DeepEMC-T2 mapping: Deep learning-enabled T2 mapping based on echo modulation curve modeling.

PURPOSE: Echo modulation curve (EMC) modeling enables accurate quantification of T2 relaxation times in multi-echo spin-echo (MESE) imaging. The standard EMC-T2 mapping framework, however, requires sufficient echoes and cumbersome pixel-wise dictionary-matching steps. This work proposes a deep learning version of EMC-T2 mapping, called DeepEMC-T2 mapping, to efficiently estimate accurate T2 maps from fewer echoes. METHODS: DeepEMC-T2 mapping was developed using a modified U-Net to estimate both T2 and proton density (PD) maps directly from MESE images. The network implements several new features to improve the accuracy of T2/PD estimation. A total of 67 MESE datasets acquired in axial orientation were used for network training and evaluation. An additional 57 datasets acquired in coronal orientation with different scan parameters were used to evaluate the generalizability of the framework. The performance of DeepEMC-T2 mapping was evaluated in seven experiments. RESULTS: Compared to the reference, DeepEMC-T2 mapping achieved T2 estimation errors from 1% to 11% and PD estimation errors from 0.4% to 1.5% with ten/seven/five/three echoes, which are more accurate than standard EMC-T2 mapping. By incorporating datasets acquired with different scan parameters and orientations for joint training, DeepEMC-T2 exhibits robust generalizability across varying imaging protocols. Increasing the echo spacing and including longer echoes improve the accuracy of parameter estimation. The new features proposed in DeepEMC-T2 mapping all enabled more accurate T2 estimation. CONCLUSIONS: DeepEMC-T2 mapping enables simplified, efficient, and accurate T2 quantification directly from MESE images without dictionary matching. Accurate T2 estimation from fewer echoes allows for increased volumetric coverage and/or higher slice resolution without prolonging total scan times.

Rapid 3D T1 mapping using deep learning-assisted Look-Locker inversion recovery MRI.

PURPOSE: Conventional 3D Look-Locker inversion recovery (LLIR) T1 mapping requires multi-repetition data acquisition to reconstruct images at different inversion times for T1 fitting. To ensure B1 robustness, sufficient time of delay (TD) is needed between repetitions, which prolongs scan time. This work proposes a novel deep learning-assisted LLIR MRI approach for rapid 3D T1 mapping without TD. THEORY AND METHODS: The proposed approach is based on the fact that T 1 * $$ {\mathrm{T}}_1^{\ast } $$ , the effective T1 in LLIR imaging, is independent of TD and can be estimated from both LLIR imaging with and without TD, while accurate conversion of T 1 * $$ {\mathrm{T}}_1^{\ast } $$ to T1 requires TD. Therefore, deep learning can be used to learn the conversion of T 1 * $$ {\mathrm{T}}_1^{\ast } $$ to T1 , which eliminates the need for TD. This idea was implemented for inversion-recovery-prepared Golden-angel RAdial Sparse Parallel T1 mapping (GraspT1 ). 39 GraspT1 datasets with a TD of 6 s (GraspT1 -TD6) were used for training, which also incorporates additional anatomical images. The trained network was applied for T1 estimation in 14 GraspT1 datasets without TD (GraspT1 -TD0). The robustness of the trained network was also tested. RESULTS: Deep learning-based T1 estimation from GraspT1 -TD0 is accurate compared to the reference. Incorporation of additional anatomical images improves the accuracy of T1 estimation. The technique is also robust against slight variation in spatial resolution, imaging orientation and scanner platform. CONCLUSION: Our approach eliminates the need for TD in 3D LLIR imaging without affecting the T1 estimation accuracy. It represents a novel use of deep learning towards more efficient and robust 3D LLIR T1 mapping.

Long-term analysis of the urban heat island effect using multisource Landsat images considering inter-class differences in land surface temperature products.

It is imperative to quantitatively analyze the long-term temporal and spatial characteristics of the urban heat island (UHI) effect on cities for applications, such as urban expansion and environmental protection. Owing to the high spatial resolution and availability of long time-series data, remote sensing images from Landsat satellites are widely used for land surface temperature (LST) retrieval. However, limited by the satellite revisit cycle and image quality, the use of multisource Landsat images in a long-term study of the UHI effect is inevitable. Nonetheless, owing to the differences among multisource sensors, such as Landsat-7 and Landsat-8, there may be apparent deviations in the LST results retrieved from different sensor data, which are obtained from the same area and under similar circumstances. Consequently, it is necessary to build a relationship between the LST results generated from multisource Landsat sensors for future research on the UHI effect. In this study, Shenzhen city was studied to explore the fitting relationship between the corresponding LST products from Landsat-7 and Landsat-8 images obtained from adjacent dates with similar climatic conditions. Furthermore, factors affecting the fitting models, such as land cover types, seasonal and inter-annual differences, were analyzed. The constructed fitting model had a strong relationship with land cover types but a relatively weak relationship with seasonal and inter-annual differences; this indicates that a pseudo Landsat-8-based LST product can be generated from a Landsat-7-based LST product using a model fitted by a Landsat-7/8 pair obtained from adjacent years (or different seasons). Finally, by considering the consistency between LST products from multisource Landsat images, the spatiotemporal variations in the UHI effect in Shenzhen can be accurately explored using long time-series data.