RATIONALE AND OBJECTIVES:Given the high volume of chest radiographs, radiologists frequently encounter heavy workloads. In outpatient imaging, a substantial portion of chest radiographs show no actionable findings. Automatically identifying these cases could improve efficiency by facilitating shorter reading workflows. PURPOSE:A large-scale study to assess the performance of AI on identifying chest radiographs with no actionable disease (NAD) in an outpatient imaging population using comprehensive, objective, and reproducible criteria for NAD. MATERIALS AND METHODS:The independent validation study includes 15000 patients with chest radiographs in posterior-anterior (PA) and lateral projections from an outpatient imaging center in the United States. Ground truth was established by reviewing CXR reports and classifying cases as NAD or actionable disease (AD). The NAD definition includes completely normal chest radiographs and radiographs with well-defined non-actionable findings. The AI NAD Analyzer1 (trained with 100 million multimodal images and fine-tuned on 1.3 million radiographs) utilizes a tandem system with image-level rule in and compartment-level rule out to provide case level output as NAD or potential actionable disease (PAD). RESULTS:A total of 14057 cases met our eligibility criteria (age 56 ± 16.1 years, 55% women and 45% men). The prevalence of NAD cases in the study population was 70.7%. The AI NAD Analyzer correctly classified NAD cases with a sensitivity of 29.1% and a yield of 20.6%. The specificity was 98.9% which corresponds to a miss rate of 0.3% of cases. Significant findings were missed in 0.06% of cases, while no cases with critical findings were missed by AI. CONCLUSION:In an outpatient population, AI can identify 20% of chest radiographs as NAD with a very low rate of missed findings. These cases could potentially be read using a streamlined protocol, thus improving efficiency and consequently reducing daily workload for radiologists.

Abstract Background: Chest radiographs are one of the most frequently performed imaging examinations in radiology. Chest radiograph reading is characterized by a high volume of cases, leading to long worklists. However, a substantial percentage of chest radiographs in outpatient imaging are without actionable findings. Identifying these cases could lead to numerous workflow efficiency improvements. Objective: To assess the performance of an AI system to identify chest radiographs with no actionable disease (NAD) in an outpatient imaging population in the United States. Materials and Methods: The study includes a random sample of 15,000 patients with chest radiographs in posterior-anterior (PA) and optional lateral projections from an outpatient imaging center with multiple locations in the Northeast United States. The ground truth was established by manually reviewing procedure reports and classifying cases as non-actionable disease (NAD) or actionable disease (AD) based on predetermined criteria. The NAD cases include both completely normal chest radiographs without any abnormal findings and radiographs with non-actionable findings. The AI NAD Analyzer1 trained on more than 1.3 million radiographs provides a binary case level output for the chest radiographs as either NAD or potential actionable disease (PAD). Two systems A (more specific) and B (more sensitive) were trained. Both systems were capable of processing either frontal only or frontal-lateral pair. Results: After excluding patients < 18 years (n=861) as well as the cases not meeting the image quality requirements of the AI NAD Analyzer (n=82), 14057 cases (average age 56±16.1 years, 7722 women and 6328 men) remained for the analysis. The AI NAD Analyzer with input consisting of PA and lateral images, correctly classified 2891 cases as NAD with concordance between ground truth and AI, which is 20.6% of all cases and 29.1% of all ground truth NAD cases. The miss rate was 0.3% and included 0.06% significant findings. With a more specific version of the AI NAD Analyzer (System A), there were 12.2% of all NAD cases were identified correctly with a miss rate of 0.1%. No cases with critical findings were missed by either system. Conclusion: The AI system can identify a meaningful number of chest radiographs with no actionable disease in an outpatient imaging population with a very low rate of missed findings. 1For research purposes only. Not for clinical use. Future commercial availability cannot be guaranteed.

Based on past experiences of the Center for Augmented Intelligence in Imaging (CAII) [Department of Radiology, Mayo Clinic Florida], depending on the project, 10 to 20 months has typically been required to realize the successful creation (data curation and algorithm development), and utilization (integration, testing, and operationalization) of an AI algorithm [[Figure 1][1]]. ![Figure 1:][2] Figure 1: AI algorithm evolution typically requires 10 to 20 months, consisting of four consecutive phases: 1. data identification and extraction; 2. data cleansing and labeling; 3. algorithm development with training and tuning; and 4. Implementation and integration with testing and operationalization. This manuscript delineates the related challenges and opportunities for greater efficiency in completing the clinical workflow implementation and integration of an AI algorithm. Strategies exploiting conventional data standards in facilitating the completion of such deployment and utilization goals within the operations of a busy Radiology practice are described. Methodologies and techniques employed during this initial phase of the CAII-Siemens D&A AI collaboration to address the previously mentioned challenges and opportunities are depicted with use-case examples. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This study did not receive any funding. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: IRB of Mayo Clinic waived ethical approval for this work. The following 2 IRB approvals were covering this project: 1. Mayo Clinic IRB ID: 20-006893 2. Mayo Clinic IRB ID: 22-011962 I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced in the present work are contained in the manuscript. [1]: #F1 [2]: pending:yes

Abstract Background Airspace disease as seen on chest X-rays is an important point in triage for patients initially presenting to the emergency department with suspected COVID-19 infection. The purpose of this study is to evaluate a previously trained interpretable deep learning algorithm for the diagnosis and prognosis of COVID-19 pneumonia from chest X-rays obtained in the ED. Methods This retrospective study included 2456 (50% RT-PCR positive for COVID-19) adult patients who received both a chest X-ray and SARS-CoV-2 RT-PCR test from January 2020 to March of 2021 in the emergency department at a single U.S. institution. A total of 2000 patients were included as an additional training cohort and 456 patients in the randomized internal holdout testing cohort for a previously trained Siemens AI-Radiology Companion deep learning convolutional neural network algorithm. Three cardiothoracic fellowship-trained radiologists systematically evaluated each chest X-ray and generated an airspace disease area-based severity score which was compared against the same score produced by artificial intelligence. The interobserver agreement, diagnostic accuracy, and predictive capability for inpatient outcomes were assessed. Principal statistical tests used in this study include both univariate and multivariate logistic regression. Results Overall ICC was 0.820 (95% CI 0.790–0.840). The diagnostic AUC for SARS-CoV-2 RT-PCR positivity was 0.890 (95% CI 0.861–0.920) for the neural network and 0.936 (95% CI 0.918–0.960) for radiologists. Airspace opacities score by AI alone predicted ICU admission (AUC = 0.870) and mortality (0.829) in all patients. Addition of age and BMI into a multivariate log model improved mortality prediction (AUC = 0.906). Conclusion The deep learning algorithm provides an accurate and interpretable assessment of the disease burden in COVID-19 pneumonia on chest radiographs. The reported severity scores correlate with expert assessment and accurately predicts important clinical outcomes. The algorithm contributes additional prognostic information not currently incorporated into patient management.

To perform a multicenter assessment of the CT Pneumonia Analysis prototype for predicting disease severity and patient outcome in COVID-19 pneumonia both without and with integration of clinical information. Our IRB-approved observational study included consecutive 241 adult patients (> 18 years; 105 females; 136 males) with RT-PCR-positive COVID-19 pneumonia who underwent non-contrast chest CT at one of the two tertiary care hospitals (site A: Massachusetts General Hospital, USA; site B: Firoozgar Hospital Iran). We recorded patient age, gender, comorbid conditions, laboratory values, intensive care unit (ICU) admission, mechanical ventilation, and final outcome (recovery or death). Two thoracic radiologists reviewed all chest CTs to record type, extent of pulmonary opacities based on the percentage of lobe involved, and severity of respiratory motion artifacts. Thin-section CT images were processed with the prototype (Siemens Healthineers) to obtain quantitative features including lung volumes, volume and percentage of all-type and high-attenuation opacities (≥ −200 HU), and mean HU and standard deviation of opacities within a given lung region. These values are estimated for the total combined lung volume, and separately for each lung and each lung lobe. Multivariable analyses of variance (MANOVA) and multiple logistic regression were performed for data analyses. About 26% of chest CTs (62/241) had moderate to severe motion artifacts. There were no significant differences in the AUCs of quantitative features for predicting disease severity with and without motion artifacts (AUC 0.94–0.97) as well as for predicting patient outcome (AUC 0.7–0.77) (p > 0.5). Combination of the volume of all-attenuation opacities and the percentage of high-attenuation opacities (AUC 0.76–0.82, 95% confidence interval (CI) 0.73–0.82) had higher AUC for predicting ICU admission than the subjective severity scores (AUC 0.69–0.77, 95% CI 0.69–0.81). Despite a high frequency of motion artifacts, quantitative features of pulmonary opacities from chest CT can help differentiate patients with favorable and adverse outcomes.

Purpose: Comparison of deep learning algorithm, radiomics and subjective assessment of chest CT for predicting outcome (death or recovery) and intensive care unit (ICU) admission in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Methods: The multicenter, ethical committee-approved, retrospective study included non-contrast-enhanced chest CT of 221 SARS-CoV-2 positive patients from Italy (n = 196 patients; mean age 64 +/- 16 years) and Denmark (n = 25; mean age 69 +/- 13 years). A thoracic radiologist graded presence, type and extent of pulmonary opacities and severity of motion artifacts in each lung lobe on all chest CTs. Thin-section CT images were processed with CT Pneumonia Analysis Prototype (Siemens Healthineers) which yielded segmentation masks from a deep learning (DL) algorithm to derive features of lung abnormalities such as opacity scores, mean HU, as well as volume and percentage of all-attenuation and high-attenuation (opacities >-200 HU) opacities. Separately, whole lung radiomics were obtained for all CT exams. Analysis of variance and multiple logistic regression were performed for data analysis. Results: Moderate to severe respiratory motion artifacts affected nearly one-quarter of chest CTs in patients. Subjective severity assessment, DL-based features and radiomics predicted patient outcome (AUC 0.76 vs AUC 0.88 vs AUC 0.83) and need for ICU admission (AUC 0.77 vs AUC 0.0.80 vs 0.82). Excluding chest CT with motion artifacts, the performance of DL-based and radiomics features improve for predicting ICU admission. Conclusion: DL-based and radiomics features of pulmonary opacities from chest CT were superior to subjective assessment for differentiating patients with favorable and adverse outcomes.

Introduction/Hypothesis: To validate an AI algorithm for rapid detection of COVID-19 pulmonary complications and prediction of negative 30-day outcomes in COVID-suspicious patients. Methods: We included 2000 chest X-rays (CXR) from patients who had a COVID-19 RT-PCR test within 14 days of the CXR. A deep learning CNN (AI-RAD Companion, Siemens) previously trained for non-COVID pneumonia was used to analyze the CXRs. A total of 1544 CXRs were first used to train the AI with COVID cases. Then, a randomized modified internal holdout of 456 patients (236 positive, 220 negative) were used as test cohort. AI results detect the presence of COVID-19 lung disease (CLD) and also report a 1 to 10 AI severity score. Positive RT-PCR within 14 days of the CXR was used as the ground-truth for COVID diagnosis. Radiologic assessment by three cardiothoracic radiologists was used to detect the presence of CLD and generate a 1 to 10 expert severity score. All-cause mortality within 30 days of the CXR was recorded. Receiver-operating characteristic (ROC) curve analysis was performed and the area under the curve (AUC) was reported. Concordance metrics included intraclass correlation coefficient (ICC) for comparison of AI and expert results. Results: In COVID+ patients, AI demonstrated a sensitivity of 99% (205/207) , specificity of 62% (18/29), PPV of 95% (205/216), and NPV of 90% (18/20), for the detection of CLD. Amongst COVID+ patients, the AI severity score had excellent agreement with the expert severity score for lung involvement (ICC=0.89, 95% CI 0.86-0.92). There were 70 deaths in the test cohort (15.3%). The AI severity score had an excellent ability to predict all-cause mortality (AUC=0.832 vs expert AUC=0.844, p >0.05). Conclusions: This CXR AI tool had an excellent sensitivity for detection of COVID-19 lung disease in PCR-positive patients and excellent correlation with expert analysis. AI severity score was able to strongly predict 30-day patient all-cause mortality.