Cardiac magnetic resonance (CMR) is an essential clinical tool for the assessment of cardiovascular disease. Deep learning (DL) has recently revolutionized the field through image reconstruction techniques that allow unprecedented data undersampling rates. These fast acquisitions have the potential to considerably impact the diagnosis and treatment of cardiovascular disease. Herein, we provide a comprehensive review of DL-based reconstruction methods for CMR. We place special emphasis on state-of-the-art unrolled networks, which are heavily based on a conventional image reconstruction framework. We review the main DL-based methods and connect them to the relevant conventional reconstruction theory. Next, we review several methods developed to tackle specific challenges that arise from the characteristics of CMR data. Then, we focus on DL-based methods developed for specific CMR applications, including flow imaging, late gadolinium enhancement, and quantitative tissue characterization. Finally, we discuss the pitfalls and future outlook of DL-based reconstructions in CMR, focusing on the robustness, interpretability, clinical deployment, and potential for new methods.

Breast cancer is the most common malignant tumor in women worldwide. In recent years, there has been an increasing use of immunohistochemistry (the process of detecting the expression of certain proteins in cytological images) to obtain useful information for diagnosis. This paper presents an efficient algorithm that automatically detects breast cancer cell nuclei and divides them into two groups: those that express the ER marker and those that do not. First, the areas that belong to the carcinoma are automatically identified. Then, the algorithm evaluates features such as size and shape to correctly segment the nuclei in these fields. Finally, the Fuzzy C-Means algorithm is used to classify the detected nuclei. The method proposed was evaluated with a set of 10 images which contained 4093 cell nuclei. The algorithm correctly identified 93.1% of the nuclei, and sensitivity and specificity of the classification were 95.7% and 93.2% respectively.

Objective: This study investigates the feasibility of non-contact retrospective respiratory gating and cardiac sensing using continuous wave Doppler radar deployed in an MRI system. The proposed technique can complement existing sensors which are difficult to apply for certain patient populations. Methods: We leverage a software-defined radio for continuous wave radar at 2.4 GHz to detect in-vivo respiratory and cardiac timescrolled signals. In-bore radar signal demodulation is verified with full electromagnetic simulations, and its functionality is validated on a test bench and within the MR bore with four normal subjects. Radar sensing was compared against well-known references: electrocardiography on a test bench, system bellows, and pulsed plethysmography sensors with in the MRI bore. Results: The feasibility of noncontact cardiac rate sensing, dynamic breathing sequence synchronization, and in-bore motion correction for retrospective respiratory gating applications was demonstrated. Optimal radar front-end system arrangement, along with spectral isolation and narrow bandwidth of operation, enable MRI-compatible and interference-free motion sensing. The signal-to-noise-ratio degradation by the radar integration was within 4.5% on phantom images. Conclusion: We confirmed that in-bore retrospective motion correction using CW Doppler radar is feasible without MRI system constraints. Significance : Non-contact motion correction sensing in MRI may provide better patient handling and through put by complementing existing system sensors and motion correction algorithms