Abstract
PURPOSE: To develop and evaluate MyoMapNet, a rapid myocardial T(1) mapping approach that uses fully connected neural networks (FCNN) to estimate T(1) values from four T(1)-weighted images collected after a single inversion pulse in four heartbeats (Look-Locker, LL4). METHOD: We implemented an FCNN for MyoMapNet to estimate T(1) values from a reduced number of T(1)-weighted images and corresponding inversion-recovery times. We studied MyoMapNet performance when trained using native, post-contrast T(1), or a combination of both. We also explored the effects of number of T(1)-weighted images (four and five) for native T(1). After rigorous training using in-vivo modified Look-Locker inversion recovery (MOLLI) T(1) mapping data of 607 patients, MyoMapNet performance was evaluated using MOLLI T(1) data from 61 patients by discarding the additional T(1)-weighted images. Subsequently, we implemented a prototype MyoMapNet and LL4 on a 3 T scanner. LL4 was used to collect T(1) mapping data in 27 subjects with inline T(1) map reconstruction by MyoMapNet. The resulting T(1) values were compared to MOLLI. RESULTS: MyoMapNet trained using a combination of native and post-contrast T(1)-weighted images had excellent native and post-contrast T(1) accuracy compared to MOLLI. The FCNN model using four T(1)-weighted images yields similar performance compared to five T(1)-weighted images, suggesting that four T(1) weighted images may be sufficient. The inline implementation of LL4 and MyoMapNet enables successful acquisition and reconstruction of T(1) maps on the scanner. Native and post-contrast myocardium T(1) by MOLLI and MyoMapNet was 1170 ± 55 ms vs. 1183 ± 57 ms (P = 0.03), and 645 ± 26 ms vs. 630 ± 30 ms (P = 0.60), and native and post-contrast blood T(1) was 1820 ± 29 ms vs. 1854 ± 34 ms (P = 0.14), and 508 ± 9 ms vs. 514 ± 15 ms (P = 0.02), respectively. CONCLUSION: A FCNN, trained using MOLLI data, can estimate T(1) values from only four T(1)-weighted images. MyoMapNet enables myocardial T(1) mapping in four heartbeats with similar accuracy as MOLLI with inline map reconstruction.