Using Deep learning to Predict Cardiovascular Magnetic Resonance Findings from Echocardiography Videos.

Background: Echocardiography is the most common modality for assessing cardiac structure and function. While cardiac magnetic resonance (CMR) imaging is less accessible, CMR can provide unique tissue characterization including late gadolinium enhancement (LGE), T1 and T2 mapping, and extracellular volume (ECV) which are associated with tissue fibrosis, infiltration, and inflammation. While deep learning has been shown to uncover findings not recognized by clinicians, it is unknown whether CMR-based tissue characteristics can be derived from echocardiography videos using deep learning. We hypothesized that deep learning applied to echocardiography could predict CMR-based measurements. Methods: In a retrospective single-center study, adult patients with CMRs and echocardiography studies within 30 days were included. A video-based convolutional neural network was trained on echocardiography videos to predict CMR-derived labels including wall motion abnormality (WMA) presence, LGE presence, and abnormal T1, T2 or ECV across echocardiography views. The model performance was evaluated in a held-out test dataset not used for training. Results: The study population included 1,453 adult patients (mean age 56±18 years, 42% female) with 2,556 paired echocardiography studies occurring on average 2 days after CMR (interquartile range 2 days prior to 6 days after). The model had high predictive capability for presence of WMA (AUC 0.873 [95%CI 0.816-0.922]), however, the model was unable to reliably detect the presence of LGE (AUC 0.699 [0.613-0.780]), native T1 (AUC 0.614 [0.500-0.715]), T2 0.553 [0.420-0.692], or ECV 0.564 [0.455-0.691]). Conclusions: Deep learning applied to echocardiography accurately identified CMR-based WMA, but was unable to predict tissue characteristics, suggesting that signal for these tissue characteristics may not be present within ultrasound videos, and that the use of CMR for tissue characterization remains essential within cardiology. Clinical Perspective: Tissue characterization of the heart muscle is useful for clinical diagnosis and prognosis by identifying myocardial fibrosis, inflammation, and infiltration, and can be measured using cardiac MRI. While echocardiography is highly accessible and provides excellent functional information, its ability to provide tissue characterization information is limited at this time. Our study using a deep learning approach to predict cardiac MRI-based tissue characteristics from echocardiography showed limited ability to do so, suggesting that alternative approaches, including non-deep learning methods should be considered in future research.

Alternating low-rank tensor reconstruction for improved multiparametric mapping with cardiovascular MR Multitasking.

PURPOSE: To develop a novel low-rank tensor reconstruction approach leveraging the complete acquired data set to improve precision and repeatability of multiparametric mapping within the cardiovascular MR Multitasking framework. METHODS: A novel approach that alternated between estimation of temporal components and spatial components using the entire data set acquired (i.e., including navigator data and imaging data) was developed to improve reconstruction. The precision and repeatability of the proposed approach were evaluated on numerical simulations, 10 healthy subjects, and 10 cardiomyopathy patients at multiple scan times for 2D myocardial T1/T2 mapping with MR Multitasking and were compared with those of the previous navigator-derived fixed-basis approach. RESULTS: In numerical simulations, the proposed approach outperformed the previous fixed-basis approach with lower T1 and T2 error against the ground truth at all scan times studied and showed better motion fidelity. In human subjects, the proposed approach showed no significantly different sharpness or T1/T2 measurement and significantly improved T1 precision by 20%-25%, T2 precision by 10%-15%, T1 repeatability by about 30%, and T2 repeatability by 25%-35% at 90-s and 50-s scan times The proposed approach at the 50-s scan time also showed comparable results with that of the previous fixed-basis approach at the 90-s scan time. CONCLUSION: The proposed approach improved precision and repeatability for quantitative imaging with MR Multitasking while maintaining comparable motion fidelity, T1/T2 measurement, and septum sharpness and had the potential for further reducing scan time from 90 s to 50 s.

High-resolution Diffusion-weighted Imaging to Detect Changes in Tumor Size and ADC, and Predict Adverse Biopsy Histology during Prostate Cancer Active Surveillance.

PURPOSE: Majority of men with low-risk prostate cancer can be managed with active surveillance (AS). This study evaluates a high-resolution diffusion-weighted imaging (HR-DWI) technique to predict adverse biopsy histology (AH), defined as Gleason score ≥7 on any biopsy or ≥3 increase in number of positive biopsy cores on systematic biopsies. We test the hypothesis that high-grade disease and progressing disease undergo subtle changes during even short intervals that can be detected by HR-DWI. EXPERIMENTAL DESIGN: In a prospective clinical trial, serial multiparametric MRIs, incorporating HR-DWI and standard DWI (S-DWI) were performed approximately 12 months apart prior to prostate biopsy (n = 59). HR-DWI, which uses reduced field-of-view and motion compensation techniques, was compared with S-DWI. RESULTS: HR-DWI had a 3-fold improvement in spacial resolution compared with S-DWI as confirmed using imaging phantoms. For detecting AH, multiparametric MRI using HR-DWI had a sensitivity of 75% and specificity of 83.9%, and MRI using S-DWI had a sensitivity of 71.4% and specificity of 54.8%. The AUC for HR-DWI was significantly higher (0.794 vs. 0.631, P = 0.014). Secondary analyses of univariable predictors of AH showed tumor size increase [OR 16.8; 95% confidence interval (CI): 4.06-69.48; P < 0.001] and apparent diffusion coefficient (ADC) decrease (OR 5.06; 95% CI: 1.39-18.38; P = 0.014) on HR-DWI were significant predictors of AH. CONCLUSION: HR-DWI outperforms S-DWI in predicting AH. Patient with AH have tumors that change in size and ADC that could be detected using HR-DWI. Future studies with longer follow-up should assess HR-DWI for predicting disease progression during AS. SIGNIFICANCE: We report on a prospective clinical trial using a MRI that has three times the resolution of standard MRI. During AS for prostate cancer, two high-resolution MRIs performed approximately a year apart can detect tumor changes that predict the presence of aggressive cancers that should be considered for curative therapy such as prostatectomy or radiation.

Enhancing Whole-Brain Magnetic Field Homogeneity for 3D-Magnetic Resonance Spectroscopic Imaging with a Novel Unified Coil: A Preliminary Study.

The spectral quality of magnetic resonance spectroscopic imaging (MRSI) can be affected by strong magnetic field inhomogeneities, posing a challenge for 3D-MRSI's widespread clinical use with standard scanner-equipped 2nd-order shim coils. To overcome this, we designed an empirical unified shim-RF head coil (32-ch RF receive and 51-ch shim) for 3D-MRSI improvement. We compared its shimming performance and 3D-MRSI brain coverages against the standard scanner shim (2nd-order spherical harmonic (SH) shim coils) and integrated parallel reception, excitation, and shimming (iPRES) 32-ch AC/DC head coil. We also simulated a theoretical 3rd-, 4th-, and 5th-order SH shim as a benchmark to assess the UNIfied shim-RF coil (UNIC) improvements. In this preliminary study, the whole-brain coverage was simulated by using B0 field maps of twenty-four healthy human subjects (n = 24). Our results demonstrated that UNIC substantially improves brain field homogeneity, reducing whole-brain frequency standard deviations by 27% compared to the standard 2nd-order scanner shim and 17% compared to the iPRES shim. Moreover, UNIC enhances whole-brain coverage of 3D-MRSI by up to 34% compared to the standard 2nd-order scanner shim and up to 13% compared to the iPRES shim. UNIC markedly increases coverage in the prefrontal cortex by 147% and 47% and in the medial temporal lobe and temporal pole by 29% and 13%, respectively, at voxel resolutions of 1.4 cc and 0.09 cc for 3D-MRSI. Furthermore, UNIC effectively reduces variations in shim quality and brain coverage among different subjects compared to scanner shim and iPRES shim. Anticipated advancements in higher-order shimming (beyond 6th order) are expected via optimized designs using dimensionality reduction methods.

AI-luminating Artificial Intelligence in Inflammatory Bowel Diseases: A Narrative Review on the Role of AI in Endoscopy, Histology, and Imaging for IBD.

Endoscopy, histology, and cross-sectional imaging serve as fundamental pillars in the detection, monitoring, and prognostication of inflammatory bowel disease (IBD). However, interpretation of these studies often relies on subjective human judgment, which can lead to delays, intra- and interobserver variability, and potential diagnostic discrepancies. With the rising incidence of IBD globally coupled with the exponential digitization of these data, there is a growing demand for innovative approaches to streamline diagnosis and elevate clinical decision-making. In this context, artificial intelligence (AI) technologies emerge as a timely solution to address the evolving challenges in IBD. Early studies using deep learning and radiomics approaches for endoscopy, histology, and imaging in IBD have demonstrated promising results for using AI to detect, diagnose, characterize, phenotype, and prognosticate IBD. Nonetheless, the available literature has inherent limitations and knowledge gaps that need to be addressed before AI can transition into a mainstream clinical tool for IBD. To better understand the potential value of integrating AI in IBD, we review the available literature to summarize our current understanding and identify gaps in knowledge to inform future investigations.

Artificial intelligence in liver cancer research: a scientometrics analysis of trends and topics.

Background and aims: With the rapid growth of artificial intelligence (AI) applications in various fields, understanding its impact on liver cancer research is paramount. This scientometrics project aims to investigate publication trends and topics in AI-related publications in liver cancer. Materials and Methods: We employed a search strategy to identify AI-related publications in liver cancer using Scopus database. We analyzed the number of publications, author affiliations, and journals that publish AI-related publications in liver cancer. Finally, the publications were grouped based on intended application. Results: We identified 3950 eligible publications (2695 articles, 366 reviews, and 889 other document types) from 1968 to August 3, 2023. There was a 12.7-fold increase in AI-related publications from 2013 to 2022. By comparison, the number of total publications on liver cancer increased by 1.7-fold. Our analysis revealed a significant shift in trends of AI-related publications on liver cancer in 2019. We also found a statistically significant consistent increase in numbers of AI-related publications over time (tau = 0.756, p < 0.0001). Eight (53%) of the top 15 journals with the most publications were radiology journals. The largest number of publications were from China (n=1156), the US (n=719), and Germany (n=236). The three most common publication categories were "medical image analysis for diagnosis" (37%), "diagnostic or prognostic biomarkers modeling & bioinformatics" (19%), and "genomic or molecular analysis" (18%). Conclusion: Our study reveals increasing interest in AI for liver cancer research, evidenced by a 12.7-fold growth in related publications over the past decade. A common application of AI is in medical imaging analysis for various purposes. China, the US, and Germany are leading contributors.

The Disproportionate Rise in Pancreatic Cancer in Younger Women Is Due to a Rise in Adenocarcinoma and Not Neuroendocrine Tumors: A Nationwide Time-Trend Analysis Using 2001-2018 United States Cancer Statistics Databases.

In previous studies, a significant increase in the incidence of pancreatic cancer among younger women compared to men in the United States was noted. However, the specific histopathologic characteristics were not delineated. This population-based study aimed to assess whether this disproportionate rise in pancreatic cancer in younger women was contributed by pancreatic ductal adenocarcinoma (PDAC) or pancreatic neuroendocrine tumors (PanNET). The United States Cancer Statistics (USCS) database was used to identify patients with pancreatic cancer between 2001 and 2018. The results showed that, in younger adults, the incidence of PDAC has increased in women [average annual percentage change (AAPC) = 0.62%], while it has remained stable in men (AAPC = -0.09%). The PDAC incidence rate among women increased at a greater rate compared to men with a statistically significant difference in AAPC (p < 0.001), with neither identical nor parallel trends. In contrast, cases of PanNET did not demonstrate a statistically significant sex-specific AAPC difference. In conclusion, this study demonstrated that the dramatic increase in the incidence rate of PDAC explains the disproportionate rise in pancreatic cancer incidence in younger women. This prompts further prospective studies to investigate the underlying reasons for these sex-specific disparities in PDAC.

Non-electrocardiogram-gated, free-breathing, off-resonance reduced, high-resolution, whole-heart myocardial T2 * mapping at 3 T within 5 min.

PURPOSE: Widely used conventional 2D T2 * approaches that are based on breath-held, electrocardiogram (ECG)-gated, multi-gradient-echo sequences are prone to motion artifacts in the presence of incomplete breath holding or arrhythmias, which is common in cardiac patients. To address these limitations, a 3D, non-ECG-gated, free-breathing T2 * technique that enables rapid whole-heart coverage was developed and validated. METHODS: A continuous random Gaussian 3D k-space sampling was implemented using a low-rank tensor framework for motion-resolved 3D T2 * imaging. This approach was tested in healthy human volunteers and in swine before and after intravenous administration of ferumoxytol. RESULTS: Spatial-resolution matched T2 * images were acquired with 2-3-fold reduction in scan time using the proposed T2 * mapping approach relative to conventional T2 * mapping. Compared with the conventional approach, T2 * images acquired with the proposed method demonstrated reduced off-resonance and flow artifacts, leading to higher image quality and lower coefficient of variation in T2 *-weighted images of the myocardium of swine and humans. Mean myocardial T2 * values determined using the proposed and conventional approaches were highly correlated and showed minimal bias. CONCLUSION: The proposed non-ECG-gated, free-breathing, 3D T2 * imaging approach can be performed within 5 min or less. It can overcome critical image artifacts from undesirable cardiac and respiratory motion and bulk off-resonance shifts at the heart-lung interface. The proposed approach is expected to facilitate faster and improved cardiac T2 * mapping in those with limited breath-holding capacity or arrhythmias.

Rapid three-dimensional quantification of high-intensity plaques from coronary atherosclerosis T1-weighted characterization to predict periprocedural myocardial injury.

BACKGROUND: High-intensity plaque (HIP) on magnetic resonance imaging (MRI) has been documented as a powerful predictor of periprocedural myocardial injury (PMI) following percutaneous coronary intervention (PCI). Despite the recent proposal of three-dimensional HIP quantification to enhance the predictive capability, the conventional pulse sequence, which necessitates the separate acquisition of anatomical reference images, hinders accurate three-dimensional segmentation along the coronary vasculature. Coronary atherosclerosis T1-weighted characterization (CATCH) enables the simultaneous acquisition of inherently coregistered dark-blood plaque and bright-blood coronary artery images. We aimed to develop a novel HIP quantification approach using CATCH and to ascertain its superior predictive performance compared to the conventional two-dimensional assessment based on plaque-to-myocardium signal intensity ratio (PMR). METHODS: In this prospective study, CATCH MRI was conducted before elective stent implantation in 137 lesions from 125 patients. On CATCH images, dedicated software automatically generated tubular three-dimensional volumes of interest on the dark-blood plaque images along the coronary vasculature, based on the precisely matched bright-blood coronary artery images, and subsequently computed PMR and HIP volume (HIPvol). Specifically, HIPvol was calculated as the volume of voxels with signal intensity exceeding that of the myocardium, weighted by their respective signal intensities. PMI was defined as post-PCI cardiac troponin-T > 5 × the upper reference limit. RESULTS: The entire analysis process was completed within 3 min per lesion. PMI occurred in 44 lesions. Based on the receiver operating characteristic curve analysis, HIPvol outperformed PMR for predicting PMI (C-statistics, 0.870 [95% CI, 0.805-0.936] vs. 0.787 [95% CI, 0.706-0.868]; p = 0.001). This result was primarily driven by the higher sensitivity HIPvol offered: 0.886 (95% CI, 0.754-0.962) vs. 0.750 for PMR (95% CI, 0.597-0.868; p = 0.034). Multivariable analysis identified HIPvol as an independent predictor of PMI (odds ratio, 1.15 per 10-µL increase; 95% CI, 1.01-1.30, p = 0.035). CONCLUSIONS: Our semi-automated method of analyzing coronary plaque using CATCH MRI provided rapid HIP quantification. Three-dimensional assessment using this approach had a better ability to predict PMI than conventional two-dimensional assessment.

Impact of Measurement Noise on Genetic Association Studies of Cardiac Function.

Recent research has effectively used quantitative traits from imaging to boost the capabilities of genome-wide association studies (GWAS), providing further understanding of disease biology and various traits. However, it's important to note that phenotyping inherently carries measurement error and noise that could influence subsequent genetic analyses. The study focused on left ventricular ejection fraction (LVEF), a vital yet potentially inaccurate quantitative measurement, to investigate how imprecision in phenotype measurement affects genetic studies. Several methods of acquiring LVEF, along with simulating measurement noise, were assessed for their effects on ensuing genetic analyses. The results showed that by introducing just 7.9% of measurement noise, all genetic associations in an LVEF GWAS with almost forty thousand individuals could be eliminated. Moreover, a 1% increase in mean absolute error (MAE) in LVEF had an effect equivalent to a 10% reduction in the sample size of the cohort on the power of GWAS. Therefore, enhancing the accuracy of phenotyping is crucial to maximize the effectiveness of genome-wide association studies.

Retrospective T2 quantification from conventional weighted MRI of the prostate based on deep learning.

Purpose: To develop a deep learning-based method to retrospectively quantify T2 from conventional T1- and T2-weighted images. Methods: Twenty-five subjects were imaged using a multi-echo spin-echo sequence to estimate reference prostate T2 maps. Conventional T1- and T2-weighted images were acquired as the input images. A U-Net based neural network was developed to directly estimate T2 maps from the weighted images using a four-fold cross-validation training strategy. The structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), mean percentage error (MPE), and Pearson correlation coefficient were calculated to evaluate the quality of network-estimated T2 maps. To explore the potential of this approach in clinical practice, a retrospective T2 quantification was performed on a high-risk prostate cancer cohort (Group 1) and a low-risk active surveillance cohort (Group 2). Tumor and non-tumor T2 values were evaluated by an experienced radiologist based on region of interest (ROI) analysis. Results: The T2 maps generated by the trained network were consistent with the corresponding reference. Prostate tissue structures and contrast were well preserved, with a PSNR of 26.41 ± 1.17 dB, an SSIM of 0.85 ± 0.02, and a Pearson correlation coefficient of 0.86. Quantitative ROI analyses performed on 38 prostate cancer patients revealed estimated T2 values of 80.4 ± 14.4 ms and 106.8 ± 16.3 ms for tumor and non-tumor regions, respectively. ROI measurements showed a significant difference between tumor and non-tumor regions of the estimated T2 maps (P < 0.001). In the two-timepoints active surveillance cohort, patients defined as progressors exhibited lower estimated T2 values of the tumor ROIs at the second time point compared to the first time point. Additionally, the T2 difference between two time points for progressors was significantly greater than that for non-progressors (P = 0.010). Conclusion: A deep learning method was developed to estimate prostate T2 maps retrospectively from clinically acquired T1- and T2-weighted images, which has the potential to improve prostate cancer diagnosis and characterization without requiring extra scans.

Retrospective quantification of clinical abdominal DCE-MRI using pharmacokinetics-informed deep learning: a proof-of-concept study.

Introduction: Dynamic contrast-enhanced (DCE) MRI has important clinical value for early detection, accurate staging, and therapeutic monitoring of cancers. However, conventional multi-phasic abdominal DCE-MRI has limited temporal resolution and provides qualitative or semi-quantitative assessments of tissue vascularity. In this study, the feasibility of retrospectively quantifying multi-phasic abdominal DCE-MRI by using pharmacokinetics-informed deep learning to improve temporal resolution was investigated. Method: Forty-five subjects consisting of healthy controls, pancreatic ductal adenocarcinoma (PDAC), and chronic pancreatitis (CP) were imaged with a 2-s temporal-resolution quantitative DCE sequence, from which 30-s temporal-resolution multi-phasic DCE-MRI was synthesized based on clinical protocol. A pharmacokinetics-informed neural network was trained to improve the temporal resolution of the multi-phasic DCE before the quantification of pharmacokinetic parameters. Through ten-fold cross-validation, the agreement between pharmacokinetic parameters estimated from synthesized multi-phasic DCE after deep learning inference was assessed against reference parameters from the corresponding quantitative DCE-MRI images. The ability of the deep learning estimated parameters to differentiate abnormal from normal tissues was assessed as well. Results: The pharmacokinetic parameters estimated after deep learning have a high level of agreement with the reference values. In the cross-validation, all three pharmacokinetic parameters (transfer constant Ktrans, fractional extravascular extracellular volume ve, and rate constant kep) achieved intraclass correlation coefficient and R2 between 0.84-0.94, and low coefficients of variation (10.1%, 12.3%, and 5.6%, respectively) relative to the reference values. Significant differences were found between healthy pancreas, PDAC tumor and non-tumor, and CP pancreas. Discussion: Retrospective quantification (RoQ) of clinical multi-phasic DCE-MRI is possible by deep learning. This technique has the potential to derive quantitative pharmacokinetic parameters from clinical multi-phasic DCE data for a more objective and precise assessment of cancer.

Quantification of carotid plaque composition with a multi-contrast atherosclerosis characterization (MATCH) MRI sequence.

Background and purpose: Carotid atherosclerotic plaques with a large lipid-rich necrotic core (LRNC), intraplaque hemorrhage (IPH), and a thin or ruptured fibrous cap are associated with increased stroke risk. Multi-sequence MRI can be used to quantify carotid atherosclerotic plaque composition. Yet, its clinical implementation is hampered by long scan times and image misregistration. Multi-contrast atherosclerosis characterization (MATCH) overcomes these limitations. This study aims to compare the quantification of plaque composition with MATCH and multi-sequence MRI. Methods: MATCH and multi-sequence MRI were used to image 54 carotid arteries of 27 symptomatic patients with ≥2 mm carotid plaque on a 3.0 T MRI scanner. The following sequence parameters for MATCH were used: repetition time/echo time (TR/TE), 10.1/4.35 ms; field of view, 160 mm × 160 mm × 2 mm; matrix size, 256 × 256; acquired in-plane resolution, 0.63 mm2× 0.63 mm2; number of slices, 18; and flip angles, 8°, 5°, and 10°. Multi-sequence MRI (black-blood pre- and post-contrast T1-weighted, time of flight, and magnetization prepared rapid acquisition gradient echo; acquired in-plane resolution: 0.63 mm2 × 0.63 mm2) was acquired according to consensus recommendations, and image quality was scored (5-point scale). The interobserver agreement in plaque composition quantification was assessed by the intraclass correlation coefficient (ICC). The sensitivity and specificity of MATCH in identifying plaque composition were calculated using multi-sequence MRI as a reference standard. Results: A significantly lower image quality of MATCH compared to that of multi-sequence MRI was observed (p < 0.05). The scan time for MATCH was shorter (7 vs. 40 min). Interobserver agreement in quantifying plaque composition on MATCH images was good to excellent (ICC ≥ 0.77) except for the total volume of calcifications and fibrous tissue that showed moderate agreement (ICC ≥ 0.61). The sensitivity and specificity of detecting plaque components on MATCH were ≥89% and ≥91% for IPH, ≥81% and 85% for LRNC, and ≥71% and ≥32% for calcifications, respectively. Overall, good-to-excellent agreement (ICC ≥ 0.76) of quantifying plaque components on MATCH with multi-sequence MRI as the reference standard was observed except for calcifications (ICC = 0.37-0.38) and fibrous tissue (ICC = 0.59-0.70). Discussion and conclusion: MATCH images can be used to quantify plaque components such as LRNC and IPH but not for calcifications. Although MATCH images showed a lower mean image quality score, short scan time and inherent co-registration are significant advantages.

Physics-informed deep learning for T2-deblurred superresolution turbo spin echo MRI.

PURPOSE: Deep learning superresolution (SR) is a promising approach to reduce MRI scan time without requiring custom sequences or iterative reconstruction. Previous deep learning SR approaches have generated low-resolution training images by simple k-space truncation, but this does not properly model in-plane turbo spin echo (TSE) MRI resolution degradation, which has variable T2 relaxation effects in different k-space regions. To fill this gap, we developed a T2 -deblurred deep learning SR method for the SR of 3D-TSE images. METHODS: A SR generative adversarial network was trained using physically realistic resolution degradation (asymmetric T2 weighting of raw high-resolution k-space data). For comparison, we trained the same network structure on previous degradation models without TSE physics modeling. We tested all models for both retrospective and prospective SR with 3 × 3 acceleration factor (in the two phase-encoding directions) of genetically engineered mouse embryo model TSE-MR images. RESULTS: The proposed method can produce high-quality 3 × 3 SR images for a typical 500-slice volume with 6-7 mouse embryos. Because 3 × 3 SR was performed, the image acquisition time can be reduced from 15 h to 1.7 h. Compared to previous SR methods without TSE modeling, the proposed method achieved the best quantitative imaging metrics for both retrospective and prospective evaluations and achieved the best imaging-quality expert scores for prospective evaluation. CONCLUSION: The proposed T2 -deblurring method improved accuracy and image quality of deep learning-based SR of TSE MRI. This method has the potential to accelerate TSE image acquisition by a factor of up to 9.

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  ±â€¯ 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.

Direct synthesis of multi-contrast brain MR images from MR multitasking spatial factors using deep learning.

PURPOSE: To develop a deep learning method to synthesize conventional contrast-weighted images in the brain from MR multitasking spatial factors. METHODS: Eighteen subjects were imaged using a whole-brain quantitative T1 -T2 -T1ρ MR multitasking sequence. Conventional contrast-weighted images consisting of T1 MPRAGE, T1 gradient echo, and T2 fluid-attenuated inversion recovery were acquired as target images. A 2D U-Net-based neural network was trained to synthesize conventional weighted images from MR multitasking spatial factors. Quantitative assessment and image quality rating by two radiologists were performed to evaluate the quality of deep-learning-based synthesis, in comparison with Bloch-equation-based synthesis from MR multitasking quantitative maps. RESULTS: The deep-learning synthetic images showed comparable contrasts of brain tissues with the reference images from true acquisitions and were substantially better than the Bloch-equation-based synthesis results. Averaging on the three contrasts, the deep learning synthesis achieved normalized root mean square error = 0.184 ± 0.075, peak SNR = 28.14 ± 2.51, and structural-similarity index = 0.918 ± 0.034, which were significantly better than Bloch-equation-based synthesis (p < 0.05). Radiologists' rating results show that compared with true acquisitions, deep learning synthesis had no notable quality degradation and was better than Bloch-equation-based synthesis. CONCLUSION: A deep learning technique was developed to synthesize conventional weighted images from MR multitasking spatial factors in the brain, enabling the simultaneous acquisition of multiparametric quantitative maps and clinical contrast-weighted images in a single scan.

Deep learning-enabled analysis of medical images identifies cardiac sphericity as an early marker of cardiomyopathy and related outcomes.

BACKGROUND: Quantification of chamber size and systolic function is a fundamental component of cardiac imaging. However, the human heart is a complex structure with significant uncharacterized phenotypic variation beyond traditional metrics of size and function. Examining variation in cardiac shape can add to our ability to understand cardiovascular risk and pathophysiology. METHODS: We measured the left ventricle (LV) sphericity index (short axis length/long axis length) using deep learning-enabled image segmentation of cardiac magnetic resonance imaging data from the UK Biobank. Subjects with abnormal LV size or systolic function were excluded. The relationship between LV sphericity and cardiomyopathy was assessed using Cox analyses, genome-wide association studies, and two-sample Mendelian randomization. FINDINGS: In a cohort of 38,897 subjects, we show that a one standard deviation increase in sphericity index is associated with a 47% increased incidence of cardiomyopathy (hazard ratio [HR]: 1.47, 95% confidence interval [CI]: 1.10-1.98, p = 0.01) and a 20% increased incidence of atrial fibrillation (HR: 1.20, 95% CI: 1.11-1.28, p < 0.001), independent of clinical factors and traditional magnetic resonance imaging (MRI) measurements. We identify four loci associated with sphericity at genome-wide significance, and Mendelian randomization supports non-ischemic cardiomyopathy as causal for LV sphericity. CONCLUSIONS: Variation in LV sphericity in otherwise normal hearts predicts risk for cardiomyopathy and related outcomes and is caused by non-ischemic cardiomyopathy. FUNDING: This study was supported by grants K99-HL157421 (D.O.) and KL2TR003143 (S.L.C.) from the National Institutes of Health.

Increasing Pancreatic Cancer Incidence in Young Women in the United States: A Population-Based Time-Trend Analysis, 2001-2018.

BACKGROUND & AIMS: Previous studies have shown an increasing incidence of pancreatic cancer (PC), especially in younger women; however, this has not been externally validated. In addition, there are limited data about contributing factors to this trend. We report age and sex-specific time-trend analysis of PC age-adjusted incidence rates (aIRs) using the National Program of Cancer Registries database without Surveillance Epidemiology and End Results data. METHODS: PC aIR, mortality rates, annual percentage change, and average annual percentage change (AAPC) were calculated and assessed for parallelism and identicalness. Age-specific analyses were conducted in older (≥55 years) and younger (<55 years) adults. PC incidence based on demographics, tumor characteristics, and mortality were evaluated in younger adults. RESULTS: A total of 454,611 patients were diagnosed with PC between 2001 and 2018 with significantly increasing aIR in women (AAPC = 1.27%) and men (AAPC = 1.14%) without a difference (P = .37). Similar results were seen in older adults. However, in younger adults (53,051 cases; 42.9% women), women experienced a greater increase in aIR than men (AAPCs = 2.36%, P < .001 vs 0.62%, P = 0.62) with nonparallel trends (P < .001) and AAPC difference of 1.74% (P < .001). This AAPC difference appears to be due to rising aIR in Blacks (2.23%; P < .001), adenocarcinoma histopathologic subtype (0.89%; P = .003), and location in the head-of-pancreas (1.64%; P < .001). PC mortality was found to be unchanged in women but decreasing in counterpart men (AAPC difference = 0.54%; P = .001). CONCLUSION: Using nationwide data, covering ≈64.5% of the U.S. population, we externally validate a rapidly increasing aIR of PC in younger women. There was a big separation of the incidence trend between women and men aged 15-34 years between 2001 and 2018 (>200% difference), and it did not show slowing down.

Pancreatic Cancer Incidence Trends by Race, Ethnicity, Age and Sex in the United States: A Population-Based Study, 2000-2018.

BACKGROUND AND AIMS: Pancreatic cancer (PC) incidence is increasing at a greater rate in young women compared to young men. We performed a race- and ethnicity-specific evaluation of incidence trends in subgroups stratified by age and sex to investigate the association of race and ethnicity with these trends. METHODS: Age-adjusted PC incidence rates (IR) from the years 2000 to 2018 were obtained from the SEER 21 database. Non-Hispanic White (White), Non-Hispanic Black (Black) and Hispanic patients were included. Age categories included older (ages ≥ 55) and younger (ages < 55) adults. Time-trends were described as annual percentage change (APC) and average APC (AAPC). RESULTS: Younger White [AAPC difference = 0.73, p = 0.01)], Black [AAPC difference = 1.96, p = 0.01)] and Hispanic [AAPC difference = 1.55, p = 0.011)] women experienced a greater rate of increase in IR compared to their counterpart men. Younger Hispanic women experienced a greater rate of increase in IR compared to younger Black women [AAPC difference = -1.28, p = 0.028)] and younger White women [AAPC difference = -1.35, p = 0.011)]. CONCLUSION: Younger women of all races and ethnicities experienced a greater rate of increase in PC IR compared to their counterpart men; however, younger Hispanic and Black women experienced a disproportionately greater increase. Hispanic women experienced a greater rate of increase in IR compared to younger Black and White women.

Impact of Measurement Imprecision on Genetic Association Studies of Cardiac Function.

Background: Recent studies have leveraged quantitative traits from imaging to amplify the power of genome-wide association studies (GWAS) to gain further insights into the biology of diseases and traits. However, measurement imprecision is intrinsic to phenotyping and can impact downstream genetic analyses. Methods: Left ventricular ejection fraction (LVEF), an important but imprecise quantitative imaging measurement, was examined to assess the impact of precision of phenotype measurement on genetic studies. Multiple approaches to obtain LVEF, as well as simulated measurement noise, were evaluated with their impact on downstream genetic analyses. Results: Even within the same population, small changes in the measurement of LVEF drastically impacted downstream genetic analyses. Introducing measurement noise as little as 7.9% can eliminate all significant genetic associations in an GWAS with almost forty thousand individuals. An increase of 1% in mean absolute error (MAE) in LVEF had an equivalent impact on GWAS power as a decrease of 10% in the cohort sample size, suggesting optimizing phenotyping precision is a cost-effective way to improve power of genetic studies. Conclusions: Improving the precision of phenotyping is important for maximizing the yield of genome-wide association studies.