Automated CT biomarkers for opportunistic prediction of future cardiovascular events and mortality in an asymptomatic screening population: a retrospective cohort study

Value-added Opportunistic CT Screening: State of the Art

Opportunistic CT screening leverages robust imaging data embedded within abdominal and thoracic scans that are generally unrelated to the specific clinical indication and have heretofore gone largely unused. This incidental imaging information may prove beneficial to patients in terms of wellness, prevention, risk profiling, and presymptomatic detection of relevant disease. The growing interest in CT-based opportunistic screening relates to a confluence of factors: the objective and generalizable nature of CT-based body composition measures, the emergence of fully automated explainable AI solutions, the sheer volume of body CT scans performed, and the increasing emphasis on precision medicine and value-added initiatives. With a systematic approach to body composition and other useful CT markers, initial evidence suggests that their ability to help radiologists assess biologic age and predict future adverse cardiometabolic events rivals even the best available clinical reference standards. Emerging data suggest that standalone "intended" CT screening over an unorganized opportunistic approach may be justified, especially when combined with established cancer screening. This review will discuss the current status of opportunistic CT screening, including specific body composition markers and the various disease processes that may be impacted. The remaining hurdles to widespread clinical adoption include generalization to more diverse patient populations, disparate technical settings, and reimbursement.

Incidence of pancreatic cancer during long-term follow-up in patients with incidental pancreatic cysts smaller than 2 cm

PURPOSE

To assess the long-term malignancy risk of incidental small pancreatic cysts.

MATERIALS AND METHODS

In this HIPAA-compliant, IRB-approved, retrospective, multi-institutional study, the long-term incidence of pancreatic cancer was compared between patients with and without small pancreatic cysts. Patients with incidental pancreatic cysts ≥ 0.5 and < 2.0 cm in maximal diameter, detected on MRI performed between 1999 and 2011, represented the "small pancreatic cyst" group. Patients that underwent MRI between 2005 and 2011 and had no reported pancreatic cysts represented the comparison "no cyst" group.

RESULTS

The "small pancreatic cyst" group included 267 patients, ages 63.4 ± 11.8 years, 166/267 (62%) women with a mean follow-up of 8.6 ± 4.3 years, median 9.2 years; the "no cyst" group included 1,459 patients, ages 64.6 ± 12 years, 794/1,459 (54%) women with a mean follow-up of 7.0 ± 4.2 years, median 7.8 (p values 0.12, 0.02, < 0.001, respectively). Two/267 (0.7%) patients developed pancreatic cancer at a separate location from the known cyst in the "small pancreatic cyst" group, with a cancer rate of 0.9 (95% CI 0.1-3.1) cases per 1,000 patient-years. In the "no cyst" cohort, 18/1,459 (1.2%) patients developed pancreatic cancer, with a cancer rate of 1.8 (95% CI 1.2-3.1) cases per 1,000 patient-years (p = 0.6). The all-cause mortality was similar in both groups: 57/267 (21%) vs. 384/1,459 (26%) (p = 0.09).

CONCLUSION

The long-term risk of pancreatic malignancy in asymptomatic patients with incidental pancreatic cysts less than 2 cm is 0.9 cases per 1,000 patient-years of follow-up, similar to those without pancreatic cysts. These very few pancreatic cancers developed at a separate location from the known cyst.

KEY POINTS

• After a median of 9.2 years of follow-up, the risk of pancreatic malignancy in patients with an asymptomatic small pancreatic cyst was 0.9 cases per 1,000 patient-years of follow-up, similar to those without pancreatic cysts. • Very few pancreatic cancer cases developed in the location separate from the known pancreatic cyst.

Novel Artificial Intelligence Applications in Cardiology: Current Landscape, Limitations, and the Road to Real-World Applications

Cardiovascular diseases are the leading cause of death globally and contribute significantly to the cost of healthcare. Artificial intelligence (AI) is poised to reshape cardiology. Using supervised and unsupervised learning, the two main branches of AI, several applications have been developed in recent years to improve risk prediction, allow large-scale analysis of medical data, and phenotype patients for personalized medicine. In this review, we examine the key advances in AI in cardiology and its limitations regarding bias in the data, standardization in reporting, data access, and model trust and accountability in cases of error. Finally, we discuss implementation methods to unleash AI's potential in making healthcare more accurate and efficient. Several steps need to be followed and challenges overcome in order to successfully integrate AI in clinical practice and ensure its longevity.

Fully Automated Abdominal CT Biomarkers for Type 2 Diabetes Using Deep Learning

Background CT biomarkers both inside and outside the pancreas can potentially be used to diagnose type 2 diabetes mellitus. Previous studies on this topic have shown significant results but were limited by manual methods and small study samples. Purpose To investigate abdominal CT biomarkers for type 2 diabetes mellitus in a large clinical data set using fully automated deep learning. Materials and Methods For external validation, noncontrast abdominal CT images were retrospectively collected from consecutive patients who underwent routine colorectal cancer screening with CT colonography from 2004 to 2016. The pancreas was segmented using a deep learning method that outputs measurements of interest, including CT attenuation, volume, fat content, and pancreas fractal dimension. Additional biomarkers assessed included visceral fat, atherosclerotic plaque, liver and muscle CT attenuation, and muscle volume. Univariable and multivariable analyses were performed, separating patients into groups based on time between type 2 diabetes diagnosis and CT date and including clinical factors such as sex, age, body mass index (BMI), BMI greater than 30 kg/m², and height. The best set of predictors for type 2 diabetes were determined using multinomial logistic regression. Results A total of 8992 patients (mean age, 57 years ± 8 [SD]; 5009 women) were evaluated in the test set, of whom 572 had type 2 diabetes mellitus. The deep learning model had a mean Dice similarity coefficient for the pancreas of 0.69 ± 0.17, similar to the interobserver Dice similarity coefficient of 0.69 ± 0.09 (P = .92). The univariable analysis showed that patients with diabetes had, on average, lower pancreatic CT attenuation (mean, 18.74 HU ± 16.54 vs 29.99 HU ± 13.41; P < .0001) and greater visceral fat volume (mean, 235.0 mL ± 108.6 vs 130.9 mL ± 96.3; P < .0001) than those without diabetes. Patients with diabetes also showed a progressive decrease in pancreatic attenuation with greater duration of disease. The final multivariable model showed pairwise areas under the receiver operating characteristic curve (AUCs) of 0.81 and 0.85 between patients without and patients with diabetes who were diagnosed 0-2499 days before and after undergoing CT, respectively. In the multivariable analysis, adding clinical data did not improve upon CT-based AUC performance (AUC = 0.67 for the CT-only model vs 0.68 for the CT and clinical model). The best predictors of type 2 diabetes mellitus included intrapancreatic fat percentage, pancreatic fractal dimension, plaque severity between the L1 and L4 vertebra levels, average liver CT attenuation, and BMI. Conclusion The diagnosis of type 2 diabetes mellitus was associated with abdominal CT biomarkers, especially measures of pancreatic CT attenuation and visceral fat. © RSNA, 2022 Online supplemental material is available for this article.

A deep learning system for automated kidney stone detection and volumetric segmentation on noncontrast CT scans

PURPOSE

Early detection and size quantification of renal calculi are important for optimizing treatment and preventing severe kidney stone disease. Prior work has shown that volumetric measurements of kidney stones are more informative and reproducible than linear measurements. Deep learning-based systems that use abdominal noncontrast computed tomography (CT) scans may assist in detection and reduce workload by removing the need for manual stone volume measurement. Prior to this work, no such system had been developed for use on noisy low-dose CT or tested on a large-scale external dataset.

METHODS

We used a dataset of 91 CT colonography (CTC) scans with manually marked kidney stones combined with 89 CTC scans without kidney stones. To compare with a prior work half the data was used for training and half for testing. A set of CTC scans from 6185 patients from a separate institution with patient-level labels were used as an external validation set. A 3D U-Net model was employed to segment the kidneys, followed by gradient-based anisotropic denoising, thresholding, and region growing. A 13 layer convolutional neural network classifier was then applied to distinguish kidney stones from false positive regions.

RESULTS

The system achieved a sensitivity of 0.86 at 0.5 false positives per scan on a challenging test set of low-dose CT with many small stones, an improvement over an earlier work that obtained a sensitivity of 0.52. The stone volume measurements correlated well with manual measurements (r 2 = 0.95 $r^2 = 0.95$ ). For patient-level classification, the system achieved an area under the receiver-operating characteristic of 0.95 on an external validation set (sensitivity = 0.88, specificity = 0.91 at the Youden point). A common cause of false positives were small atherosclerotic plaques in the renal sinus that simulated kidney stones.

CONCLUSIONS

Our deep-learning-based system showed improvements over a previously developed system that did not use deep learning, with even higher performance on an external validation set.

An overview of artificial intelligence in oncology

Cancer is associated with significant morbimortality globally. Advances in screening, diagnosis, management and survivorship were substantial in the last decades, however, challenges in providing personalized and data-oriented care remain. Artificial intelligence (AI), a branch of computer science used for predictions and automation, has emerged as potential solution to improve the healthcare journey and to promote precision in healthcare. AI applications in oncology include, but are not limited to, optimization of cancer research, improvement of clinical practice (eg., prediction of the association of multiple parameters and outcomes - prognosis and response) and better understanding of tumor molecular biology. In this review, we examine the current state of AI in oncology, including fundamentals, current applications, limitations and future perspectives.

Computed Tomography Assessment of Sarcopenic Myosteatosis for Predicting Overall Survival in Colorectal Carcinoma: Systematic Review

BACKGROUND

As the US population ages, cancer incidence and prevalence are projected to increase. In the last decade, there has been an increased interest in the opportunistic use of computed tomography (CT) scan data to predict cancer prognosis and inform treatment based on body composition measures, especially muscle measures for sarcopenia.

OBJECTIVE

This article aimed to perform a systematic review of current literature related to CT assessment of muscle attenuation values for myosteatosis in colorectal cancer (CRC) survival prediction.

RESULTS

Initial broad search of CT and CRC yielded 4234 results. A more focused search strategy narrowed this to 129 research papers, and 13 articles met the final inclusion criteria. Twelve of 13 studies found a statistically significant decrease in overall survival according to Hounsfield unit (HU)-based sarcopenia, with hazard ratios ranging from 1.36 to 2.94 (mean, 1.78). However, the specific criteria used to define myosteatosis by CT varied widely, with attenuation thresholds ranging from 22.5 to 47.3 HU, often further subdivided by sex and/or body mass index.

CONCLUSIONS

Current evidence suggests that a strong association between CT-based muscle attenuation values for myosteatosis assessment correlates with overall survival in CRC. However, more research is needed to verify these findings and determine appropriate threshold values for more diverse patient populations. Because CRC patients are staged and followed by CT, the opportunity exists for routine objective myosteatosis assessment in the clinical setting.

Epicardial and pericardial fat analysis on CT images and artificial intelligence: a literature review

The present review summarizes the available evidence on artificial intelligence (AI) algorithms aimed to the segmentation of epicardial and pericardial adipose tissues on computed tomography (CT) images. Body composition imaging is a novel concept based on quantitative analysis of body tissues. Manual segmentation of medical images allows to obtain quantitative and qualitative data on several tissues including epicardial and pericardial fat. However, since manual segmentation requires a considerable amount of time, the analysis of adipose tissue compartments based on AI has been proposed as an automatic, reliable, accurate and fast tool. The literature research was performed on March 2021 using MEDLINE PubMed Central and "adipose tissue artificial intelligence", "adipose tissue deep learning" or "adipose tissue machine learning" as keywords for articles search. Relevant articles concerning epicardial adipose tissue, pericardial adipose tissue and AI were selected. The evaluation of adipose tissue compartments can provide additional information on the pathogenesis and prognosis of several diseases, including cardiovascular. AI can assist physicians to obtain important information, possibly improving the patient's quality of life and identifying patients at risk of developing variable disorders.

A Fully Automated Deep Learning Pipeline for Multi-Vertebral Level Quantification and Characterization of Muscle and Adipose Tissue on Chest CT Scans

Body composition on chest CT scans encompasses a set of important imaging biomarkers. This study developed and validated a fully automated analysis pipeline for multi-vertebral level assessment of muscle and adipose tissue on routine chest CT scans. This study retrospectively trained two convolutional neural networks on 629 chest CT scans from 629 patients (55% women; mean age, 67 years ± 10 [standard deviation]) obtained between 2014 and 2017 prior to lobectomy for primary lung cancer at three institutions. A slice-selection network was developed to identify an axial image at the level of the fifth, eighth, and 10th thoracic vertebral bodies. A segmentation network (U-Net) was trained to segment muscle and adipose tissue on an axial image. Radiologist-guided manual-level selection and segmentation generated ground truth. The authors then assessed the predictive performance of their approach for cross-sectional area (CSA) (in centimeters squared) and attenuation (in Hounsfield units) on an independent test set. For the pipeline, median absolute error and intraclass correlation coefficients for both tissues were 3.6% (interquartile range, 1.3%-7.0%) and 0.959-0.998 for the CSA and 1.0 HU (interquartile range, 0.0-2.0 HU) and 0.95-0.99 for median attenuation. This study demonstrates accurate and reliable fully automated multi-vertebral level quantification and characterization of muscle and adipose tissue on routine chest CT scans. Keywords: Skeletal Muscle, Adipose Tissue, CT, Chest, Body Composition Analysis, Convolutional Neural Network (CNN), Supervised Learning Supplemental material is available for this article. © RSNA, 2022.

Percentile-based averaging and skeletal muscle gauge improve body composition analysis: validation at multiple vertebral levels

BACKGROUND

Skeletal muscle metrics on computed tomography (CT) correlate with clinical and patient-reported outcomes. We hypothesize that aggregating skeletal muscle measurements from multiple vertebral levels and skeletal muscle gauge (SMG) better predict outcomes than skeletal muscle radioattenuation (SMRA) or -index (SMI) at a single vertebral level.

METHODS

We performed a secondary analysis of prospectively collected clinical (overall survival, hospital readmission, time to unplanned hospital readmission or death, and readmission or death within 90 days) and patient-reported outcomes (physical and psychological symptom burden captured as Edmonton Symptom Assessment Scale and Patient Health Questionnaire) of patients with advanced cancer who experienced an unplanned admission to Massachusetts General Hospital from 2014 to 2016. First, we assessed the correlation of skeletal muscle cross-sectional area, SMRA, SMI, and SMG at one or more of the following thoracic (T) or lumbar (L) vertebral levels: T5, T8, T10, and L3 on CT scans obtained ≤50 days before index assessment. Second, we aggregated measurements across all available vertebral levels using percentile-based averaging (PBA) to create the average percentile. Third, we constructed one regression model adjusted for age, sex, sociodemographic factors, cancer type, body mass index, and intravenous contrast for each combination of (i) vertebral level and average percentile, (ii) muscle metrics (SMRA, SMI, & SMG), and (iii) clinical and patient-reported outcomes. Fourth, we compared the performance of vertebral levels and muscle metrics by ranking otherwise identical models by concordance statistic, number of included patients, coefficient of determination, and significance of muscle metric.

RESULTS

We included 846 patients (mean age: 63.5 ± 12.9 years, 50.5% males) with advanced cancer [predominantly gastrointestinal (32.9%) or lung (18.9%)]. The correlation of muscle measurements between vertebral levels ranged from 0.71 to 0.84 for SMRA and 0.67 to 0.81 for SMI. The correlation of individual levels with the average percentile was 0.90-0.93 for SMRA and 0.86-0.92 for SMI. The intrapatient correlation of SMRA with SMI was 0.21-0.40. PBA allowed for inclusion of 8-47% more patients than any single-level analysis. PBA outperformed single-level analyses across all comparisons with average ranks 2.6, 2.9, and 1.6 for concordance statistic, coefficient of determination, and significance (range 1-5, μ = 3), respectively. On average, SMG outperformed SMRA and SMI across outcomes and vertebral levels: the average rank of SMG was 1.4, 1.4, and 1.4 for concordance statistic, coefficient of determination, and significance (range 1-3, μ = 2), respectively.

CONCLUSIONS

Multivertebral level skeletal muscle analyses using PBA and SMG independently and additively outperform analyses using individual levels and SMRA or SMI.

Proposed diagnostic volumetric bone mineral density thresholds for osteoporosis and osteopenia at the cervicothoracic spine in correlation to the lumbar spine

OBJECTIVES

To determine the correlation between cervicothoracic and lumbar volumetric bone mineral density (vBMD) in an average cohort of adults and to identify specific diagnostic thresholds for the cervicothoracic spine on the individual subject level.

METHODS

In this HIPPA-compliant study, we retrospectively included 260 patients (59.7 ± 18.3 years, 105 women), who received a contrast-enhanced or non-contrast-enhanced CT scan. vBMD was extracted using an automated pipeline ( https://anduin.bonescreen.de ). The association of vBMD between each vertebra spanning C2-T12 and the averaged values at the lumbar spine (L1-L3) was analyzed before and after semiquantitative assessment of fracture status and degeneration, and respective vertebra-specific cut-off values for osteoporosis were calculated using linear regression.

RESULTS

In both women and men, trabecular vBMD decreased with age in the cervical, thoracic, and lumbar regions. vBMD values of cervicothoracic vertebrae showed strong correlations with lumbar vertebrae (L1-L3), with a median Pearson value of r = 0.87 (range: r_C2 = 0.76 to r_T12 = 0.96). The correlation coefficients were significantly lower (p < 0.0001) without excluding fractured and degenerated vertebrae, median r = 0.82 (range: r_C2 = 0.69 to r_T12 = 0.93). Respective cut-off values for osteoporosis peaked at C4 (209.2 mg/ml) and decreased to 83.8 mg/ml at T12.

CONCLUSION

Our data show a high correlation between clinically used mean L1-L3 values and vBMD values elsewhere in the spine, independent of age. The proposed cut-off values for the cervicothoracic spine therefore may allow the determination of low bone mass even in clinical cases where only parts of the spine are imaged.

KEY POINTS

vBMD of all cervicothoracic vertebrae showed strong correlation with lumbar vertebrae (L1-L3), with a median Pearson's correlation coefficient of r = 0.87 (range: r_C2 = 0.76 to r_T12 = 0.96). The correlation coefficients were significantly lower (p < 0.0001) without excluding fractured and moderate to severely degenerated vertebrae, median r = 0.82 (range: r_C2 = 0.69 to r_T12 = 0.93). We postulate that trabecular vBMD < 200 mg/ml for the cervical spine and < 100 mg/ml for the thoracic spine are strong indicators of osteoporosis, similar to < 80 mg/ml at the lumbar spine.

RSNA-MICCAI Panel Discussion: 2. Leveraging the Full Potential of AI-Radiologists and Data Scientists Working Together

In March 2021, the Radiological Society of North America hosted a virtual panel discussion with members of the Medical Image Computing and Computer Assisted Intervention Society. Both organizations share a vision to develop radiologic and medical imaging techniques through advanced quantitative imaging biomarkers and artificial intelligence. The panel addressed how radiologists and data scientists can collaborate to advance the science of AI in radiology. Keywords: Adults and Pediatrics, Segmentation, Feature Detection, Quantification, Diagnosis/Classification, Prognosis/Classification © RSNA, 2021.

Metabolic Regulation of Inflammation and Its Resolution: Current Status, Clinical Needs, Challenges, and Opportunities

Metabolism and inflammation have been viewed as two separate processes with distinct but critical functions for our survival: metabolism regulates the utilization of nutrients, and inflammation is responsible for defense and repair. Both respond to an organism's stressors to restore homeostasis. The interplay between metabolic status and immune response (immunometabolism) plays an important role in maintaining health or promoting disease development. Understanding these interactions is critical in developing tools for facilitating novel preventative and therapeutic approaches for diseases, including cancer. This trans-National Institutes of Health workshop brought together basic scientists, technology developers, and clinicians to discuss state-of-the-art, innovative approaches, challenges, and opportunities to understand and harness immunometabolism in modulating inflammation and its resolution.

Different prognostic impact of glucose uptake in visceral adipose tissue according to sex in patients with colorectal cancer

The purpose of this study was to investigate whether sex differences in visceral fat volume and glucose uptake measured by positron emission tomography/computed tomography (PET/CT) in abdominal visceral fat can stratify overall survival (OS) in patients with colorectal cancer (CRC). We retrospectively enrolled 293 patients diagnosed with CRC who underwent PET/CT before surgical resection. Fluorodeoxyglucose uptake of visceral adipose tissue (VAT-SUV) and subcutaneous adiposity tissue (SAT-SUV) were measured using PET/CT. The relative VAT (rVAT) was defined as the visceral fat volume normalized to the total volume of fat (VAT plus SAT). We defined sex-specific cutoff values for VAT-SUV, SAT-SUV, and rVAT. Univariate and multivariate analyses using Cox proportional hazard regression analysis were performed to identify the independent prognostic factors. The study population comprised 181 men and 112 women. The rVAT (0.40 vs. 0.29, p < 0.001) and VAT-SUV (0.55 vs. 0.48, p = 0.007) were significantly greater in men than in women. High rVAT (than low rVAT) and high VAT-SUV (than low VAT-SUV) showed a worse prognosis in male and female patients, respectively. Multivariate analysis indicated that the combination of rVAT and VAT-SUV was an independent prognostic factor for predicting OS in both male and female patients. The combination of rVAT and VAT-SUV could differentiate the patients with the best survival outcome from the other three individual groups in female patients, but not in males. Glucose uptake and relative volume of visceral fat may provide a new risk stratification for patients with CRC, especially female patients.

Atherosclerotic Plaque Burden on Abdominal CT: Automated Assessment With Deep Learning on Noncontrast and Contrast-enhanced Scans

BACKGROUND

Abdominal aortic atherosclerotic plaque burden may have clinical significance but manual measurement is time-consuming and impractical.

PURPOSE

To perform external validation on an automated atherosclerotic plaque detector for noncontrast and postcontrast abdominal CT.

MATERIALS AND METHODS

The training data consisted of 114 noncontrast CT scans and 23 postcontrast CT urography scans. The testing data set consisted of 922 CT colonography (CTC) scans, and 1207 paired noncontrast and postcontrast CT scans from renal donors from a second institution. Reference standard data included manual plaque segmentations in the 137 training scans and manual plaque burden measurements in the 922 CTC scans. The total Agatston score and group (0-3) was determined using fully-automated deep learning software. Performance was assessed by measures of agreement, linear regression, and paired evaluations.

RESULTS

On CTC scans, automated Agatston scoring correlated highly with manual assessment (R² = 0.94). On paired renal donor CT scans, automated Agatston scoring on postcontrast CT correlated highly with noncontrast CT (R² = 0.95). When plaque burden was expressed as a group score, there was excellent agreement for both the CTC (weighted kappa 0.80 ± 0.01 [95% confidence interval: 0.78-0.83]) and renal donor (0.83 ± 0.02 [0.79-0.86]) assessments.

CONCLUSION

Fully automated detection, segmentation, and scoring of abdominal aortic atherosclerotic plaques on both pre- and post-contrast CT was validated and may have application for population-based studies.

Quantification of Liver Fat Content with CT and MRI: State of the Art

Hepatic steatosis is defined as pathologically elevated liver fat content and has many underlying causes. Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease worldwide, with an increasing prevalence among adults and children. Abnormal liver fat accumulation has serious consequences, including cirrhosis, liver failure, and hepatocellular carcinoma. In addition, hepatic steatosis is increasingly recognized as an independent risk factor for the metabolic syndrome, type 2 diabetes, and, most important, cardiovascular mortality. During the past 2 decades, noninvasive imaging-based methods for the evaluation of hepatic steatosis have been developed and disseminated. Chemical shift-encoded MRI is now established as the most accurate and precise method for liver fat quantification. CT is important for the detection and quantification of incidental steatosis and may play an increasingly prominent role in risk stratification, particularly with the emergence of CT-based screening and artificial intelligence. Quantitative imaging methods are increasingly used for diagnostic work-up and management of steatosis, including treatment monitoring. The purpose of this state-of-the-art review is to provide an overview of recent progress and current state of the art for liver fat quantification using CT and MRI, as well as important practical considerations related to clinical implementation.

Clinical Artificial Intelligence Applications in Radiology: Chest and Abdomen

Organ segmentation, chest radiograph classification, and lung and liver nodule detections are some of the popular artificial intelligence (AI) tasks in chest and abdominal radiology due to the wide availability of public datasets. AI algorithms have achieved performance comparable to humans in less time for several organ segmentation tasks, and some lesion detection and classification tasks. This article introduces the current published articles of AI applied to chest and abdominal radiology, including organ segmentation, lesion detection, classification, and predicting prognosis.

Automated detection of the contrast phase in MDCT by an artificial neural network improves the accuracy of opportunistic bone mineral density measurements

Objectives

To determine the accuracy of an artificial neural network (ANN) for fully automated detection of the presence and phase of iodinated contrast agent in routine abdominal multidetector computed tomography (MDCT) scans and evaluate the effect of contrast correction for osteoporosis screening.

Methods

This HIPPA-compliant study retrospectively included 579 MDCT scans in 193 patients (62.4 ± 14.6 years, 48 women). Three different ANN models (2D DenseNet with random slice selection, 2D DenseNet with anatomy-guided slice selection, 3D DenseNet) were trained in 462 MDCT scans of 154 patients (threefold cross-validation), who underwent triphasic CT. All ANN models were tested in 117 unseen triphasic scans of 39 patients, as well as in a public MDCT dataset containing 311 patients. In the triphasic test scans, trabecular volumetric bone mineral density (BMD) was calculated using a fully automated pipeline. Root-mean-square errors (RMSE) of BMD measurements with and without correction for contrast application were calculated in comparison to nonenhanced (NE) scans.

Results

The 2D DenseNet with anatomy-guided slice selection outperformed the competing models and achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set (public dataset: F1 score 0.93; accuracy 94.2%). Application of contrast agent resulted in significant BMD biases (all p < .001; portal-venous (PV): RMSE 18.7 mg/ml, mean difference 17.5 mg/ml; arterial (AR): RMSE 6.92 mg/ml, mean difference 5.68 mg/ml). After the fully automated correction, this bias was no longer significant (p > .05; PV: RMSE 9.45 mg/ml, mean difference 1.28 mg/ml; AR: RMSE 3.98 mg/ml, mean difference 0.94 mg/ml).

Conclusion

Automatic detection of the contrast phase in multicenter CT data was achieved with high accuracy, minimizing the contrast-induced error in BMD measurements.

Key Points

• A 2D DenseNet with anatomy-guided slice selection achieved an F1 score of 0.98 and an accuracy of 98.3% in the test set. In a public dataset, an F1 score of 0.93 and an accuracy of 94.2% were obtained.

• Automated adjustment for contrast injection improved the accuracy of lumbar bone mineral density measurements (RMSE 18.7 mg/ml vs. 9.45 mg/ml respectively, in the portal-venous phase).

• An artificial neural network can reliably reveal the presence and phase of iodinated contrast agent in multidetector CT scans (https://github.com/ferchonavarro/anatomy_guided_contrast_c). This allows minimizing the contrast-induced error in opportunistic bone mineral density measurements.

Segmentation and characterization of visceral and abdominal subcutaneous adipose tissue on CT with and without contrast medium: influence of 2D- and 3D-segmentation

Background

Adipose tissue is a valuable biomarker. Although validation and correlation to clinical data have mostly been performed on non-enhanced scans (NES), a previous study has shown conversion of values of contrast enhanced scan (CES) into those of NES to be feasible with segmentation of the entire abdomen (3D-segmentation). In this study we analyzed if density and area of abdominal adipose tissue segmented in a single slice (2D-segmentation) of CES may be converted into that of NES. Furthermore, we compared the precision of conversion between 2D- and 3D-segmentation.

Methods

Thirty-one multi-phasic abdominal CT examinations at identical scan settings were retrospectively included. Exams included NES (n=31), arterial (ART) (n=23), portal-venous (PVN) (n=10), and/or venous scan (VEN) (n=31). Density and area of visceral (VAT) and subcutaneous adipose tissue (SAT) were quantified semi-automatically with fixed thresholds. For conversion of values from CES into those of NES regression analyses were performed and tested. 2D- and 3D-segmentation were compared with respect to conversion accuracy (normalized deviations of converted NES values from original measurements).

Results

After the application of contrast medium 2D-segmented adipose tissue increased in density (max. +5.6±2.4 HU) and decreased in area (max. -10.91%) (10.47%), with few exceptions (P<0.05). This was more pronounced in later scans (VEN ≈ PVN > ART) and more marked in VAT than SAT. Density and area in CES correlated very well with NES, allowing for conversion with only small error. While converted density is slightly more precise applying 3D-segmentation, conversion error of quantity was occasionally smaller with 2D-segmentation.

Conclusions

Contrast medium changes density and quantity of segmented adipose tissue in differing degrees between compartments, contrast phases and 2D- and 3D-segmentation. However, changes are fairly constant for a given compartment, contrast phase and mode of segmentation. Therefore, conversion of values into those of NES may be achieved with comparable precision for 2D- and 3D-segmentation.

Impact of Image Reconstruction Parameters on Abdominal Aortic Calcification Measurement Using Abdominal Computed Tomography

BACKGROUND AND AIMS

Abdominal aortic calcification (AAC) is correlated with cardiovascular outcomes independent of traditional risk factors. Quantification of AAC on computed tomography (CT) has not been standardized. Reconstruction parameters have been shown to impact coronary calcium scores. The aim of our study was to assess the impact of abdominal CT reconstruction parameters, slice thickness (ST), and display field of view (DFOV) on AAC quantitative scoring on abdominal CT examinations.

METHODS

We retrospectively measured AAC on noncontrast CT of 46 patients (mean age, 64.1 years; 35 males) using 5 different reconstruction protocols with a range of ST and DFOV: protocol A, 2.5 mm ST, 35 cm DFOV; protocol B, 2.5 mm ST, 50 cm DFOV; protocol C, 2.5 mm ST, 25 cm DFOV; protocol D, 5 mm ST, 35 cm DFOV; and protocol E: 0.625 mm ST, 35 cm DFOV. The AAC scores from each protocol were compared using concordance correlation coefficient and Bland-Altman agreement analyses.

RESULTS

The AAC mean (SD) scores for each protocol were as follows: A, 2022 (2418); B, 2022 (2412); C, 1939 (2310); D, 2220 (2695); and E, 1862 (2234). The AAC mean score differences between protocols and reference protocol A were -0.47, 82.01, -198.94, and 160 for protocols B, C, D, and E, respectively, with differences between protocols C to E statistically significantly different (P < 0.05). The different protocols showed overall excellent correlation (concordance correlation coefficient, >0.9) between AAC scores.

CONCLUSIONS

Slice thickness and DFOV can impact AAC score measurement. A description of reconstruction parameters is important to allow comparisons across different cohorts.

Fully Automated Deep Learning Tool for Sarcopenia Assessment on CT: L1 Versus L3 Vertebral Level Muscle Measurements for Opportunistic Prediction of Adverse Clinical Outcomes

BACKGROUND. Sarcopenia is associated with adverse clinical outcomes. CT-based skeletal muscle measurements for sarcopenia assessment are most commonly performed at the L3 vertebral level. OBJECTIVE. The purpose of this article is to compare the utility of fully automated deep learning CT-based muscle quantitation at the L1 versus L3 level for predicting future hip fractures and death. METHODS. This retrospective study included 9223 asymptomatic adults (mean age, 57 ± 8 [SD] years; 4071 men, 5152 women) who underwent unenhanced low-dose abdominal CT. A previously validated fully automated deep learning tool was used to assess muscle for myosteatosis (by mean attenuation) and myopenia (by cross-sectional area) at the L1 and L3 levels. Performance for predicting hip fractures and death was compared between L1 and L3 measures. Performance for predicting hip fractures and death was also evaluated using the established clinical risk scores from the fracture risk assessment tool (FRAX) and Framingham risk score (FRS), respectively. RESULTS. Median clinical follow-up interval after CT was 8.8 years (interquartile range, 5.1-11.6 years), yielding hip fractures and death in 219 (2.4%) and 549 (6.0%) patients, respectively. L1-level and L3-level muscle attenuation measurements were not different in 2-, 5-, or 10-year AUC for hip fracture (p = .18-.98) or death (p = .19-.95). For hip fracture, 5-year AUCs for L1-level muscle attenuation, L3-level muscle attenuation, and FRAX score were 0.717, 0.709, and 0.708, respectively. For death, 5-year AUCs for L1-level muscle attenuation, L3-level muscle attenuation, and FRS were 0.737, 0.721, and 0.688, respectively. Lowest quartile hazard ratios (HRs) for hip fracture were 2.20 (L1 attenuation), 2.45 (L3 attenuation), and 2.53 (FRAX score), and for death were 3.25 (L1 attenuation), 3.58 (L3 attenuation), and 2.82 (FRS). CT-based muscle cross-sectional area measurements at L1 and L3 were less predictive for hip fracture and death (5-year AUC ≤ 0.571; HR ≤ 1.56). CONCLUSION. Automated CT-based measurements of muscle attenuation for myosteatosis at the L1 level compare favorably with previously established L3-level measurements and clinical risk scores for predicting hip fracture and death. Assessment for myopenia was less predictive of outcomes at both levels. CLINICAL IMPACT. Alternative use of the L1 rather than L3 level for CT-based muscle measurements allows sarcopenia assessment using both chest and abdominal CT scans, greatly increasing the potential yield of opportunistic CT screening.

Artificial intelligence and machine learning in aortic disease

PURPOSE OF REVIEW

Discuss foundational concepts for artificial intelligence (AI) and review recent literature on its application to aortic disease.

RECENT FINDINGS

Machine learning (ML) techniques are rapidly evolving for the evaluation of aortic disease - broadly categorized as algorithms for aortic segmentation, detection of pathology, and risk stratification. Advances in deep learning, particularly U-Net architectures, have revolutionized segmentation of the aorta and show potential for monitoring the size of aortic aneurysm and characterizing aortic dissection. These algorithms also facilitate application of more complex technologies including analysis of flow dynamics with 4D Flow magnetic resonance imaging (MRI) and computational simulation of fluid dynamics for aortic coarctation. In addition, AI algorithms have been proposed to assist in 'opportunistic' screening from routine imaging exams, including automated aortic calcification score, which has emerged as a strong predictor of cardiovascular risk. Finally, several ML algorithms are being explored for risk stratification of patients with aortic aneurysm and dissection, in addition to prediction of postprocedural complications.

SUMMARY

Multiple ML techniques have potential for characterization and risk prediction of aortic aneurysm, dissection, coarctation, and atherosclerotic disease on computed tomography and MRI. This nascent field shows considerable promise with many applications in development and in early preclinical evaluation.

Progress and Research Priorities in Imaging Genomics for Heart and Lung Disease: Summary of an NHLBI Workshop

Imaging genomics is a rapidly evolving field that combines state-of-the-art bioimaging with genomic information to resolve phenotypic heterogeneity associated with genomic variation, improve risk prediction, discover prevention approaches, and enable precision diagnosis and treatment. Contemporary bioimaging methods provide exceptional resolution generating discrete and quantitative high-dimensional phenotypes for genomics investigation. Despite substantial progress in combining high-dimensional bioimaging and genomic data, methods for imaging genomics are evolving. Recognizing the potential impact of imaging genomics on the study of heart and lung disease, the National Heart, Lung, and Blood Institute convened a workshop to review cutting-edge approaches and methodologies in imaging genomics studies, and to establish research priorities for future investigation. This report summarizes the presentations and discussions at the workshop. In particular, we highlight the need for increased availability of imaging genomics data in diverse populations, dedicated focus on less common conditions, and centralization of efforts around specific disease areas.

Current Status of Radiomics and Deep Learning in Liver Imaging

ABSTRACT

Artificial intelligence is poised to revolutionize medical image. It takes advantage of the high-dimensional quantitative features present in medical images that may not be fully appreciated by humans. Artificial intelligence has the potential to facilitate automatic organ segmentation, disease detection and characterization, and prediction of disease recurrence. This article reviews the current status of artificial intelligence in liver imaging and reviews the opportunities and challenges in clinical implementation.

The predictive value of coronary artery calcification score combined with bone mineral density for the 2-year risk of cardiovascular events in maintenance hemodialysis patients

PURPOSE

Cardiovascular disease is the leading cause of death in maintenance hemodialysis (MHD) patients. The aim of this study is to investigate the predictive value of coronary artery calcification score (CACs) combined with bone mineral density (BMD) for the risk of cardiovascular diseases in MHD patients.

METHODS

From January 2017 to January 2019, we enrolled 112 MHD patients and 112 controls in Ningbo First Hospital, and retrospectively counted the cardiovascular events in the next 2 years after enrollment. According to the occurrence of cardiovascular events, the MHD patients were divided into CVD group and non-CVD group. The differences of vertebral BMD and CACs between the two groups were compared. ROC curve, Kaplan-Meier curve and Cox regression analyses were used for assess the predictive value of 2-year cardiovascular events in MHD patients.

RESULTS

Among 112 MHD patients, 49 (43.75%) patients had cardiovascular events. The results showed that the average value of BMD in MHD patients was significantly lower than that in the control group (99.88 ± 30.99 VS. 108.35 ± 23.98, P = 0.0231). The CACs in MHD patients were significantly higher than that in the control group (317.81 ± 211.53 VS. 190.03 ± 100.50, P < 0.001). The results between CVD group and the non-CVD group were to the same direction (BMD: 81.12 ± 31.28 VS. 114.48 ± 21.61, P < 0.001; CACs: 447.16 ± 234.11 VS. 217.21 ± 119.03, P < 0.001). Besides, CACs combined with BMD yield an AUC of 0.875 with a sensitivity of 79.60%, a specificity of 82.50%. Kaplan-Meier curve and Cox regression analyses indicated that CACs and BMD were independently associated with high risk of cardiovascular events in MHD patients.

CONCLUSION

The combination of CACs and vertebral BMD could predict the occurrence of cardiovascular events in MHD patients to some extent.

Heart failure risk estimation based on novel biomarkers

Introduction: Despite advances in medical care, heart failure (HF)-associated morbidity and mortality remains high. Consequently, there is increased effort to find better ways for predicting, screening, and prognosticating HF in order to facilitate effective primary and secondary prevention.Areas covered: In this review, we describe the various biomarkers associated with different etiologic pathways implicated in HF, and discuss their roles in screening, diagnosing, prognosticating and predicting HF. We explore the emerging role of multi-omic approaches. We performed electronic searches in databases (PubMed and Google Scholar) through December 2020, using the following key terms: biomarker, novel, heart failure, risk, prediction, and estimation.Circulating BNP and troponin concentrations have been established in clinical care as key biomarkers for diagnosing and prognosticating HF. Emerging biomarkers (such as galectin-3 and ST-2) have gained further recognition for use in evaluating prognosis of HF patients. Promising biomarkers that are yet to be part of clinical recommendations include biomarkers of cardiorenal disease.Expert opinion: Increasing recognition of the complex and interdependent nature of pathophysiological pathways of HF has led to the application of multi-marker approaches including multi-omic high throughput assays. These newer approaches have the potential for new therapeutic discoveries and improving precision medicine in HF.

Leveraging Machine Learning and Artificial Intelligence to Improve Peripheral Artery Disease Detection, Treatment, and Outcomes

Peripheral artery disease is an atherosclerotic disorder which, when present, portends poor patient outcomes. Low diagnosis rates perpetuate poor management, leading to limb loss and excess rates of cardiovascular morbidity and death. Machine learning algorithms and artificially intelligent systems have shown great promise in application to many areas in health care, such as accurately detecting disease, predicting patient outcomes, and automating image interpretation. Although the application of these technologies to peripheral artery disease are in their infancy, their promises are tremendous. In this review, we provide an introduction to important concepts in the fields of machine learning and artificial intelligence, detail the current state of how these technologies have been applied to peripheral artery disease, and discuss potential areas for future care enhancement with advanced analytics.

Automated assessment of longitudinal biomarker changes at abdominal CT: correlation with subsequent cardiovascular events in an asymptomatic adult screening cohort

BACKGROUND

Cardiovascular (CV) disease is a major public health concern, and automated methods can potentially capture relevant longitudinal changes on CT for opportunistic CV screening purposes.

METHODS

Fully-automated and validated algorithms that quantify abdominal fat, muscle, bone, liver, and aortic calcium were retrospectively applied to a longitudinal adult screening cohort undergoing serial non-contrast CT examination between 2005 and 2016. Downstream major adverse events (MI/CVA/CHF/death) were identified via algorithmic EHR search. Logistic regression, ROC curve, and Cox survival analyses assessed for associations between changes in CT variables and adverse events.

RESULTS

Final cohort included 1949 adults (942 M/1007F; mean age, 56.2 ± 6.2 years at initial CT). Mean interval between CT scans was 5.8 ± 2.0 years. Mean clinical follow-up interval from initial CT was 10.4 ± 2.7 years. Major CV events occurred after follow-up CT in 230 total subjects (11.8%). Mean change in aortic calcium Agatston score was significantly higher in CV(+) cohort (591.6 ± 1095.3 vs. 261.1 ± 764.3), as was annualized Agatston change (120.5 ± 263.6 vs. 46.7 ± 143.9) (p < 0.001 for both). 5-year area under the ROC curve (AUC) for Agatston change was 0.611. Hazard ratio for Agatston score change > 500 was 2.8 (95% CI 1.5-4.0) relative to < 500. Agatston score change was the only significant univariate CT biomarker in the survival analysis. Changes in fat and bone measures added no meaningful prediction.

CONCLUSION

Interval change in automated CT-based abdominal aortic calcium load represents a promising predictive longitudinal tool for assessing cardiovascular and mortality risks. Changes in other body composition measures were less predictive of adverse events.

CT-determined sarcopenia as a predictor of post-operative outcomes in patients undergoing an emergency laparotomy

PURPOSE

Emergency laparotomy has a high reported thirty-day mortality, ranging from 11 to 15%. Current peri-operative risk assessment tools may poorly estimate the risk of perioperative mortality. We sought to determine if CT-determined sarcopenia may be a useful predictor of post-operative outcomes in patients undergoing an emergency laparotomy.

METHODS

A retrospective review of a prospectively maintained database of consecutive adult patients who underwent an emergency laparotomy at our institution was performed. Post-operative mortality (90-day mortality), admission to HDU or ICU and APACHE-II scores were recorded for these patients. Sarcopenia was calculated by determining psoas area and density at the level of the third lumbar vertebra. The lowest quartile was determined to be sarcopenic. Univariate statistical analysis investigated associations between sarcopenia and these outcome measures.

RESULTS

Eighty patients were included in the study following application of exclusion criteria. Thirty-eight were male. The 90-day mortality rate was 11%. Compared to their non-sarcopenic counterparts, sarcopenic patients were significantly more likely to have died by 90 days post-operatively (χ² = 9.51, p = 0.002) and to require admission to either the HDU or ICU in the post-operative period (χ² = 10.62, p = 0.001).

CONCLUSIONS

CT determined sarcopenia is significantly associated with 90-day mortality and post-operative admission to HDU or ICU in patients undergoing an emergency laparotomy. The future development of a validated scoring tool incorporating sarcopenia could help to better select out those patients who will require higher levels of post-operative care as well as identifying those for whom surgery may not confer a survival benefit and be deemed futile.

Muscle Loss Is Associated with Overall Survival in Patients with Metastatic Colorectal Cancer Independent of Tumor Mutational Status and Weight Loss

Background

Survival in patients with metastatic colorectal cancer (mCRC) has been associated with tumor mutational status, muscle loss, and weight loss. We sought to explore the combined effects of these variables on overall survival.

Materials and Methods

We performed an observational cohort study, prospectively enrolling patients receiving chemotherapy for mCRC. We retrospectively assessed changes in muscle (using computed tomography) and weight, each dichotomized as >5% or ≤5% loss, at 3, 6, and 12 months after diagnosis of mCRC. We used regression models to assess relationships between tumor mutational status, muscle loss, weight loss, and overall survival. Additionally, we evaluated associations between muscle loss, weight loss, and tumor mutational status.

Results

We included 226 patients (mean age 59 ± 13 years, 53% male). Tumor mutational status included 44% wild type, 42% RAS‐mutant, and 14% BRAF‐mutant. Patients with >5% muscle loss at 3 and 12 months experienced worse survival controlling for mutational status and weight (3 months hazard ratio, 2.66; p < .001; 12 months hazard ratio, 2.10; p = .031). We found an association of >5% muscle loss with BRAF‐mutational status at 6 and 12 months. Weight loss was not associated with survival nor mutational status.

Conclusion

Increased muscle loss at 3 and 12 months may identify patients with mCRC at risk for decreased overall survival, independent of tumor mutational status. Specifically, >5% muscle loss identifies patients within each category of tumor mutational status with decreased overall survival in our sample. Our findings suggest that quantifying muscle loss on serial computed tomography scans may refine survival estimates in patients with mCRC.

Implications for Practice

In this study of 226 patients with metastatic colorectal cancer, it was found that losing >5% skeletal muscle at 3 and 12 months after the diagnosis of metastatic disease was associated with worse overall survival, independent of tumor mutational status and weight loss. Interestingly, results did not show a significant association between weight loss and overall survival. These findings suggest that muscle quantification on serial computed tomography may refine survival estimates in patients with metastatic colorectal cancer beyond mutational status.

Cancer cachexia has traditionally been defined using weight loss; however, loss of skeletal muscle may be a more objective measure. This article reports the results of a retrospective study that assessed whether skeletal muscle loss is associated with overall survival in patients with metastatic colorectal cancer, independent of tumor mutational status and weight loss.

Use of Variational Autoencoders with Unsupervised Learning to Detect Incorrect Organ Segmentations at CT

Purpose

To develop a deep learning model to detect incorrect organ segmentations at CT.

Materials and Methods

In this retrospective study, a deep learning method was developed using variational autoencoders (VAEs) to identify problematic organ segmentations. First, three different three-dimensional (3D) U-Nets were trained on segmented CT images of the liver (n = 141), spleen (n = 51), and kidney (n = 66). A total of 12 495 CT images then were segmented by the 3D U-Nets, and output segmentations were used to train three different VAEs for the detection of problematic segmentations. Automatic reconstruction errors (Dice scores) were then calculated. A random sampling of 2510 segmented images each for the liver, spleen, and kidney models were assessed manually by a human reader to determine problematic and correct segmentations. The ability of the VAEs to identify unusual or problematic segmentations was evaluated using receiver operating characteristic curve analysis and compared with traditional non-deep learning methods for outlier detection. Using the VAE outputs, passive and active learning approaches were performed on the original 3D U-Nets to determine if training could decrease segmentation error rates (15 CT scans were added to the original training data, according to each approach).

Results

The mean area under the receiver operating characteristic curve (AUC) for detecting problematic segmentations using the VAE method was 0.90 (95% CI: 0.89, 0.92) for kidney, 0.94 (95% CI: 0.93, 0.95) for liver, and 0.81 (95% CI: 0.80, 0.82) for spleen. The VAE performance was higher compared with traditional methods in most cases. For example, for liver segmentation, the highest performing non-deep learning method for outlier detection had an AUC of 0.83 (95% CI: 0.77, 0.90) compared with 0.94 (95% CI: 0.93, 0.95) using the VAE method (P < .05). Using the information on problematic segmentations for active learning approaches decreased 3D U-Net segmentation error rates (original error rate, 7.1%; passive learning, 6.0%; active learning, 5.7%).

Conclusion

A method was developed to screen for unusual and problematic automatic organ segmentations using a 3D VAE.Keywords: Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms, Segmentation, CT© RSNA, 2021.

A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises

Since its renaissance, deep learning has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high performance computing. However, medical imaging presents unique challenges that confront deep learning approaches. In this survey paper, we first present traits of medical imaging, highlight both clinical needs and technical challenges in medical imaging, and describe how emerging trends in deep learning are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, etc. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.

Cardiovascular Informatics: building a bridge to data harmony

The search for new strategies for better understanding cardiovascular disease is a constant one, spanning multitudinous types of observations and studies. A comprehensive characterization of each disease state and its biomolecular underpinnings relies upon insights gleaned from extensive information collection of various types of data. Researchers and clinicians in cardiovascular biomedicine repeatedly face questions regarding which types of data may best answer their questions, how to integrate information from multiple datasets of various types, and how to adapt emerging advances in machine learning and/or artificial intelligence to their needs in data processing. Frequently lauded as a field with great practical and translational potential, the interface between biomedical informatics and cardiovascular medicine is challenged with staggeringly massive datasets. Successful application of computational approaches to decode these complex and gigantic amounts of information becomes an essential step toward realizing the desired benefits. In this review, we examine recent efforts to adapt informatics strategies to cardiovascular biomedical research: automated information extraction and unification of multifaceted -omics data. We discuss how and why this interdisciplinary space of Cardiovascular Informatics is particularly relevant to and supportive of current experimental and clinical research. We describe in detail how open data sources and methods can drive discovery while demanding few initial resources, an advantage afforded by widespread availability of cloud computing-driven platforms. Subsequently, we provide examples of how interoperable computational systems facilitate exploration of data from multiple sources, including both consistently-formatted structured data and unstructured data. Taken together, these approaches for achieving data harmony enable molecular phenotyping of cardiovascular (CV) diseases and unification of cardiovascular knowledge.

Opportunistic Screening at Abdominal CT: Use of Automated Body Composition Biomarkers for Added Cardiometabolic Value

Abdominal CT is a frequently performed imaging examination for a wide variety of clinical indications. In addition to the immediate reason for scanning, each CT examination contains robust additional data on body composition that generally go unused in routine clinical practice. There is now growing interest in harnessing this additional information. Prime examples of cardiometabolic information include measurement of bone mineral density for osteoporosis screening, quantification of aortic calcium for assessment of cardiovascular risk, quantification of visceral fat for evaluation of metabolic syndrome, assessment of muscle bulk and density for diagnosis of sarcopenia, and quantification of liver fat for assessment of hepatic steatosis. All of these relevant biometric measures can now be fully automated through the use of artificial intelligence algorithms, which provide rapid and objective assessment and allow large-scale population-based screening. Initial investigations into these measures of body composition have demonstrated promising performance for prediction of future adverse events that matches or exceeds the best available clinical prediction models, particularly when these CT-based measures are used in combination. In this review, the concept of CT-based opportunistic screening is discussed, and an overview of the various automated biomarkers that can be derived from essentially all abdominal CT examinations is provided, drawing heavily on the authors' experience. As radiology transitions from a volume-based to a value-based practice, opportunistic screening represents a promising example of adding value to services that are already provided. If the potentially high added value of these objective CT-based automated measures is ultimately confirmed in subsequent investigations, this opportunistic screening approach could be considered for intentional CT-based screening. ^©RSNA, 2021.

Fully automated CT imaging biomarkers of bone, muscle, and fat: correcting for the effect of intravenous contrast

PURPOSE

Fully automated CT-based algorithms for quantifying bone, muscle, and fat have been validated for unenhanced abdominal scans. The purpose of this study was to determine and correct for the effect of intravenous (IV) contrast on these automated body composition measures.

MATERIALS AND METHODS

Initial study cohort consisted of 1211 healthy adults (mean age, 45.2 years; 733 women) undergoing abdominal CT for potential renal donation. Multiphasic CT protocol consisted of pre-contrast, arterial, and parenchymal phases. Fully automated CT-based algorithms for quantifying bone mineral density (BMD, L1 trabecular HU), muscle area and density (L3-level MA and M-HU), and fat (visceral/subcutaneous (V/S) fat ratio) were applied to pre-contrast and parenchymal phases. Effect of IV contrast upon these body composition measures was analyzed. Square of the Pearson correlation coefficient (r²) was generated for each comparison.

RESULTS

Mean changes (± SD) in L1 BMD, L3-level MA and M-HU, and V/S fat ratio were 26.7 ± 27.2 HU, 2.9 ± 10.2 cm², 18.8 ± 6.0 HU, - 0.1 ± 0.2, respectively. Good linear correlation between pre- and post-contrast values was observed for all automated measures: BMD (pre = 0.87 × post; r² = 0.72), MA (pre = 0.98 × post; r² = 0.92), M-HU (pre = 0.75 × post  + 5.7; r² = 0.75), and V/S (pre = 1.11 × post; r² = 0.94); p < 0.001 for all r² values. There were no significant trends according to patient age or gender that required further correction.

CONCLUSION

Fully automated quantitative tissue measures of bone, muscle, and fat at contrast-enhanced abdominal CT can be correlated with non-contrast equivalents using simple, linear relationships. These findings will facilitate evaluation of mixed CT cohorts involving larger patient populations and could greatly expand the potential for opportunistic screening.

Assisting scalable diagnosis automatically via CT images in the combat against COVID-19

The pandemic of Coronavirus Disease 2019 (COVID-19) is causing enormous loss of life globally. Prompt case identification is critical. The reference method is the real-time reverse transcription PCR (RT-PCR) assay, whose limitations may curb its prompt large-scale application. COVID-19 manifests with chest computed tomography (CT) abnormalities, some even before the onset of symptoms. We tested the hypothesis that the application of deep learning (DL) to 3D CT images could help identify COVID-19 infections. Using data from 920 COVID-19 and 1,073 non-COVID-19 pneumonia patients, we developed a modified DenseNet-264 model, COVIDNet, to classify CT images to either class. When tested on an independent set of 233 COVID-19 and 289 non-COVID-19 pneumonia patients, COVIDNet achieved an accuracy rate of 94.3% and an area under the curve of 0.98. As of March 23, 2020, the COVIDNet system had been used 11,966 times with a sensitivity of 91.12% and a specificity of 88.50% in six hospitals with PCR confirmation. Application of DL to CT images may improve both efficiency and capacity of case detection and long-term surveillance.

Assessing the Role of Pericardial Fat as a Biomarker Connected to Coronary Calcification-A Deep Learning Based Approach Using Fully Automated Body Composition Analysis

(1) Background: Epi- and Paracardial Adipose Tissue (EAT, PAT) have been spotlighted as important biomarkers in cardiological assessment in recent years. Since biomarker quantification is an increasingly important method for clinical use, we wanted to examine fully automated EAT and PAT quantification for possible use in cardiovascular risk stratification. (2) Methods: 966 patients with intermediate Framingham risk scores for Coronary Artery Disease referred for coronary calcium scans were included in clinical routine retrospectively. The Coronary Artery Calcium Score (CACS) was extracted and tissue quantification was performed by a deep learning network. (3) Results: The Computed Tomography (CT) segmentations predicted by the network indicated no significant correlation between EAT volume and EAT radiodensity when compared to Agatston score (r = 0.18, r = −0.09). CACS 0 category patients showed significantly lower levels of total EAT and PAT volumes and higher EAT and PAT densities than CACS 1–99 category patients (p < 0.01). Notably, this difference did not reach significance regarding EAT attenuation in male patients. Women older than 50 years, thus more likely to be postmenopausal, were shown to be at higher risk of coronary calcification (p < 0.01, OR = 4.59). CACS 1–99 vs. CACS ≥100 category patients remained below significance level (EAT volume: p = 0.087, EAT attenuation: p = 0.98). (4) Conclusions: Our study proves the feasibility of a fully automated adipose tissue analysis in clinical cardiac CT and confirms in a large clinical cohort that volume and attenuation of EAT and PAT are not correlated with CACS. Broadly available deep learning based rapid and reliable tissue quantification should thus be discussed as a method to assess this biomarker as a supplementary risk predictor in cardiac CT.

Assessment of Intramyocardial Fat Content Using Computed Tomography: Is There a Relationship With Obesity?

BACKGROUND

Fat deposition in the liver and the skeletal muscle are linked to cardiovascular risk factors. Fat content in tissues can be estimated by measuring attenuation on noncontrast computed tomography (CT). Quantifying intramyocardial fat content is of interest as it may be related to myocardial dysfunction or development of heart failure. We hypothesized that myocardial fat content would correlate with severity of obesity, liver fat, and components of the metabolic syndrome.

METHODS

We measured attenuation values on 121 noncontrast CT scans from the spleen, liver, erector spinae muscle, and myocardial septum. A chart review was performed for patient demographics and clinical characteristics. We tested for correlations between attenuation values in each of the tissues and various clinical parameters.

RESULTS

We studied 78 females and 43 males, with a mean age of 54.5±11.2 years. Weak, but significant inverse Spearman correlation between body mass index and attenuation values were found in the liver (ρ=-0.228, P=0.012), spleen (ρ=-0.225, P=0.017), and erector spinae muscle (ρ=-0.211, P=0.022) but not in the myocardial septum (ρ=0.012, P=0.897). Mean attenuation in the nonobese group versus obese group (body mass index >30 kg/m2) were 41.1±5.0 versus 42.3±6.9 (P=0.270) in myocardial septum, 56.1±8.7 versus 51.7±10.9 (P=0.016) in the liver, 43.9±8.9 versus 40.1±10.4 (P=0.043) in the spleen, and 41.7±8.3 versus 39.0±8.8 (P=0.087) in the erector spinae muscle.

CONCLUSIONS

Although CT is a theoretically appealing modality to assess fat content of the myocardium, we did not find a relationship between myocardial CT attenuation and obesity, or other cardiovascular risk factors. These findings suggest that the degree of myocardial fat accumulation in obesity or metabolic syndrome is too small to be detected with this modality.

Population-Scale CT-based Body Composition Analysis of a Large Outpatient Population Using Deep Learning to Derive Age-, Sex-, and Race-specific Reference Curves

Background Although CT-based body composition (BC) metrics may inform disease risk and outcomes, obtaining these metrics has been too resource intensive for large-scale use. Thus, population-wide distributions of BC remain uncertain. Purpose To demonstrate the validity of fully automated, deep learning BC analysis from abdominal CT examinations, to define demographically adjusted BC reference curves, and to illustrate the advantage of use of these curves compared with standard methods, along with their biologic significance in predicting survival. Materials and Methods After external validation and equivalency testing with manual segmentation, a fully automated deep learning BC analysis pipeline was applied to a cross-sectional population cohort that included any outpatient without a cardiovascular disease or cancer who underwent abdominal CT examination at one of three hospitals in 2012. Demographically adjusted population reference curves were generated for each BC area. The z scores derived from these curves were compared with sex-specific thresholds for sarcopenia by using χ² tests and used to predict 2-year survival in multivariable Cox proportional hazards models that included weight and body mass index (BMI). Results External validation showed excellent correlation (R = 0.99) and equivalency (P < .001) of the fully automated deep learning BC analysis method with manual segmentation. With use of the fully automated BC data from 12 128 outpatients (mean age, 52 years; 6936 [57%] women), age-, race-, and sex-normalized BC reference curves were generated. All BC areas varied significantly with these variables (P < .001 except for subcutaneous fat area vs age [P = .003]). Sex-specific thresholds for sarcopenia demonstrated that age and race bias were not present if z scores derived from the reference curves were used (P < .001). Skeletal muscle area z scores were significantly predictive of 2-year survival (P = .04) in combined models that included BMI. Conclusion Fully automated body composition (BC) metrics vary significantly by age, race, and sex. The z scores derived from reference curves for BC parameters better capture the demographic distribution of BC compared with standard methods and can help predict survival. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Summers in this issue.

Utilizing Fully Automated Abdominal CT-Based Biomarkers for Opportunistic Screening for Metabolic Syndrome in Adults Without Symptoms

OBJECTIVE

Metabolic syndrome describes a constellation of reversible cardiometabolic abnormalities associated with cardiovascular risk and diabetes. The present study investigates the use of fully automated abdominal CT-based biometric measures for opportunistic identification of metabolic syndrome in adults without symptoms.

MATERIALS AND METHODS

International Diabetes Federation criteria were applied to a cohort of 9223 adults without symptoms who underwent unenhanced abdominal CT. After patients with insufficient clinical data for diagnosis were excluded, the final cohort consisted of 7785 adults (mean age, 57.0 years; 4361 women and 3424 men). Previously validated and fully automated CT-based algorithms for quantifying muscle, visceral and subcutaneous fat, liver fat, and abdominal aortic calcification were applied to this final cohort.

RESULTS

A total of 738 subjects (9.5% of all subjects; mean age, 56.7 years; 372 women and 366 men) met the clinical criteria for metabolic syndrome. Subsequent major cardiovascular events occurred more frequently in the cohort with metabolic syndrome (p < 0.001). Significant differences were observed between the two groups for all CT-based biomarkers (p < 0.001). Univariate L1-level total abdominal fat (area under the ROC curve [AUROC] = 0.909; odds ratio [OR] = 27.2), L3-level skeletal muscle index (AUROC = 0.776; OR = 5.8), and volumetric liver attenuation (AUROC = 0.738; OR = 5.1) performed well when compared with abdominal aortic calcification scoring (AUROC = 0.578; OR = 1.6). An L1-level total abdominal fat threshold of 460.6 cm² was 80.1% sensitive and 85.4% specific for metabolic syndrome. For women, the AUROC was 0.930 when fat and muscle measures were combined.

CONCLUSION

Fully automated quantitative tissue measures of fat, muscle, and liver derived from abdominal CT scans can help identify individuals who are at risk for metabolic syndrome. These visceral measures can be opportunistically applied to CT scans obtained for other clinical indications, and they may ultimately provide a more direct and useful definition of metabolic syndrome.

Liver fat quantification: where do we stand?

Excessive intracellular accumulation of triglycerides in the liver, or hepatic steatosis, is a highly prevalent condition affecting approximately one billion people worldwide. In the absence of secondary cause, the term nonalcoholic fatty liver disease (NAFLD) is used. Hepatic steatosis may progress into nonalcoholic steatohepatitis, the more aggressive form of NAFLD, associated with hepatic complications such as fibrosis, liver failure and hepatocellular carcinoma. Hepatic steatosis is associated with metabolic syndrome, cardiovascular disease and represents an independent risk factor for type 2 diabetes, cardiovascular disease and malignancy. Percutaneous liver biopsy is the current reference standard for NAFLD assessment; however, it is an invasive procedure associated with complications and suffers from high sampling variability, impractical for clinical routine and drug efficiency studies. Therefore, noninvasive imaging methods are increasingly used for the diagnosis and monitoring of NAFLD. Among the methods quantifying liver fat, chemical-shift-encoded MRI (CSE-MRI)-based proton density fat-fraction (PDFF) has shown the most promise. MRI-PDFF is increasingly accepted as quantitative imaging biomarker of liver fat that is transforming daily clinical practice and influencing the development of new treatments for NAFLD. Furthermore, CT is an important imaging method for detection of incidental steatosis, and the practical advantages of quantitative ultrasound hold great promise for the future. Understanding the disease burden of NAFLD and the role of imaging may initiate important interventions aimed at avoiding the hepatic and extrahepatic complications of NAFLD. This article reviews clinical burden of NAFLD, and the role of noninvasive imaging techniques for quantification of liver fat.

Automated Abdominal CT Imaging Biomarkers for Opportunistic Prediction of Future Major Osteoporotic Fractures in Asymptomatic Adults

Background Body composition data from abdominal CT scans have the potential to opportunistically identify those at risk for future fracture. Purpose To apply automated bone, muscle, and fat tools to noncontrast CT to assess performance for predicting major osteoporotic fractures and to compare with the Fracture Risk Assessment Tool (FRAX) reference standard. Materials and Methods Fully automated bone attenuation (L1-level attenuation), muscle attenuation (L3-level attenuation), and fat (L1-level visceral-to-subcutaneous [V/S] ratio) measures were derived from noncontrast low-dose abdominal CT scans in a generally healthy asymptomatic adult outpatient cohort from 2004 to 2016. The FRAX score was calculated from data derived from an algorithmic electronic health record search. The cohort was assessed for subsequent future fragility fractures. Subset analysis was performed for patients evaluated with dual x-ray absorptiometry (n = 2106). Hazard ratios (HRs) and receiver operating characteristic curve analyses were performed. Results A total of 9223 adults were evaluated (mean age, 57 years ± 8 [standard deviation]; 5152 women) at CT and were followed over a median time of 8.8 years (interquartile range, 5.1-11.6 years), with documented subsequent major osteoporotic fractures in 7.4% (n = 686), including hip fractures in 2.4% (n = 219). Comparing the highest-risk quartile with the other three quartiles, HRs for bone attenuation, muscle attenuation, V/S fat ratio, and FRAX were 2.1, 1.9, 0.98, and 2.5 for any fragility fracture and 2.0, 2.5, 1.1, and 2.5 for femoral fractures, respectively (P < .001 for all except V/S ratio, which was P ≥ .51). Area under the receiver operating characteristic curve (AUC) values for fragility fracture were 0.71, 0.65, 0.51, and 0.72 at 2 years and 0.63, 0.62, 0.52, and 0.65 at 10 years, respectively. For hip fractures, 2-year AUC for muscle attenuation alone was 0.75 compared with 0.73 for FRAX (P = .43). Multivariable 2-year AUC combining bone and muscle attenuation was 0.73 for any fragility fracture and 0.76 for hip fractures, respectively (P ≥ .73 compared with FRAX). For the subset with dual x-ray absorptiometry T-scores, 2-year AUC was 0.74 for bone attenuation and 0.65 for FRAX (P = .11). Conclusion Automated bone and muscle imaging biomarkers derived from CT scans provided comparable performance to Fracture Risk Assessment Tool score for presymptomatic prediction of future osteoporotic fractures. Muscle attenuation alone provided effective hip fracture prediction. © RSNA, 2020 See also the editorial by Smith in this issue.

Liver Steatosis Categorization on Contrast-Enhanced CT Using a Fully Automated Deep Learning Volumetric Segmentation Tool: Evaluation in 1204 Healthy Adults Using Unenhanced CT as a Reference Standard

BACKGROUND. Hepatic attenuation at unenhanced CT is linearly correlated with the MRI proton density fat fraction (PDFF). Liver fat quantification at contrast-enhanced CT is more challenging. OBJECTIVE. The purpose of this article is to evaluate liver steatosis categorization on contrast-enhanced CT using a fully automated deep learning volumetric hepatosplenic segmentation algorithm and unenhanced CT as the reference standard. METHODS. A fully automated volumetric hepatosplenic segmentation algorithm using 3D convolutional neural networks was applied to unenhanced and contrast-enhanced series from a sample of 1204 healthy adults (mean age, 45.2 years; 726 women, 478 men) undergoing CT evaluation for renal donation. The mean volumetric attenuation was computed from all designated liver and spleen voxels. PDFF was estimated from unenhanced CT attenuation and served as the reference standard. Contrast-enhanced attenuations were evaluated for detecting PDFF thresholds of 5% (mild steatosis, 10% and 15% (moderate steatosis); PDFF less than 5% was considered normal. RESULTS. Using unenhanced CT as reference, estimated PDFF was ≥ 5% (mild steatosis), ≥ 10%, and ≥ 15% (moderate steatosis) in 50.1% (n = 603), 12.5% (n = 151) and 4.8% (n = 58) of patients, respectively. ROC AUC values for predicting PDFF thresholds of 5%, 10%, and 15% using contrast-enhanced liver attenuation were 0.669, 0.854, and 0.962, respectively, and using contrast-enhanced liver-spleen attenuation difference were 0.662, 0.866, and 0.986, respectively. A total of 96.8% (90/93) of patients with contrast-enhanced liver attenuation less than 90 HU had steatosis (PDFF ≥ 5%); this threshold of less than 90 HU achieved sensitivity of 75.9% and specificity of 95.7% for moderate steatosis (PDFF ≥ 15%). Liver attenuation less than 100 HU achieved sensitivity of 34.0% and specificity of 94.2% for any steatosis (PDFF ≥ 5%). A total of 93.8% (30/32) of patients with contrast-enhanced liver-spleen attenuation difference 10 HU or less had moderate steatosis (PDFF ≥ 15%); a liver-spleen difference less than 5 HU achieved sensitivity of 91.4% and specificity of 95.0% for moderate steatosis. Liver-spleen difference less than 10 HU achieved sensitivity of 29.5% and specificity of 95.5% for any steatosis (PDFF ≥ 5%). CONCLUSION. Contrast-enhanced volumetric hepatosplenic attenuation derived using a fully automated deep learning CT tool may allow objective categoric assessment of hepatic steatosis. Accuracy was better for moderate than mild steatosis. Further confirmation using different scanning protocols and vendors is warranted. CLINICAL IMPACT. If these results are confirmed in independent patient samples, this automated approach could prove useful for both individualized and population-based steatosis assessment.

Diagnostic Performance of Multitarget Stool DNA and CT Colonography for Noninvasive Colorectal Cancer Screening

BackgroundMultitarget stool DNA (mt-sDNA) screening has increased rapidly since simultaneous approval by the U.S. Food and Drug Administration and Centers for Medicare and Medicaid Services in 2014, whereas CT colonography screening remains underused and is not covered by Centers for Medicare and Medicaid Services.PurposeTo report postapproval clinical experience with mt-sDNA screening for colorectal cancer (CRC) and compare results with CT colonography screening at the same center.Materials and MethodsIn this retrospective cohort study, asymptomatic adults underwent clinical mt-sDNA screening during a 5-year interval (2014-2019). Electronic medical records were searched to verify test results and document subsequent optical colonoscopy and histopathologic findings. A similar analysis was performed for CT colonography screening during a 15-year interval (2004-2019), with consideration of thresholds for positivity of both 6-mm and 10-mm polyp sizes. χ² or two-sample t tests were used for group comparisons.ResultsA total of 3987 asymptomatic adult patients (mean age, 64 years ± 9 [standard deviation]; 2567 women) underwent mt-sDNA screening and 9656 patients (mean age, 57 years ± 8; 5200 women) underwent CT colonography. Test-positive rates for mt-sDNA and for 6-mm- and 10-mm-threshold CT colonography were 15.2%, 16.4%, and 6.7%, respectively. Optical colonoscopy follow-up rates for positive results of mt-sDNA and 6-mm- and 10-mm-threshold CT colonography were 13.1%, 12.3%, and 5.9%, respectively. Positive predictive values (PPVs) for any neoplasm 6 mm or greater, advanced neoplasia, and CRC for mt-sDNA were 54.2%, 22.7%, and 1.9% respectively; for 6-mm-threshold CT colonography, PPVs were 76.8%, 44.3%, and 2.7%; for 10-mm-threshold CT colonography, PPVs were 84.5%, 75.2%, and 5.2%, respectively (P < .001 for mt-sDNA vs CT colonography for all except 6-mm CRC at CT colonography). For mt-sDNA versus 6-mm-threshold CT colonography, overall detection rates for advanced neoplasia were 2.7% and 5.0%, respectively (P < .001); corresponding detection rates for CRC were 0.23% and 0.31%, respectively (P = .43).ConclusionThe detection rates of advanced neoplasia at CT colonography screening were greater than those of multitarget stool DNA. Detection rates were similar for colorectal cancer.© RSNA, 2020See also the editorial by Yee in this issue.