Holistic AI analysis of hybrid cardiac perfusion images for mortality prediction

Background

While low-dose computed tomography scans are traditionally used for attenuation correction in hybrid myocardial perfusion imaging (MPI), they also contain additional anatomic and pathologic information not utilized in clinical assessment. We seek to uncover the full potential of these scans utilizing a holistic artificial intelligence (AI)-driven image framework for image assessment.

Methods

Patients with SPECT/CT MPI from 4 REFINE SPECT registry sites were studied. A multi-structure model segmented 33 structures and quantified 15 radiomics features for each on CT attenuation correction (CTAC) scans. Coronary artery calcium and epicardial adipose tissue scores were obtained from separate deep-learning models. Normal standard quantitative MPI features were derived by clinical software. Extreme Gradient Boosting derived all-cause mortality risk scores from SPECT, CT, stress test, and clinical features utilizing a 10-fold cross-validation regimen to separate training from testing data. The performance of the models for the prediction of all-cause mortality was evaluated using area under the receiver-operating characteristic curves (AUCs).

Results

Of 10,480 patients, 5,745 (54.8%) were male, and median age was 65 (interquartile range [IQR] 57-73) years. During the median follow-up of 2.9 years (1.6-4.0), 651 (6.2%) patients died. The AUC for mortality prediction of the model (combining CTAC, MPI, and clinical data) was 0.80 (95% confidence interval [0.74-0.87]), which was higher than that of an AI CTAC model (0.78 [0.71-0.85]), and AI hybrid model (0.79 [0.72-0.86]) incorporating CTAC and MPI data (p<0.001 for all).

Conclusion

In patients with normal perfusion, the comprehensive model (0.76 [0.65-0.86]) had significantly better performance than the AI CTAC (0.72 [0.61-0.83]) and AI hybrid (0.73 [0.62-0.84]) models (p<0.001, for all).CTAC significantly enhances AI risk stratification with MPI SPECT/CT beyond its primary role - attenuation correction. A comprehensive multimodality approach can significantly improve mortality prediction compared to MPI information alone in patients undergoing cardiac SPECT/CT.

Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography

Chest computed tomography is one of the most common diagnostic tests, with 15 million scans performed annually in the United States. Coronary calcium can be visualized on these scans, but other measures of cardiac risk such as atrial and ventricular volumes have classically required administration of contrast. Here we show that a fully automated pipeline, incorporating two artificial intelligence models, automatically quantifies coronary calcium, left atrial volume, left ventricular mass, and other cardiac chamber volumes in 29,687 patients from three cohorts. The model processes chamber volumes and coronary artery calcium with an end-to-end time of ~18 s, while failing to segment only 0.1% of cases. Coronary calcium, left atrial volume, and left ventricular mass index are independently associated with all-cause and cardiovascular mortality and significantly improve risk classification compared to identification of abnormalities by a radiologist. This automated approach can be integrated into clinical workflows to improve identification of abnormalities and risk stratification, allowing physicians to improve clinical decision-making.

AI-Defined Cardiac Anatomy Improves Risk Stratification of Hybrid Perfusion Imaging

BACKGROUND

Computed tomography attenuation correction (CTAC) improves perfusion quantification of hybrid myocardial perfusion imaging by correcting for attenuation artifacts. Artificial intelligence (AI) can automatically measure coronary artery calcium (CAC) from CTAC to improve risk prediction but could potentially derive additional anatomic features.

OBJECTIVES

The authors evaluated AI-based derivation of cardiac anatomy from CTAC and assessed its added prognostic utility.

METHODS

The authors considered consecutive patients without known coronary artery disease who underwent single-photon emission computed tomography/computed tomography (CT) myocardial perfusion imaging at 3 separate centers. Previously validated AI models were used to segment CAC and cardiac structures (left atrium, left ventricle, right atrium, right ventricular volume, and left ventricular [LV] mass) from CTAC. They evaluated associations with major adverse cardiovascular events (MACEs), which included death, myocardial infarction, unstable angina, or revascularization.

RESULTS

In total, 7,613 patients were included with a median age of 64 years. During a median follow-up of 2.4 years (IQR: 1.3-3.4 years), MACEs occurred in 1,045 (13.7%) patients. Fully automated AI processing took an average of 6.2 ± 0.2 seconds for CAC and 15.8 ± 3.2 seconds for cardiac volumes and LV mass. Patients in the highest quartile of LV mass and left atrium, LV, right atrium, and right ventricular volume were at significantly increased risk of MACEs compared to patients in the lowest quartile, with HR ranging from 1.46 to 3.31. The addition of all CT-based volumes and CT-based LV mass improved the continuous net reclassification index by 23.1%.

CONCLUSIONS

AI can automatically derive LV mass and cardiac chamber volumes from CT attenuation imaging, significantly improving cardiovascular risk assessment for hybrid perfusion imaging.

Automated vessel specific coronary artery calcification quantification with deep learning in a large multi-center registry

AIMS

Vessel specific coronary artery calcification (CAC) is additive to global CAC for prognostic assessment. We assessed accuracy and prognostic implications of vessel-specific automated deep learning (DL) CAC analysis on electrocardiogram gated and attenuation correction computed tomography (CT) in a large multicenter registry.

METHODS AND RESULTS

Vessel-specific CAC was assessed in the left main/left anterior descending (LM/LAD), left circumflex (LCX) and right coronary artery (RCA) using a DL model trained on 3000 gated CT and tested on 2094 gated CT and 5969 non-gated attenuation correction CT. Vessel-specific agreement was assessed with linear weighted Cohen's Kappa for CAC zero, 1-100, 101-400 and >400 Agatston units (AU). Risk of major adverse cardiovascular events (MACE) was assessed during 2.4±1.4 years follow-up, with hazard ratios (HR) and 95% confidence intervals (CI). There was strong to excellent agreement between DL and expert ground truth for CAC in LM/LAD, LCX and RCA on gated CT [0.90 (95% CI 0.89 to 0.92); 0.70 (0.68 to 0.73); 0.79 (0.77 to 0.81)] and attenuation correction CT [(0.78 (0.77 to 0.80); 0.60 (0.58 to 0.62); 0.70 (0.68 to 0.71)]. MACE occurred in 242 (12%) undergoing gated CT and 841(14%) of undergoing attenuation correction CT. LM/LAD CAC >400 AU was associated with the highest risk of MACE on gated (HR 12.0, 95% CI 7.96, 18.0, p<0.001) and attenuation correction CT (HR 4.21, 95% CI 3.48, 5.08, p<0.001).

CONCLUSION

Vessel-specific CAC assessment with DL can be performed accurately and rapidly on gated CT and attenuation correction CT and provides important prognostic information.

AI-derived epicardial fat measurements improve cardiovascular risk prediction from myocardial perfusion imaging

Epicardial adipose tissue (EAT) volume and attenuation are associated with cardiovascular risk, but manual annotation is time-consuming. We evaluated whether automated deep learning-based EAT measurements from ungated computed tomography (CT) are associated with death or myocardial infarction (MI). We included 8781 patients from 4 sites without known coronary artery disease who underwent hybrid myocardial perfusion imaging. Of those, 500 patients from one site were used for model training and validation, with the remaining patients held out for testing (n = 3511 internal testing, n = 4770 external testing). We modified an existing deep learning model to first identify the cardiac silhouette, then automatically segment EAT based on attenuation thresholds. Deep learning EAT measurements were obtained in <2 s compared to 15 min for expert annotations. There was excellent agreement between EAT attenuation (Spearman correlation 0.90 internal, 0.82 external) and volume (Spearman correlation 0.90 internal, 0.91 external) by deep learning and expert segmentation in all 3 sites (Spearman correlation 0.90-0.98). During median follow-up of 2.7 years (IQR 1.6-4.9), 565 patients experienced death or MI. Elevated EAT volume and attenuation were independently associated with an increased risk of death or MI after adjustment for relevant confounders. Deep learning can automatically measure EAT volume and attenuation from low-dose, ungated CT with excellent correlation with expert annotations, but in a fraction of the time. EAT measurements offer additional prognostic insights within the context of hybrid perfusion imaging.

Artificial intelligence-assisted quantification of COVID-19 pneumonia burden from computed tomography improves prediction of adverse outcomes over visual scoring systems

OBJECTIVE

We aimed to evaluate the effectiveness of utilizing artificial intelligence (AI) to quantify the extent of pneumonia from chest CT scans, and to determine its ability to predict clinical deterioration or mortality in patients admitted to the hospital with COVID-19 in comparison to semi-quantitative visual scoring systems.

METHODS

A deep-learning algorithm was utilized to quantify the pneumonia burden, while semi-quantitative pneumonia severity scores were estimated through visual means. The primary outcome was clinical deterioration, the composite end point including admission to the intensive care unit, need for invasive mechanical ventilation, or vasopressor therapy, as well as in-hospital death.

RESULTS

The final population comprised 743 patients (mean age 65  ±  17 years, 55% men), of whom 175 (23.5%) experienced clinical deterioration or death. The area under the receiver operating characteristic curve (AUC) for predicting the primary outcome was significantly higher for AI-assisted quantitative pneumonia burden (0.739, p = 0.021) compared with the visual lobar severity score (0.711, p < 0.001) and visual segmental severity score (0.722, p = 0.042). AI-assisted pneumonia assessment exhibited lower performance when applied for calculation of the lobar severity score (AUC of 0.723, p = 0.021). Time taken for AI-assisted quantification of pneumonia burden was lower (38 ± 10 s) compared to that of visual lobar (328 ± 54 s, p < 0.001) and segmental (698 ± 147 s, p < 0.001) severity scores.

CONCLUSION

Utilizing AI-assisted quantification of pneumonia burden from chest CT scans offers a more accurate prediction of clinical deterioration in patients with COVID-19 compared to semi-quantitative severity scores, while requiring only a fraction of the analysis time.

ADVANCES IN KNOWLEDGE

Quantitative pneumonia burden assessed using AI demonstrated higher performance for predicting clinical deterioration compared to current semi-quantitative scoring systems. Such an AI system has the potential to be applied for image-based triage of COVID-19 patients in clinical practice.

Deep Learning Coronary Artery Calcium Scores from SPECT/CT Attenuation Maps Improve Prediction of Major Adverse Cardiac Events

Low-dose ungated CT attenuation correction (CTAC) scans are commonly obtained with SPECT/CT myocardial perfusion imaging. Despite the characteristically low image quality of CTAC, deep learning (DL) can potentially quantify coronary artery calcium (CAC) from these scans in an automatic manner. We evaluated CAC quantification derived with a DL model, including correlation with expert annotations and associations with major adverse cardiovascular events (MACE). Methods: We trained a convolutional long short-term memory DL model to automatically quantify CAC on CTAC scans using 6,608 studies (2 centers) and evaluated the model in an external cohort of patients without known coronary artery disease (n = 2,271) obtained in a separate center. We assessed agreement between DL and expert annotated CAC scores. We also assessed associations between MACE (death, revascularization, myocardial infarction, or unstable angina) and CAC categories (0, 1-100, 101-400, or >400) for scores manually derived by experienced readers and scores obtained fully automatically by DL using multivariable Cox models (adjusted for age, sex, past medical history, perfusion, and ejection fraction) and net reclassification index. Results: In the external testing population, DL CAC was 0 in 908 patients (40.0%), 1-100 in 596 (26.2%), 100-400 in 354 (15.6%), and >400 in 413 (18.2%). Agreement in CAC category by DL CAC and expert annotation was excellent (linear weighted κ, 0.80), but DL CAC was obtained automatically in less than 2 s compared with about 2.5 min for expert CAC. DL CAC category was an independent risk factor for MACE with hazard ratios in comparison to a CAC of zero: CAC of 1-100 (2.20; 95% CI, 1.54-3.14; P < 0.001), CAC of 101-400 (4.58; 95% CI, 3.23-6.48; P < 0.001), and CAC of more than 400 (5.92; 95% CI, 4.27-8.22; P < 0.001). Overall, the net reclassification index was 0.494 for DL CAC, which was similar to expert annotated CAC (0.503). Conclusion: DL CAC from SPECT/CT attenuation maps agrees well with expert CAC annotations and provides a similar risk stratification but can be obtained automatically. DL CAC scores improved classification of a significant proportion of patients as compared with SPECT myocardial perfusion alone.

Automated nonlinear registration of coronary PET to CT angiography using pseudo-CT generated from PET with generative adversarial networks

BACKGROUND

Coronary 18F-sodium-fluoride (18F-NaF) positron emission tomography (PET) showed promise in imaging coronary artery disease activity. Currently image processing remains subjective due to the need for manual registration of PET and computed tomography (CT) angiography data. We aimed to develop a novel fully automated method to register coronary 18F-NaF PET to CT angiography using pseudo-CT generated by generative adversarial networks (GAN).

METHODS

A total of 169 patients, 139 in the training and 30 in the testing sets were considered for generation of pseudo-CT from non-attenuation corrected (NAC) PET using GAN. Non-rigid registration was used to register pseudo-CT to CT angiography and the resulting transformation was used to align PET with CT angiography. We compared translations, maximal standard uptake value (SUVmax) and target to background ratio (TBRmax) at the location of plaques, obtained after observer and automated alignment.

RESULTS

Automatic end-to-end registration was performed for 30 patients with 88 coronary vessels and took 27.5 seconds per patient. Difference in displacement motion vectors between GAN-based and observer-based registration in the x-, y-, and z-directions was 0.8 ± 3.0, 0.7 ± 3.0, and 1.7 ± 3.9 mm, respectively. TBRmax had a coefficient of repeatability (CR) of 0.31, mean bias of 0.03 and narrow limits of agreement (LOA) (95% LOA: - 0.29 to 0.33). SUVmax had CR of 0.26, mean bias of 0 and narrow LOA (95% LOA: - 0.26 to 0.26).

CONCLUSION

Pseudo-CT generated by GAN are perfectly registered to PET can be used to facilitate quick and fully automated registration of PET and CT angiography.

Rapid quantification of COVID-19 pneumonia burden from computed tomography with convolutional long short-term memory networks

Purpose: Quantitative lung measures derived from computed tomography (CT) have been demonstrated to improve prognostication in coronavirus disease 2019 (COVID-19) patients but are not part of clinical routine because the required manual segmentation of lung lesions is prohibitively time consuming. We aim to automatically segment ground-glass opacities and high opacities (comprising consolidation and pleural effusion). Approach: We propose a new fully automated deep-learning framework for fast multi-class segmentation of lung lesions in COVID-19 pneumonia from both contrast and non-contrast CT images using convolutional long short-term memory (ConvLSTM) networks. Utilizing the expert annotations, model training was performed using five-fold cross-validation to segment COVID-19 lesions. The performance of the method was evaluated on CT datasets from 197 patients with a positive reverse transcription polymerase chain reaction test result for SARS-CoV-2, 68 unseen test cases, and 695 independent controls. Results: Strong agreement between expert manual and automatic segmentation was obtained for lung lesions with a Dice score of 0.89 ± 0.07 ; excellent correlations of 0.93 and 0.98 for ground-glass opacity (GGO) and high opacity volumes, respectively, were obtained. In the external testing set of 68 patients, we observed a Dice score of 0.89 ± 0.06 as well as excellent correlations of 0.99 and 0.98 for GGO and high opacity volumes, respectively. Computations for a CT scan comprising 120 slices were performed under 3 s on a computer equipped with an NVIDIA TITAN RTX GPU. Diagnostically, the automated quantification of the lung burden % discriminate COVID-19 patients from controls with an area under the receiver operating curve of 0.96 (0.95-0.98). Conclusions: Our method allows for the rapid fully automated quantitative measurement of the pneumonia burden from CT, which can be used to rapidly assess the severity of COVID-19 pneumonia on chest CT.

Deep Learning of Coronary Calcium Scores From PET/CT Attenuation Maps Accurately Predicts Adverse Cardiovascular Events

BACKGROUND

Assessment of coronary artery calcium (CAC) by computed tomographic (CT) imaging provides an accurate measure of atherosclerotic burden. CAC is also visible in computed tomographic attenuation correction (CTAC) scans, always acquired with cardiac positron emission tomographic (PET) imaging.

OBJECTIVES

The aim of this study was to develop a deep-learning (DL) model capable of fully automated CAC definition from PET CTAC scans.

METHODS

The novel DL model, originally developed for video applications, was adapted to rapidly quantify CAC. The model was trained using 9,543 expert-annotated CT scans and was tested in 4,331 patients from an external cohort undergoing PET/CT imaging with major adverse cardiac events (MACEs) (follow-up 4.3 years), including same-day paired electrocardiographically gated CAC scans available in 2,737 patients. MACE risk stratification in 4 CAC score categories (0, 1-100, 101-400, and >400) was analyzed and CAC scores derived from electrocardiographically gated CT scans (standard scores) by expert observers were compared with automatic DL scores from CTAC scans.

RESULTS

Automatic DL scoring required <6 seconds per scan. DL CTAC scores provided stepwise increase in the risk for MACE across the CAC score categories (HR up to 3.2; P < 0.001). Net reclassification improvement of standard CAC scores over DL CTAC scores was nonsignificant (-0.02; 95% CI: -0.11 to 0.07). The negative predictive values for MACE of zero CAC with standard (85%) and DL CTAC (83%) CAC scores were similar (P = 0.19).

CONCLUSIONS

DL CTAC scores predict cardiovascular risk similarly to standard CAC scores quantified manually by experienced operators from dedicated electrocardiographically gated CAC scans and can be obtained almost instantly, with no changes to PET/CT scanning protocol.

Deep Learning for Explainable Estimation of Mortality Risk From Myocardial Positron Emission Tomography Images

BACKGROUND

We aim to develop an explainable deep learning (DL) network for the prediction of all-cause mortality directly from positron emission tomography myocardial perfusion imaging flow and perfusion polar map data and evaluate it using prospective testing.

METHODS

A total of 4735 consecutive patients referred for stress and rest ⁸²Rb positron emission tomography between 2010 and 2018 were followed up for all-cause mortality for 4.15 (2.24-6.3) years. DL network utilized polar maps of stress and rest perfusion, myocardial blood flow, myocardial flow reserve, and spill-over fraction combined with cardiac volumes, singular indices, and sex. Patients scanned from 2010 to 2016 were used for training and validation. The network was tested in a set of 1135 patients scanned from 2017 to 2018 to simulate prospective clinical implementation.

RESULTS

In prospective testing, the area under the receiver operating characteristic curve for all-cause mortality prediction by DL (0.82 [95% CI, 0.77-0.86]) was higher than ischemia (0.60 [95% CI, 0.54-0.66]; P <0.001), myocardial flow reserve (0.70 [95% CI, 0.64-0.76], P <0.001) or a comprehensive logistic regression model (0.75 [95% CI, 0.69-0.80], P <0.05). The highest quartile of patients by DL had an annual all-cause mortality rate of 11.87% and had a 16.8 ([95% CI, 6.12%-46.3%]; P <0.001)-fold increase in the risk of death compared with the lowest quartile patients. DL showed a 21.6% overall reclassification improvement as compared with established measures of ischemia.

CONCLUSIONS

The DL model trained directly on polar maps allows improved patient risk stratification in comparison with established methods for positron emission tomography flow or perfusion assessments.

Radiomorphological signs and clinical severity of SARS-CoV-2 lineage B.1.1.7

Objective

We aimed to assess the differences in the severity and chest-CT radiomorphological signs of SARS-CoV-2 B.1.1.7 and non-B.1.1.7 variants.

Methods

We collected clinical data of consecutive patients with laboratory-confirmed COVID-19 and chest-CT imaging who were admitted to the Emergency Department between September 1- November 13, 2020 (non-B.1.1.7 cases) and March 1-March 18, 2021 (B.1.1.7 cases). We also examined the differences in the severity and radiomorphological features associated with COVID-19 pneumonia. Total pneumonia burden (%), mean attenuation of ground-glass opacities and consolidation were quantified using deep-learning research software.

Results

The final population comprised 500 B.1.1.7 and 500 non-B.1.1.7 cases. Patients with B.1.1.7 infection were younger (58.5 ± 15.6 vs 64.8 ± 17.3; p < .001) and had less comorbidities. Total pneumonia burden was higher in the B.1.1.7 patient group (16.1% [interquartile range (IQR):6.0-34.2%] vs 6.6% [IQR:1.2-18.3%]; p < .001). In the age-specific analysis, in patients <60 years B.1.1.7 pneumonia had increased consolidation burden (0.1% [IQR:0.0-0.7%] vs 0.1% [IQR:0.0-0.2%]; p < .001), and severe COVID-19 was more prevalent (11.5% vs  4.9%; p = .032). Mortality rate was similar in all age groups.

Conclusion

Despite B.1.1.7 patients were younger and had fewer comorbidities, they experienced more severe disease than non-B.1.1.7 patients, however, the risk of death was the same between the two groups.

Advances in knowledge

Our study provides data on deep-learning based quantitative lung lesion burden and clinical outcomes of patients infected by B.1.1.7 VOC. Our findings might serve as a model for later investigations, as new variants are emerging across the globe.

Deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction: an international multicentre study

BACKGROUND

Atherosclerotic plaque quantification from coronary CT angiography (CCTA) enables accurate assessment of coronary artery disease burden and prognosis. We sought to develop and validate a deep learning system for CCTA-derived measures of plaque volume and stenosis severity.

METHODS

This international, multicentre study included nine cohorts of patients undergoing CCTA at 11 sites, who were assigned into training and test sets. Data were retrospectively collected on patients with a wide range of clinical presentations of coronary artery disease who underwent CCTA between Nov 18, 2010, and Jan 25, 2019. A novel deep learning convolutional neural network was trained to segment coronary plaque in 921 patients (5045 lesions). The deep learning network was then applied to an independent test set, which included an external validation cohort of 175 patients (1081 lesions) and 50 patients (84 lesions) assessed by intravascular ultrasound within 1 month of CCTA. We evaluated the prognostic value of deep learning-based plaque measurements for fatal or non-fatal myocardial infarction (our primary outcome) in 1611 patients from the prospective SCOT-HEART trial, assessed as dichotomous variables using multivariable Cox regression analysis, with adjustment for the ASSIGN clinical risk score.

FINDINGS

In the overall test set, there was excellent or good agreement, respectively, between deep learning and expert reader measurements of total plaque volume (intraclass correlation coefficient [ICC] 0·964) and percent diameter stenosis (ICC 0·879; both p<0·0001). When compared with intravascular ultrasound, there was excellent agreement for deep learning total plaque volume (ICC 0·949) and minimal luminal area (ICC 0·904). The mean per-patient deep learning plaque analysis time was 5·65 s (SD 1·87) versus 25·66 min (6·79) taken by experts. Over a median follow-up of 4·7 years (IQR 4·0-5·7), myocardial infarction occurred in 41 (2·5%) of 1611 patients from the SCOT-HEART trial. A deep learning-based total plaque volume of 238·5 mm3 or higher was associated with an increased risk of myocardial infarction (hazard ratio [HR] 5·36, 95% CI 1·70-16·86; p=0·0042) after adjustment for the presence of deep learning-based obstructive stenosis (HR 2·49, 1·07-5·50; p=0·0089) and the ASSIGN clinical risk score (HR 1·01, 0·99-1·04; p=0·35).

INTERPRETATION

Our novel, externally validated deep learning system provides rapid measurements of plaque volume and stenosis severity from CCTA that agree closely with expert readers and intravascular ultrasound, and could have prognostic value for future myocardial infarction.

FUNDING

National Heart, Lung, and Blood Institute and the Miriam & Sheldon G Adelson Medical Research Foundation.

Calcium scoring in low-dose ungated chest CT scans using convolutional long-short term memory networks

We aimed to develop a novel deep-learning based method for automatic coronary artery calcium (CAC) quantification in low-dose ungated computed tomography attenuation correction maps (CTAC). In this study, we used convolutional long-short -term memory deep neural network (conv-LSTM) to automatically derive coronary artery calcium score (CAC) from both standard CAC scans and low-dose ungated scans (CT-attenuation correction maps). We trained convLSTM to segment CAC using 9543 scans. A U-Net model was trained as a reference method. Both models were validated in the OrCaCs dataset (n=32) and in the held-out cohort (n=507) without prior coronary interventions who had CTAC standard CAC scan acquired contemporarily. Cohen's kappa coefficients and concordance matrices were used to assess agreement in four CAC score categories (very low: <10, low:10-100; moderate:101-400 and high >400). The median time to derive results on a central processing unit (CPU) was significantly shorter for the conv-LSTM model- 6.18s (inter quartile range [IQR]: 5.99, 6.3) than for UNet (10.1s, IQR: 9.82, 15.9s, p<0.0001). The memory consumption during training was much lower for our model (13.11Gb) in comparison with UNet (22.31 Gb). Conv-LSTM performed comparably to UNet in terms of agreement with expert annotations, but with significantly shorter inference times and lower memory consumption.

Deep learning-based plaque quantification from coronary computed tomography angiography: external validation and comparison with intravascular ultrasound

Background

Atherosclerotic plaque quantification from coronary computed tomography angiography (CTA) enables accurate assessment of coronary artery disease burden, progression, and prognosis. However, quantitative plaque analysis is time-consuming and requires high expertise. We sought to develop and externally validate an artificial intelligence (AI)-based deep learning (DL) approach for CTA-derived measures of plaque volume and stenosis severity. We compared the performance of DL to expert readers and the gold standard of intravascular ultrasound (IVUS).

Methods

This was a multicenter study of patients undergoing coronary CTA at 11 sites, with software-based quantitative plaque measurements performed at a per-lesion level by expert readers. AI-based plaque analysis was performed by a DL novel convolutional neural network which automatically segmented the coronary artery wall, lumen, and plaque for the computation of plaque volume and stenosis severity. Using expert measurements as ground truth, the DL algorithm was trained on 887 patients (4,686 lesions). Thereafter, the algorithm was applied to an independent test set of 221 patients (1,234 lesions), which included an external validation cohort of 171 patients from the SCOT-HEART (Scottish Computed Tomography of the Heart) trial as well as 50 patients who underwent IVUS within one month of CTA. We report the performance of AI-based plaque analysis in the independent test set.

Results

Within the external validation cohort, there was excellent agreement between DL and expert reader measurements of total plaque volume (intraclass correlation coefficient [ICC] 0.876), noncalcified plaque volume (ICC 0.869), and percent diameter stenosis (ICC 0.850; all p&lt;0.001). When compared with IVUS, there was excellent agreement for DL total plaque volume (ICC 0.945), total plaque burden (ICC 0.853), minimal luminal area (ICC 0.864), and percent area stenosis (ICC 0.805; all p&lt;0.001); with strong correlation between DL and IVUS for total plaque volume (r=0.915; p&lt;0.001; Figure). The average DL plaque analysis time was 20 seconds per patient, compared with 25–30 minutes taken by experts.

Conclusions

AI-based plaque quantification from coronary CTA using an externally validated DL approach enables rapid measurements of plaque volume and stenosis severity in close agreement with expert readers and IVUS.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Heart, Lung, and Blood Institute, United States

Rapid quantification of COVID-19 pneumonia burden from computed tomography with convolutional LSTM networks

Quantitative lung measures derived from computed tomography (CT) have been demonstrated to improve prognostication in Coronavirus disease 2019 (COVID-19) patients, but are not part of the clinical routine since required manual segmentation of lung lesions is prohibitively time-consuming. We propose a new fully automated deep learning framework for quantification and differentiation between lung lesions in COVID-19 pneumonia from both contrast and non-contrast CT images using convolutional Long Short-Term Memory (LSTM) networks. Utilizing the expert annotations, model training was performed using 5-fold cross-validation to segment ground-glass opacity and high opacity (including consolidation and pleural effusion). The performance of the method was evaluated on CT data sets from 197 patients with positive reverse transcription polymerase chain reaction test result for SARS-CoV-2. Strong agreement between expert manual and automatic segmentation was obtained for lung lesions with a Dice score coefficient of 0.876 ± 0.005; excellent correlations of 0.978 and 0.981 for ground-glass opacity and high opacity volumes. In the external validation set of 67 patients, there was dice score coefficient of 0.767 ± 0.009 as well as excellent correlations of 0.989 and 0.996 for ground-glass opacity and high opacity volumes. Computations for a CT scan comprising 120 slices were performed under 2 seconds on a personal computer equipped with NVIDIA Titan RTX graphics processing unit. Therefore, our deep learning-based method allows rapid fully-automated quantitative measurement of pneumonia burden from CT and may generate the big data with an accuracy similar to the expert readers.