LUCAID: Agentic Multimodal AI for Lung Cancer Precision Pathology
为解决肺癌病理诊断复杂且现有AI工具性能有限的问题,开发了LUCAID系统,集成多种功能模块,实现精准诊断,并在临床验证中表现出高一致性。
为解决肺癌病理诊断复杂且现有AI工具性能有限的问题,开发了LUCAID系统,集成多种功能模块,实现精准诊断,并在临床验证中表现出高一致性。
研究使用DNA甲基化分析方法,通过机器学习模型区分不同类型的黑色素瘤病变,并预测临床治疗组别,为诊断和疾病分层提供信息。
This study addresses the fragmentation of specialties and limited modality support in existing medical foundation models by proposing a unified visual foundation model integrating pathology and radiology. Leveraging a multi-dimensional context attention mechanism, the framework unifies sparse and dense prediction tasks across 2D and high-dimensional inputs. Combined with multi-task joint training and parameter-efficient fine-tuning (PEFT), it enables federated learning on consumer-grade hardware. This work represents the first cross-specialty, multi-dimensional unified modeling approach for medical vision, achieving state-of-the-art performance across 12 benchmark datasets. Notably, fine-tuning less than 2.5% of parameters yields results comparable to full fine-tuning, while federated learning performance closely approximates centralized training, significantly enhancing adaptation efficiency in low-resource settings.
This work addresses the integration challenges posed by consumer-grade wearable devices in clinical settings—stemming from device heterogeneity, proprietary data formats, and regulatory compliance requirements—by proposing an event-driven, cloud-native high-throughput system. The system leverages a cross-platform mobile application to collect high-frequency physiological data and employs a microservices architecture coupled with a stream processing engine to enable FHIR-compliant data standardization, real-time analytics, and end-to-end machine learning support. It introduces a novel dependency-aware FHIR minimization strategy that significantly reduces storage overhead while preserving lossless data reconstruction, thereby establishing a vendor-agnostic, scalable clinical integration framework. Evaluated performance demonstrates support for up to 50 ingestion requests per second with a median response latency under 8 milliseconds, satisfying stringent low-latency monitoring demands while adhering to healthcare regulatory standards.
Existing methods struggle to consistently define and accurately evaluate the separation between aleatoric and epistemic uncertainty, often relying on imperfect proxy tasks due to the absence of ground-truth uncertainty targets. This work proposes a unified definition of uncertainty as the pointwise posterior risk—the expected loss of a predictor with respect to the true function distribution given observed data—thereby integrating Bayesian functional uncertainty with estimation bias. Building on this formulation, we introduce the first semi-synthetic benchmark that provides direct access to ground-truth uncertainty targets, eliminating dependence on proxy tasks. Experiments reveal that predictive accuracy does not necessarily correlate with uncertainty reliability, enabling clear identification of methods aligned with true uncertainty while exposing their sensitivity to data and modeling choices.
为解决肺癌病理诊断复杂且现有AI工具性能有限的问题,开发了LUCAID系统,集成多种功能模块,实现精准诊断,并在临床验证中表现出高一致性。
研究使用DNA甲基化分析方法,通过机器学习模型区分不同类型的黑色素瘤病变,并预测临床治疗组别,为诊断和疾病分层提供信息。
This study addresses the fragmentation of specialties and limited modality support in existing medical foundation models by proposing a unified visual foundation model integrating pathology and radiology. Leveraging a multi-dimensional context attention mechanism, the framework unifies sparse and dense prediction tasks across 2D and high-dimensional inputs. Combined with multi-task joint training and parameter-efficient fine-tuning (PEFT), it enables federated learning on consumer-grade hardware. This work represents the first cross-specialty, multi-dimensional unified modeling approach for medical vision, achieving state-of-the-art performance across 12 benchmark datasets. Notably, fine-tuning less than 2.5% of parameters yields results comparable to full fine-tuning, while federated learning performance closely approximates centralized training, significantly enhancing adaptation efficiency in low-resource settings.
This work addresses the integration challenges posed by consumer-grade wearable devices in clinical settings—stemming from device heterogeneity, proprietary data formats, and regulatory compliance requirements—by proposing an event-driven, cloud-native high-throughput system. The system leverages a cross-platform mobile application to collect high-frequency physiological data and employs a microservices architecture coupled with a stream processing engine to enable FHIR-compliant data standardization, real-time analytics, and end-to-end machine learning support. It introduces a novel dependency-aware FHIR minimization strategy that significantly reduces storage overhead while preserving lossless data reconstruction, thereby establishing a vendor-agnostic, scalable clinical integration framework. Evaluated performance demonstrates support for up to 50 ingestion requests per second with a median response latency under 8 milliseconds, satisfying stringent low-latency monitoring demands while adhering to healthcare regulatory standards.
Existing methods struggle to consistently define and accurately evaluate the separation between aleatoric and epistemic uncertainty, often relying on imperfect proxy tasks due to the absence of ground-truth uncertainty targets. This work proposes a unified definition of uncertainty as the pointwise posterior risk—the expected loss of a predictor with respect to the true function distribution given observed data—thereby integrating Bayesian functional uncertainty with estimation bias. Building on this formulation, we introduce the first semi-synthetic benchmark that provides direct access to ground-truth uncertainty targets, eliminating dependence on proxy tasks. Experiments reveal that predictive accuracy does not necessarily correlate with uncertainty reliability, enabling clear identification of methods aligned with true uncertainty while exposing their sensitivity to data and modeling choices.