Search for: All records

Emergency chest X-ray (CXR) triage has a structural modality gap: reports arrive after triage decisions, yet multimodal foundation models require image-text inputs. We present Variational Risk Minimization (VRM), a distillation framework that treats LVLM-generated report variants as Monte Carlo samples of latent clinical interpretations. Rather than distilling from a single teacher target, VRM learns from a variationally marginalized teacher distribution, enabling uncertainty-aware supervision under missing-modality constraints. Under matched encoder families, VRM outperforms direct fine-tuning baselines and improves calibration with strong recovery from hallucinated supervision. Marginalized supervision reduces report-selection instability. In our compact edge-student instantiation, a confidence-gated cascade reaches AUC 0.941 at 103ms average latency with 20.3% cloud escalation, yielding an explicit reliability-latency operating point for cloud–edge clinical workflows. 

(1) Background: With technological advancements, the integration of wireless sensing and artificial intelligence (AI) has significant potential for real-time monitoring and intervention. Wireless sensing devices have been applied to various medical areas for early diagnosis, monitoring, and treatment response. This review focuses on the latest advancements in wireless, AI-incorporated methods applied to clinical medicine. (2) Methods: We conducted a comprehensive search in PubMed, IEEEXplore, Embase, and Scopus for articles that describe AI-incorporated wireless sensing devices for clinical applications. We analyzed the strengths and limitations within their respective medical domains, highlighting the value of wireless sensing in precision medicine, and synthesized the literature to provide areas for future work. (3) Results: We identified 10,691 articles and selected 34 that met our inclusion criteria, focusing on real-world validation of wireless sensing. The findings indicate that these technologies demonstrate significant potential in improving diagnosis, treatment monitoring, and disease prevention. Notably, the use of acoustic signals, channel state information, and radar emerged as leading techniques, showing promising results in detecting physiological changes without invasive procedures. (4) Conclusions: This review highlights the role of wireless sensing in clinical care and suggests a growing trend towards integrating these technologies into routine healthcare, particularly patient monitoring and diagnostic support.