With the recent advancement of novel biomedical technologies such as high-throughput sequencing and wearable devices, multi-modal biomedical data ranging from multi-omics molecular data to real-time continuous bio-signals are generated at an unprecedented speed and scale every day. For the first time, these multi-modal biomedical data are able to make precision medicine close to reality. However, due to data volume and complexity, making good use of these multi-modal biomedical data requires major effort. Researchers and clinicians are actively developing artificial intelligence (AI) approaches for data-driven knowledge discovery and causal inference using a variety of biomedical data modalities. These AI-based approaches have demonstrated promising results in various biomedical and healthcare applications. In this review paper, we summarize the state-of-the-art AI models for integrating multi-omics data and electronic health records (EHRs) for precision medicine. We discuss the challenges and opportunities in integrating multi-omics data with EHRs and future directions. We hope this review can inspire future research and development in integrating multi-omics data with EHRs for precision medicine.

The modern pattern of the Asian monsoon is thought to have formed around the Oligocene/Miocene transition and is generally attributed to Himalaya–Tibetan Plateau (H–TP) uplift. However, the timing of the ancient Asian monsoon over the TP and its response to astronomical forcing and TP uplift remains poorly known because of the paucity of well-dated high-resolution geological records from the TP interior. Here, we present a precession-scale cyclostratigraphic sedimentary section of 27.32 to 23.24 million years ago (Ma) during the late Oligocene epoch from the Nima Basin to show that the South Asian monsoon (SAM) had already advanced to the central TP (32°N) at least by 27.3 Ma, which is indicated by cyclic arid–humid fluctuations based on environmental magnetism proxies. A shift of lithology and astronomically orbital periods and amplified amplitude of proxy measurements as well as a hydroclimate transition around 25.8 Ma suggest that the SAM intensified at ~25.8 Ma and that the TP reached a paleoelevation threshold for enhancing the coupling between the uplifted plateau and the SAM. Orbital short eccentricity-paced precipitation variability is argued to be mainly driven by orbital eccentricity-modulated low-latitude summer insolation rather than glacial-interglacial Antarctic ice sheet fluctuations. The monsoon data from the TP interior provide key evidence to link the greatly enhanced tropical SAM at 25.8 Ma with TP uplift rather than global climate change and suggest that SAM’s northward expansion to the boreal subtropics was dominated by a combination of tectonic and astronomical forcing at multiple timescales in the late Oligocene epoch.

Recent advances in artificial intelligence (AI) have sparked interest in developing explainable AI (XAI) methods for clinical decision support systems, especially in translational research. Although using XAI methods may enhance trust in black-box models, evaluating their effectiveness has been challenging, primarily due to the absence of human (expert) intervention, additional annotations, and automated strategies. In order to conduct a thorough assessment, we propose a patch perturbation-based approach to automatically evaluate the quality of explanations in medical imaging analysis. To eliminate the need for human efforts in conventional evaluation methods, our approach executes poisoning attacks during model retraining by generating both static and dynamic triggers. We then propose a comprehensive set of evaluation metrics during the model inference stage to facilitate the evaluation from multiple perspectives, covering a wide range of correctness, completeness, consistency, and complexity. In addition, we include an extensive case study to showcase the proposed evaluation strategy by applying widely-used XAI methods on COVID-19 X-ray imaging classification tasks, as well as a thorough review of existing XAI methods in medical imaging analysis with evaluation availability. The proposed patch perturbation-based workflow offers model developers an automated and generalizable evaluation strategy to identify potential pitfalls and optimize their proposed explainable solutions, while also aiding end-users in comparing and selecting appropriate XAI methods that meet specific clinical needs in real-world clinical research and practice.

The Paleocene‐Eocene Thermal Maximum (PETM; ∼56 Ma)–a geologically rapid carbon‐release event similar to anthropogenic carbon emissions–is ideal for investigating weathering responses to rapid carbon release and associated climate change. Here, for the first time, we present high‐resolution lacustrine lithium isotope data that reflect basin‐scale silicate weathering across the PETM in the Nanyang Basin, eastern China. Our results reveal an ∼100% increase in regional silicate weathering intensity through the PETM relative to the pre‐PETM level. Synchronous variations between carbon isotope values and weathering intensity demonstrate a strong coupling between carbon cycle, climate, and chemical weathering on millennial timescales. These findings argue that strong negative weathering feedback and massive removal of CO₂ maintained Earth's climate within a habitable range during the PETM.

The uplift of the Tibetan Plateau (TP) is generally considered to have triggered the Asian monsoon system. However, a long‐standing scarcity of high‐resolution paleoclimatic records on the TP makes it unclear when the South Asian monsoon (SAM) reached the central TP. Here, we investigated the environmental magnetic properties of a lacustrine sequence (∼25.5–19.8 Ma) from the Lunpola Basin. Hematite is the dominant magnetic mineral with concentrations that vary with “box‐shaped” peaks that mainly resulted from preferential preservation and dissolution of detrital iron oxides in alternating nonsteady state redox environments. Environmental magnetic sequences reflect distinct dry‐wet cycles with a ∼400 kyr eccentricity periodicity, which are proposed to have been driven by SAM intensity. Therefore, our study reveals that the SAM intensified and advanced northward to the Lunpola Basin by at least ∼25.5 Ma.