Association of Coronary Artery Calcium Detected by Routine Ungated CT Imaging With Cardiovascular Outcomes.

BACKGROUND: Coronary artery calcium (CAC) is a strong predictor of cardiovascular events across all racial and ethnic groups. CAC can be quantified on nonelectrocardiography (ECG)-gated computed tomography (CT) performed for other reasons, allowing for opportunistic screening for subclinical atherosclerosis. OBJECTIVES: The authors investigated whether incidental CAC quantified on routine non-ECG-gated CTs using a deep-learning (DL) algorithm provided cardiovascular risk stratification beyond traditional risk prediction methods. METHODS: Incidental CAC was quantified using a DL algorithm (DL-CAC) on non-ECG-gated chest CTs performed for routine care in all settings at a large academic medical center from 2014 to 2019. We measured the association between DL-CAC (0, 1-99, or ≥100) with all-cause death (primary outcome), and the secondary composite outcomes of death/myocardial infarction (MI)/stroke and death/MI/stroke/revascularization using Cox regression. We adjusted for age, sex, race, ethnicity, comorbidities, systolic blood pressure, lipid levels, smoking status, and antihypertensive use. Ten-year atherosclerotic cardiovascular disease risk was calculated using the pooled cohort equations. RESULTS: Of 5,678 adults without ASCVD (51% women, 18% Asian, 13% Hispanic/Latinx), 52% had DL-CAC >0. Those with DL-CAC ≥100 had an average 10-year ASCVD risk of 24%; yet, only 26% were on statins. After adjustment, patients with DL-CAC ≥100 had increased risk of death (HR: 1.51; 95% CI: 1.28-1.79), death/MI/stroke (HR: 1.57; 95% CI: 1.33-1.84), and death/MI/stroke/revascularization (HR: 1.69; 95% CI: 1.45-1.98) compared with DL-CAC = 0. CONCLUSIONS: Incidental CAC ≥100 was associated with an increased risk of all-cause death and adverse cardiovascular outcomes, beyond traditional risk factors. DL-CAC from routine non-ECG-gated CTs identifies patients at increased cardiovascular risk and holds promise as a tool for opportunistic screening to facilitate earlier intervention.

Opportunistic detection of type 2 diabetes using deep learning from frontal chest radiographs.

Deep learning (DL) models can harness electronic health records (EHRs) to predict diseases and extract radiologic findings for diagnosis. With ambulatory chest radiographs (CXRs) frequently ordered, we investigated detecting type 2 diabetes (T2D) by combining radiographic and EHR data using a DL model. Our model, developed from 271,065 CXRs and 160,244 patients, was tested on a prospective dataset of 9,943 CXRs. Here we show the model effectively detected T2D with a ROC AUC of 0.84 and a 16% prevalence. The algorithm flagged 1,381 cases (14%) as suspicious for T2D. External validation at a distinct institution yielded a ROC AUC of 0.77, with 5% of patients subsequently diagnosed with T2D. Explainable AI techniques revealed correlations between specific adiposity measures and high predictivity, suggesting CXRs' potential for enhanced T2D screening.

Incidental Coronary Artery Calcium: Opportunistic Screening of Previous Nongated Chest Computed Tomography Scans to Improve Statin Rates (NOTIFY-1 Project).

BACKGROUND: Coronary artery calcium (CAC) can be identified on nongated chest computed tomography (CT) scans, but this finding is not consistently incorporated into care. A deep learning algorithm enables opportunistic CAC screening of nongated chest CT scans. Our objective was to evaluate the effect of notifying clinicians and patients of incidental CAC on statin initiation. METHODS: NOTIFY-1 (Incidental Coronary Calcification Quality Improvement Project) was a randomized quality improvement project in the Stanford Health Care System. Patients without known atherosclerotic cardiovascular disease or a previous statin prescription were screened for CAC on a previous nongated chest CT scan from 2014 to 2019 using a validated deep learning algorithm with radiologist confirmation. Patients with incidental CAC were randomly assigned to notification of the primary care clinician and patient versus usual care. Notification included a patient-specific image of CAC and guideline recommendations regarding statin use. The primary outcome was statin prescription within 6 months. RESULTS: Among 2113 patients who met initial clinical inclusion criteria, CAC was identified by the algorithm in 424 patients. After chart review and additional exclusions were made, a radiologist confirmed CAC among 173 of 194 patients (89.2%) who were randomly assigned to notification or usual care. At 6 months, the statin prescription rate was 51.2% (44/86) in the notification arm versus 6.9% (6/87) with usual care (P<0.001). There was also more coronary artery disease testing in the notification arm (15.1% [13/86] versus 2.3% [2/87]; P=0.008). CONCLUSIONS: Opportunistic CAC screening of previous nongated chest CT scans followed by clinician and patient notification led to a significant increase in statin prescriptions. Further research is needed to determine whether this approach can reduce atherosclerotic cardiovascular disease events. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT04789278.

CheXED: Comparison of a Deep Learning Model to a Clinical Decision Support System for Pneumonia in the Emergency Department.

PURPOSE: Patients with pneumonia often present to the emergency department (ED) and require prompt diagnosis and treatment. Clinical decision support systems for the diagnosis and management of pneumonia are commonly utilized in EDs to improve patient care. The purpose of this study is to investigate whether a deep learning model for detecting radiographic pneumonia and pleural effusions can improve functionality of a clinical decision support system (CDSS) for pneumonia management (ePNa) operating in 20 EDs. MATERIALS AND METHODS: In this retrospective cohort study, a dataset of 7434 prior chest radiographic studies from 6551 ED patients was used to develop and validate a deep learning model to identify radiographic pneumonia, pleural effusions, and evidence of multilobar pneumonia. Model performance was evaluated against 3 radiologists' adjudicated interpretation and compared with performance of the natural language processing of radiology reports used by ePNa. RESULTS: The deep learning model achieved an area under the receiver operating characteristic curve of 0.833 (95% confidence interval [CI]: 0.795, 0.868) for detecting radiographic pneumonia, 0.939 (95% CI: 0.911, 0.962) for detecting pleural effusions and 0.847 (95% CI: 0.800, 0.890) for identifying multilobar pneumonia. On all 3 tasks, the model achieved higher agreement with the adjudicated radiologist interpretation compared with ePNa. CONCLUSIONS: A deep learning model demonstrated higher agreement with radiologists than the ePNa CDSS in detecting radiographic pneumonia and related findings. Incorporating deep learning models into pneumonia CDSS could enhance diagnostic performance and improve pneumonia management.

Artificial Intelligence Algorithm Improves Radiologist Performance in Skeletal Age Assessment: A Prospective Multicenter Randomized Controlled Trial.

Background Previous studies suggest that use of artificial intelligence (AI) algorithms as diagnostic aids may improve the quality of skeletal age assessment, though these studies lack evidence from clinical practice. Purpose To compare the accuracy and interpretation time of skeletal age assessment on hand radiograph examinations with and without the use of an AI algorithm as a diagnostic aid. Materials and Methods In this prospective randomized controlled trial, the accuracy of skeletal age assessment on hand radiograph examinations was performed with (n = 792) and without (n = 739) the AI algorithm as a diagnostic aid. For examinations with the AI algorithm, the radiologist was shown the AI interpretation as part of their routine clinical work and was permitted to accept or modify it. Hand radiographs were interpreted by 93 radiologists from six centers. The primary efficacy outcome was the mean absolute difference between the skeletal age dictated into the radiologists' signed report and the average interpretation of a panel of four radiologists not using a diagnostic aid. The secondary outcome was the interpretation time. A linear mixed-effects regression model with random center- and radiologist-level effects was used to compare the two experimental groups. Results Overall mean absolute difference was lower when radiologists used the AI algorithm compared with when they did not (5.36 months vs 5.95 months; P = .04). The proportions at which the absolute difference exceeded 12 months (9.3% vs 13.0%, P = .02) and 24 months (0.5% vs 1.8%, P = .02) were lower with the AI algorithm than without it. Median radiologist interpretation time was lower with the AI algorithm than without it (102 seconds vs 142 seconds, P = .001). Conclusion Use of an artificial intelligence algorithm improved skeletal age assessment accuracy and reduced interpretation times for radiologists, although differences were observed between centers. Clinical trial registration no. NCT03530098 © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Rubin in this issue.

Automated coronary calcium scoring using deep learning with multicenter external validation.

Coronary artery disease (CAD), the most common manifestation of cardiovascular disease, remains the most common cause of mortality in the United States. Risk assessment is key for primary prevention of coronary events and coronary artery calcium (CAC) scoring using computed tomography (CT) is one such non-invasive tool. Despite the proven clinical value of CAC, the current clinical practice implementation for CAC has limitations such as the lack of insurance coverage for the test, need for capital-intensive CT machines, specialized imaging protocols, and accredited 3D imaging labs for analysis (including personnel and software). Perhaps the greatest gap is the millions of patients who undergo routine chest CT exams and demonstrate coronary artery calcification, but their presence is not often reported or quantitation is not feasible. We present two deep learning models that automate CAC scoring demonstrating advantages in automated scoring for both dedicated gated coronary CT exams and routine non-gated chest CTs performed for other reasons to allow opportunistic screening. First, we trained a gated coronary CT model for CAC scoring that showed near perfect agreement (mean difference in scores = -2.86; Cohen's Kappa = 0.89, P < 0.0001) with current conventional manual scoring on a retrospective dataset of 79 patients and was found to perform the task faster (average time for automated CAC scoring using a graphics processing unit (GPU) was 3.5 ± 2.1 s vs. 261 s for manual scoring) in a prospective trial of 55 patients with little difference in scores compared to three technologists (mean difference in scores = 3.24, 5.12, and 5.48, respectively). Then using CAC scores from paired gated coronary CT as a reference standard, we trained a deep learning model on our internal data and a cohort from the Multi-Ethnic Study of Atherosclerosis (MESA) study (total training n = 341, Stanford test n = 42, MESA test n = 46) to perform CAC scoring on routine non-gated chest CT exams with validation on external datasets (total n = 303) obtained from four geographically disparate health systems. On identifying patients with any CAC (i.e., CAC ≥ 1), sensitivity and PPV was high across all datasets (ranges: 80-100% and 87-100%, respectively). For CAC ≥ 100 on routine non-gated chest CTs, which is the latest recommended threshold to initiate statin therapy, our model showed sensitivities of 71-94% and positive predictive values in the range of 88-100% across all the sites. Adoption of this model could allow more patients to be screened with CAC scoring, potentially allowing opportunistic early preventive interventions.

Cross-Modal Data Programming Enables Rapid Medical Machine Learning.

A major bottleneck in developing clinically impactful machine learning models is a lack of labeled training data for model supervision. Thus, medical researchers increasingly turn to weaker, noisier sources of supervision, such as leveraging extractions from unstructured text reports to supervise image classification. A key challenge in weak supervision is combining sources of information that may differ in quality and have correlated errors. Recently, a statistical theory of weak supervision called data programming has shown promise in addressing this challenge. Data programming now underpins many deployed machine-learning systems in the technology industry, even for critical applications. We propose a new technique for applying data programming to the problem of cross-modal weak supervision in medicine, wherein weak labels derived from an auxiliary modality (e.g., text) are used to train models over a different target modality (e.g., images). We evaluate our approach on diverse clinical tasks via direct comparison to institution-scale, hand-labeled datasets. We find that our supervision technique increases model performance by up to 6 points area under the receiver operating characteristic curve (ROC-AUC) over baseline methods by improving both coverage and quality of the weak labels. Our approach yields models that on average perform within 1.75 points ROC-AUC of those supervised with physician-years of hand labeling and outperform those supervised with physician-months of hand labeling by 10.25 points ROC-AUC, while using only person-days of developer time and clinician work-a time saving of 96%. Our results suggest that modern weak supervision techniques such as data programming may enable more rapid development and deployment of clinically useful machine-learning models.

Inferring Generative Model Structure with Static Analysis.

Obtaining enough labeled data to robustly train complex discriminative models is a major bottleneck in the machine learning pipeline. A popular solution is combining multiple sources of weak supervision using generative models. The structure of these models affects training label quality, but is difficult to learn without any ground truth labels. We instead rely on these weak supervision sources having some structure by virtue of being encoded programmatically. We present Coral, a paradigm that infers generative model structure by statically analyzing the code for these heuristics, thus reducing the data required to learn structure significantly. We prove that Coral's sample complexity scales quasilinearly with the number of heuristics and number of relations found, improving over the standard sample complexity, which is exponential in n for identifying nth degree relations. Experimentally, Coral matches or outperforms traditional structure learning approaches by up to 3.81 F1 points. Using Coral to model dependencies instead of assuming independence results in better performance than a fully supervised model by 3.07 accuracy points when heuristics are used to label radiology data without ground truth labels.