A new study develops and validates a nomogram to predict the need for continuous renal replacement therapy (CRRT) in patients with acute kidney injury (AKI) after lung transplantation. By integrating perioperative and early postoperative variables, this tool supports timely, individualized intervention, helping clinicians potentially reduce CRRT needs and improve patient outcomes in the complex post-lung transplant setting.

The diabetes drug, metformin, may mimic key benefits of exercise in prostate cancer patients, raising a molecule linked to weight and energy control even without activity. The finding suggests a new way to support metabolic health during hormone therapy, when patients often cannot exercise.

Antivirals given for chronic hepatitis B include nucleos(t)ide analog monotherapy and pegylated interferon-alpha therapy, but the safety of interferon in patients with cirrhosis has been unclear. A study in Portal Hypertension and Cirrhosis analyzed data from two large Chinese cohorts, including 920 patients with compensated cirrhosis, and found that interferon therapy was not associated with a higher risk of serious complications, supporting its use with appropriate monitoring.

Researchers from Osaka Metropolitan University and Japan System Techniques Co., Ltd. evaluated the 2019 Osaka-Higashi Line station openings using a natural experiment and health insurance claims. While the overall rail-line effect was not significant, Shigino Station, a well-connected transfer hub, showed an estimated cumulative reduction of about JPY 62,500 per medically treated person over four years, highlighting that station connectivity and local context may determinewhether new stations deliver healthcare cost benefits.

General-purpose large language models (LLMs) that rely on in-context learning do not reliably deliver the scientific understanding and performance required for drug discovery tasks. Simply increasing model size or introducing reasoning tokens does not yield significant performance gains. To address this gap, we introduce the MMAI Gym for Science, a one-stop shop molecular data formats and modalities as well as task-specific reasoning, training, and benchmarking recipes designed to teach foundation models the 'language of molecules' in order to solve practical drug discovery problems. We use MMAI Gym to train an efficient Liquid Foundation Model (LFM) for these applications, demonstrating that smaller, purpose-trained foundation models can outperform substantially larger general-purpose or specialist models on molecular benchmarks. Across essential drug discovery tasks - including molecular optimization, ADMET property prediction, retrosynthesis, drug-target activity prediction, and functional group reasoning - the resulting model achieves near specialist-level performance and, in the majority of settings, surpasses larger models, while remaining more efficient and broadly applicable in the domain.