Tandem light-emitting diodes (LEDs), achieved by vertically stacking multiple units in series to combine the luminance of individual light-emitting elements, are effective for improving efficiency and lifespan compared to single-unit devices1-3. In particular, tandem perovskite LEDs benefit from the small Stokes shifts of perovskites4, which in principle can enable significant photon recycling between individual perovskite layers and enhance light extraction from trapped modes. However, a tandem structure that effectively merges the luminance of each perovskite units still remains a significant challenge. Here, we demonstrate efficient and stable tandem LEDs by combining two solution-processed perovskite light-emitting units. This tandem structure effectively combines the original luminance of each light-emitting units; we argue that the emissions are also significantly enhanced through photon recycling between the individual light-emitting units. Consequently, we achieve tandem perovskite LEDs with a low turn-on voltage of 3.2 V, a high peak external quantum efficiency (EQE) of 45.5% (even 20% higher than the sum of peak EQEs of single-unit devices), an average peak EQE of 40.9%, and a half-lifetime of 64 h at an initial radiance of 70 W Sr-1 m-2. These findings represent a significant advancement in achieving high-performance and multicolor LEDs through the stacking of perovskite LEDs.

Echocardiography is the most widely used cardiac imaging modality, capturing ultrasound video data to assess cardiac structure and function1. Artificial intelligence (AI) in echocardiography has the potential to streamline manual tasks and improve reproducibility and precision2. However, most echocardiography AI models are single-view, single-task systems that do not synthesize complementary information from multiple views captured during a full exam3,4, and thus lead to limited performance and scope of applications. To address this problem, we introduce EchoPrime, a multi-view, view-informed, video-based vision-language foundation model trained on over 12 million video-report pairs. EchoPrime uses contrastive learning to train a unified embedding model for all standard views in a comprehensive echocardiogram study with representation of both rare and common diseases and diagnoses. EchoPrime then utilizes view-classification and a view-informed anatomic attention module to weight video-specific embeddings that accurately map the relationship between echocardiographic views and anatomical structures. With retrieval-augmented interpretation, EchoPrime integrates information from all echocardiogram videos in a comprehensive study and performs holistic clinical interpretation. In datasets from five international independent healthcare systems, EchoPrime achieves state-of-the art performance on 23 diverse benchmarks of cardiac form and function, surpassing the performance of both task-specific approaches and prior foundation models. Following rigorous clinical evaluation, EchoPrime can assist physicians in the automated preliminary assessment of comprehensive echocardiography.