Cardiovascular medicine is inherently multimodal, but the data available at the clinical decision point are rarely as complete or as synchronized as benchmark evaluations assume. Recent reviews identify incomplete datasets, heterogeneous data characteristics, and missing modalities as persistent barriers to multimodal clinical AI. A correspondence published in Medicine Discovery argues that the central problem is not how to fuse more data, but how to reason when clinically relevant data are incomplete, temporally misaligned, or no longer representative of the patient's current state. The authors distinguish five linked but non-equivalent problems — conventional multimodal fusion, robustness to missing modalities, temporally aware modeling, uncertainty-aware selective prediction, and clinically informed interpretation of why a modality is unavailable — and treat modality availability as an observation state rather than a data defect. They define deferral operationally, separate it from calibration, uncertainty estimation, and distribution-shift detection, propose hybrid deferral rules that combine non-compensatory clinical hard stops with graded model-derived signals, and set out five candidate conditions for an intended-use-specific minimum evaluation suite. The framework is offered as a complement to existing clinical AI guidance (TRIPOD+AI, PROBAST+AI, FUTURE-AI, DECIDE-AI, CONSORT-AI, SPIRIT-AI) rather than a replacement, and as a conceptual proposal requiring prospective evaluation and multi-stakeholder consensus.

A new study compares archived real-time and post-processed precise satellite products for multi-constellation, multi-frequency precise point positioning with ambiguity resolution (PPP-AR). 

A research paper by scientists at Guangzhou Medical University presented a  large-language-model-based multimorbidity score (LLM-MMS), a prompt-based large-language-model-derived multimorbidity phenotype score.

For patients with EGFR-mutant non-small cell lung cancer, resistance to tyrosine kinase inhibitors (TKIs)—most often driven by MET amplification—remains a major clinical obstacle. Researchers have now developed a structure-guided rational design method that converts existing monoclonal antibodies into common-light-chain bispecific antibodies, without large-scale library screening. Applying this framework, the team generated ALK101, an EGFR×c-MET bispecific antibody with binding affinity and biological activity comparable to the approved benchmark amivantamab. ALK101 showed robust antitumor efficacy across multiple tumor models, including tumors resistant to first-, second-, and third-generation TKIs, together with favorable pharmacokinetic and safety profiles in non-human primates. The study establishes a generalizable route to accelerate bispecific antibody development and nominates ALK101 as a promising clinical candidate.

Diabetic kidney disease (DKD) is commonly associated with metabolic disorders and inflammatory injury, but gut-resident bacteria may also contribute to renal damage. Researchers found that Escherichia coli/Escherichia-Shigella was enriched in the gut microbiota of patients and mice with DKD. In diabetic mice, viable enteroaggregative Escherichia coli (EAEC) crossed the impaired intestinal barrier and colonized the kidneys. Oral administration of mouse- or patient-derived EAEC increased renal bacterial burden, inflammation, and fibrosis. The researchers subsequently isolated an EAEC-specific bacteriophage, EAWsc9, that selectively lysed EAEC and carried no known virulence or antimicrobial resistance genes. After three weeks of oral treatment with EAWsc9, db/db mice showed reductions in renal EAEC burden, serum creatinine, blood urea nitrogen, indoxyl sulfate, inflammatory markers, and fibrosis. These findings extend the gut–kidney axis beyond microbial metabolites to include the translocation of viable bacteria and suggest a precision, microbiota-directed antibacterial strategy for addressing infection-associated renal injury in DKD.

Lead-based piezoelectric materials still dominate high-performance ultrasonic transducers, creating demand for environmentally friendly alternatives that can deliver comparable device performance. A research team led by Guangzhi Dong at Xidian University used B-site Fe³⁺ acceptor doping to tune the phase structure and electromechanical response of BNKBT lead-free ceramics. The optimized BNKBT-Fe(0.01) ceramic reached d₃₃ = 180 pC/N and k_t = 0.42. A 10 MHz transducer achieved an insertion loss of 16 dB, while acoustic focusing enabled 182 μm axial and 264 μm lateral resolution and clear porcine-eyeball imaging.

Skin is an entry point for tick-borne pathogens, but the responses of its different cell types remain incompletely understood. Researchers reporting in hLife mapped skin-cell responses to Rickettsia heilongjiangensis in wild-type mice, mice lacking type I or both type I and type II interferon receptors. They also examined patient samples and clinical data. Macrophage gene expression varied across interferon receptor backgrounds, whereas neutrophils retained signatures related to oxidative phosphorylation and protein degradation across all three mouse models. Fibroblast responses also differed in timing. In evaluable human tick-bite skin samples, the lipid peroxidation marker MDA appeared only in lesions from rickettsial PCR-positive patients, while MPO and OGG1 staining occurred in both PCR-positive and PCR-negative patients. Among patients with confirmed spotted fever group rickettsial infection and clear lesion and fever information, fever was less frequent in those with local skin lesions. The findings identify distinct cellular responses and a clinical association, while leaving open whether these local responses influence infection outcomes.

The enhancement of electrochemical performance for protonic ceramic cells is hindered by sluggish oxygen-electrode kinetics. Increasing specific surface area is a widely-adopted strategy to create more catalytically-active sites. Differently, a research team from Nanjing Tech University proposed a novel design strategy with large-grain for Sc-Ce co-doped BaCoO₃-based (BCSCe) electrodes system. Where the proton-coupled oxygen reduction reaction of BCSCe system follows the hydrogenation-dehydrogenation mechanism, micron-sized electrodes deliver superior electrochemistry compared with nanostructured electrode, achieving peak power density of 2.28 W cm^-2 and stable operation over 120 hours, offering new guidance for oxygen electrode design.

A growing surplus of refinery sulfur could find a new use in asphalt production, reducing reliance on petroleum-based bitumen and lowering costs. But a new academic review highlights unanswered questions about emissions, human exposure and long-term health implications, calling for safer formulations and life-cycle assessment of sulfur-modified asphalt.