of the imaging chain, from image creation to image interpretation to report generation. In order to responsibly manage radiology AI systems and make informed purchase decisions about them, radiologists must understand the underlying principles of AI. Our task force was formed by the Radiology Research Alliance (RRA) of the Association of University Radiologists to identify and summarize a curated list of current educational materials available for radiologists.

Neck CT imaging and correlation with thyroid cancer incidence across age, gender and race.

OBJECTIVE: Incidental detection of thyroid cancers has been proposed as a cause of thyroid cancer increases over past decades, but few studies assess the impact of imaging utilization on thyroid cancer incidence. This study quantifies neck CT prevalence and its relationship with thyroid cancer incidence as a function of age, sex and race. DESIGN AND PATIENTS: Medical records of over 1 million patients at our institution were retrospectively analysed to quantify neck CT prevalence from 2004 to 2011 (study period). A national cancer database was used to compute thyroid cancer incidences over the study period and a reference period (1974-81) and to calculate change in thyroid incidence between the two periods. Both populations were partitioned into demographic subgroups of varying age, sex and race. Linear correlation between neck imaging and thyroid cancer incidence changes among subgroups was assessed using Pearson's correlation. RESULTS: Neck CT imaging and change in thyroid cancer incidence varied across all examined demographic variables, particularly age. When stratifying by age, CT use correlated strongly with recent national thyroid cancer incidence (R = .97) and with 30-year change in thyroid cancer incidence (R = .87). Across all demographic subgroups, CT prevalence correlated strongly and positively with change in thyroid cancer incidence (R = .60), greater for whites (R = .60) and blacks (R = .70) than other races (R = .28). CONCLUSION: Differences in neck CT usage strongly and positively correlates with the variation in thyroid cancer trends based on age, gender and race.

Noninterpretive Uses of Artificial Intelligence in Radiology.

We deem a computer to exhibit artificial intelligence (AI) when it performs a task that would normally require intelligent action by a human. Much of the recent excitement about AI in the medical literature has revolved around the ability of AI models to recognize anatomy and detect pathology on medical images, sometimes at the level of expert physicians. However, AI can also be used to solve a wide range of noninterpretive problems that are relevant to radiologists and their patients. This review summarizes some of the newer noninterpretive uses of AI in radiology.

Predicting Mental Decline Rates in Mild Cognitive Impairment From Baseline MRI Volumetric Data.

PURPOSE: In mild cognitive impairment (MCI), identifying individuals at high risk for progressive cognitive deterioration can be useful for prognostication and intervention. This study quantitatively characterizes cognitive decline rates in MCI and tests whether volumetric data from baseline magnetic resonance imaging (MRI) can predict accelerated cognitive decline. METHODS: The authors retrospectively examined Alzheimer Disease Neuroimaging Initiative data to obtain serial Mini-Mental Status Exam (MMSE) scores, diagnoses, and the following baseline MRI volumes: total intracranial volume, whole-brain and ventricular volumes, and volumes of the hippocampus, entorhinal cortex, fusiform gyrus, and medial temporal lobe. Subjects with <24 months or <4 measurements of MMSE data were excluded. Predictive modeling of fast cognitive decline (defined as >0.6/year) from baseline volumetric data was performed on subjects with MCI using a single hidden layer neural network. RESULTS: Among 698 baseline MCI subjects, the median annual decline in the MMSE score was 1.3 for converters to dementia versus 0.11 for stable MCI (P<0.001). A 0.6/year threshold captured dementia conversion with 82% accuracy (sensitivity 79%, specificity 85%, area under the receiver operating characteristic curve 0.88). Regional volumes on baseline MRI predicted fast cognitive decline with a test accuracy of 71%. DISCUSSION: An MMSE score decrease of >0.6/year is associated with MCI-to-dementia conversion and can be predicted from baseline MRI.

Thyroid Incidentalomas: Practice Considerations for Radiologists in the Age of Incidental Findings.

Radiologists very frequently encounter incidental findings related to the thyroid gland. Given increases in imaging use over the past several decades, thyroid incidentalomas are increasingly encountered in clinical practice, and it is important for radiologists to be aware of recent developments with respect to workup and diagnosis of incidental thyroid abnormalities. Recent reporting and management guidelines, such as those from the American College of Radiology and American Thyroid Association, are reviewed along with applicable evidence in the literature. Trending topics, such as artificial intelligence approaches to guide thyroid incidentaloma workup, are also discussed.

Predicting rate of cognitive decline at baseline using a deep neural network with multidata analysis.

Purpose: Our study investigates whether a machine-learning-based system can predict the rate of cognitive decline in mildly cognitively impaired patients by processing only the clinical and imaging data collected at the initial visit. Approach: We built a predictive model based on a supervised hybrid neural network utilizing a three-dimensional convolutional neural network to perform volume analysis of magnetic resonance imaging (MRI) and integration of nonimaging clinical data at the fully connected layer of the architecture. The experiments are conducted on the Alzheimer's Disease Neuroimaging Initiative dataset. Results: Experimental results confirm that there is a correlation between cognitive decline and the data obtained at the first visit. The system achieved an area under the receiver operator curve of 0.70 for cognitive decline class prediction. Conclusion: To our knowledge, this is the first study that predicts "slowly deteriorating/stable" or "rapidly deteriorating" classes by processing routinely collected baseline clinical and demographic data [baseline MRI, baseline mini-mental state examination (MMSE), scalar volumetric data, age, gender, education, ethnicity, and race]. The training data are built based on MMSE-rate values. Unlike the studies in the literature that focus on predicting mild cognitive impairment (MCI)-to-Alzheimer's disease conversion and disease classification, we approach the problem as an early prediction of cognitive decline rate in MCI patients.