Deformation-encoding Deep Learning Transformer for High-Frame-Rate Cardiac Cine MRI.

Purpose To develop a deep learning model for increasing cardiac cine frame rate while maintaining spatial resolution and scan time. Materials and Methods A transformer-based model was trained and tested on a retrospective sample of cine images from 5840 patients (mean age, 55 years ± 19 [SD]; 3527 male patients) referred for clinical cardiac MRI from 2003 to 2021 at nine centers; images were acquired using 1.5- and 3-T scanners from three vendors. Data from three centers were used for training and testing (4:1 ratio). The remaining data were used for external testing. Cines with downsampled frame rates were restored using linear, bicubic, and model-based interpolation. The root mean square error between interpolated and original cine images was modeled using ordinary least squares regression. In a prospective study of 49 participants referred for clinical cardiac MRI (mean age, 56 years ± 13; 25 male participants) and 12 healthy participants (mean age, 51 years ± 16; eight male participants), the model was applied to cines acquired at 25 frames per second (fps), thereby doubling the frame rate, and these interpolated cines were compared with actual 50-fps cines. The preference of two readers based on perceived temporal smoothness and image quality was evaluated using a noninferiority margin of 10%. Results The model generated artifact-free interpolated images. Ordinary least squares regression analysis accounting for vendor and field strength showed lower error (P < .001) with model-based interpolation compared with linear and bicubic interpolation in internal and external test sets. The highest proportion of reader choices was "no preference" (84 of 122) between actual and interpolated 50-fps cines. The 90% CI for the difference between reader proportions favoring collected (15 of 122) and interpolated (23 of 122) high-frame-rate cines was -0.01 to 0.14, indicating noninferiority. Conclusion A transformer-based deep learning model increased cardiac cine frame rates while preserving both spatial resolution and scan time, resulting in images with quality comparable to that of images obtained at actual high frame rates. Keywords: Functional MRI, Heart, Cardiac, Deep Learning, High Frame Rate Supplemental material is available for this article. © RSNA, 2024.

Present and Future Innovations in AI and Cardiac MRI.

Cardiac MRI is used to diagnose and treat patients with a multitude of cardiovascular diseases. Despite the growth of clinical cardiac MRI, complicated image prescriptions and long acquisition protocols limit the specialty and restrain its impact on the practice of medicine. Artificial intelligence (AI)-the ability to mimic human intelligence in learning and performing tasks-will impact nearly all aspects of MRI. Deep learning (DL) primarily uses an artificial neural network to learn a specific task from example data sets. Self-driving scanners are increasingly available, where AI automatically controls cardiac image prescriptions. These scanners offer faster image collection with higher spatial and temporal resolution, eliminating the need for cardiac triggering or breath holding. In the future, fully automated inline image analysis will most likely provide all contour drawings and initial measurements to the reader. Advanced analysis using radiomic or DL features may provide new insights and information not typically extracted in the current analysis workflow. AI may further help integrate these features with clinical, genetic, wearable-device, and "omics" data to improve patient outcomes. This article presents an overview of AI and its application in cardiac MRI, including in image acquisition, reconstruction, and processing, and opportunities for more personalized cardiovascular care through extraction of novel imaging markers.

MRI Assessment of Myocardial Deformation for Risk Stratification of Major Arrhythmic Events in Patients With Non-Ischemic Cardiomyopathy Eligible for Primary Prevention Implantable Cardioverter Defibrillators.

BACKGROUND: Implantable cardioverter-defibrillator (ICD) intervention is an established prophylactic measure. Identifying high-benefit patients poses challenges. PURPOSE: To assess the prognostic value of cardiac magnetic resonance imaging (MRI) parameters including myocardial deformation for risk stratification of ICD intervention in non-ischemic cardiomyopathy (NICM) while accounting for competing mortality risk. STUDY TYPE: Retrospective and prospective. POPULATION: One hundred and fifty-nine NICM patients eligible for primary ICD (117 male, 54 ± 13 years) and 49 control subjects (38 male, 53 ± 5 years). FIELD STRENGTH/SEQUENCE: Balanced steady state free precession (bSSFP) and three-dimensional phase-sensitive inversion-recovery late gadolinium enhancement (LGE) sequences at 1.5 T or 3 T. ASSESSMENT: Patients underwent MRI before ICD implantation and were followed up. Functional parameters, left ventricular global radial, circumferential and longitudinal strain, right ventricular free wall longitudinal strain (RV FWLS) and left atrial strain were measured (Circle, cvi42). LGE presence was assessed visually. The primary endpoint was appropriate ICD intervention. Models were developed to determine outcome, with and without accounting for competing risk (non-sudden cardiac death), and compared to a baseline model including LGE and clinical features. STATISTICAL TESTS: Wilcoxon non-parametric test, Cox's proportional hazards regression, Fine-Gray competing risk model, and cumulative incidence functions. Harrell's c statistic was used for model selection. A P value <0.05 was considered statistically significant. RESULTS: Follow-up duration was 1176 ± 960 days (median: 896). Twenty-six patients (16%) met the primary endpoint. RV FWLS demonstrated a significant difference between patients with and without events (-12.5% ± 5 vs. -16.4% ± 5.5). Univariable analyses showed LGE and RV FWLS were significantly associated with outcome (LGE: hazard ratio [HR] = 3.69, 95% CI = 1.28-10.62; RV FWLS: HR = 2.04, 95% CI = 1.30-3.22). RV FWLS significantly improved the prognostic value of baseline model and remained significant in multivariable analysis, accounting for competing risk (HR = 1.73, 95% CI = 1.12-2.66). DATA CONCLUSIONS: In NICM, RV FWLS may provide additional predictive value for predicting appropriate ICD intervention. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 5.

Uso de isotretinoina para manejo de la piel gruesa en rinoplastia. Revisión de alcance / Use of isotretinoin to manage thick skin in rhinoplasty. Scoping review

Introducción y objetivo: El cirujano plástico que realiza rinoplastia se enfrenta a pacientes con diferentes características según su etnia. Entre las estrategias implementadas para pacientes sometidos a rinoplastia se ha descrito el uso de isotretinoina para el manejo de pieles gruesas. Actualmente existen diferentes vacíos en la literatura sobre el beneficio y las complicaciones asociadas a este tipo de tratamiento en rinoplastia. El objetivo del presente trabajo es revisar la evidencia disponible en la literatura sobre los beneficios y riesgos derivados del uso de isotretinoina y rinoplastia Material y método: Desarrollamos una revisión del alcance del conocimiento en las diferentes bases de datos: Cochrane, PubMed, Embase, SAGE, Sciencie Direct, Taylor and Francis, utilizando los términos Mesh “Rhinoplasty AND Isotretinoin” para posteriormente evaluar los diferentes criterios de inclusión y exclusión que permitieron filtrar artículos evaluados en inglés y español. Resultados: Recogimos en total 2621 artículos publicados entre enero del 2005 y junio del 2022, de los cuales 13 cumplieron con los criterios establecidos por los investigadores para su análisis: 5 reportes de caso, 2 revisiones narrativas, 2 artículos tipo experimental, 1 consenso de expertos, 1 de casos y controles, 1 estudio multicéntrico y 1 revisión sistemática de la literatura. Conclusiones: La isotretinoina es un medicamento que, en diferentes artículos revisados, se demuestra seguro en su uso concomitante con procedimientos faciales como la rinoplastia y genera resultados favorables estéticos en paciente con nariz de piel gruesa en los 6 primeros meses de postoperatorio. Sin embargo, su uso debe ser analizado de manera individual puesto que puede llegar a producir efectos adversos tales como piel seca y adelgazamiento excesivo de la piel nasal. (AU)

Background and objective: Plastic surgeons who perform rhinoplasty face with patients with different characteristics depending on their ethnicity. Among the strategies implemented for these patients, the use of isotretinoin has been described for the management of thick skin. Currently, there are multiple gaps in the literature regarding the benefit and associated complications of this type of treatment in rhinoplasty. The objective of this study is to review the evidence available in the literature on the benefits and risks derived from the use of isotretinoin and rhinoplasty. Methods: A scoping review of the literature was developed on data base: PubMed Central (PMC), Cochrane, Sciencie Direct, Taylor and Francis Databases using Mesh terms “Rhinoplsaty AND Isotretinoin”. The articles were evaluated and classified to identify the available evidence in English and Spanish. Results: A review of 2621 articles published on different databases between January 2005 and June 2022 was performed, and 13 articles met the criteria established by the researchers for their analysis: 5 case reports, 2 narrative reviews, 2 controlled trials, 1 expert consensus, 1 case and control, 1 multicenter study, and 1 systematic review of the literature. Conclusions: Isotretinoin is a safe drug in concomitant use with facial procedures such as rhinoplasty and generates favorable aesthetic results in patients with a thick-skinned nose in the first 6 postoperative months. However, its use must be analyzed individually in every case since it can produce adverse effects such as dry skin and excessive thinning of the nasal skin. (AU)