image pipeline construction

To address persistent challenges in multi-center medical imaging AI collaboration—including difficult cross-institutional algorithm deployment, strong environmental dependencies, implicit data assumptions, and incomplete documentation leading to frequent manual intervention—this study systematically identifies comprehensive packaging requirements and reveals critical interoperability and reusability gaps in existing APIs and standards. We propose the first FAIR-aligned (particularly Interoperability and Reusability) packaging specification framework for medical imaging AI algorithms. Methodologically, we integrate domain ontologies (DICOM/SNOMED CT), Docker-based containerization, standardized interfaces (FHIR/MONAI Deploy), and FAIR data practices. Our contributions include an open-source packaging requirements checklist and evaluation matrix, which significantly reduce integration overhead in federated learning, multi-center validation, and clinical deployment. This work establishes foundational interoperability support for a sustainable medical AI algorithm ecosystem.

Clinical workflow fragmentation severely impedes efficiency: heterogeneous scripting, ad-hoc model ensembles, and lack of data-driven modality identification and standardized outputs result in high deployment overhead, costly monitoring, and poor interoperability. To address this, we propose a healthcare-first vision-language unified framework that pioneers the use of a single vision-language model (VLM) for two-tier clinical decision-making—first, an auditable, three-stage routing mechanism matches inputs to expert-defined model cards; second, domain-specific multi-task joint inference (with early-exit capability and candidate arbitration) adheres to clinical risk constraints. Leveraging phased prompting, a candidate answer selector, and specialty-specific fine-tuning, our framework unifies modality identification, abnormality classification, model selection, and multi-task reasoning. Evaluated across gastroenterology, hematology, ophthalmology, and pathology, our single-model solution achieves performance on par with specialized models while substantially reducing deployment complexity, operational overhead, and integration effort.

This work addresses the lack of standardized, user-friendly, and reproducible software environments in medical image analysis, which hinders the widespread adoption of advanced methodologies. To overcome this limitation, the authors propose a modular, zero-code platform that enables seamless integration of image reading, visualization, registration, segmentation, radiomics feature extraction, and machine learning through intuitive graphical workflows. The platform allows users to construct, execute, and share end-to-end analysis pipelines without programming expertise. It introduces a unified, executable, and shareable cross-modality workflow system that ensures full transparency and facilitates collaborative reuse across disciplines. Supporting classification, regression, and clustering tasks, the platform significantly enhances analytical consistency, reproducibility, and interdisciplinary collaboration in clinical and translational research, as demonstrated by experimental validation.

This work addresses the lack of a unified, machine-verifiable data specification in medical imaging AI, which hinders consistent dataset structure, annotation provenance, quality documentation, and ML-readiness. To bridge this gap, we propose VIDS—an open standard that, for the first time, integrates standardized folder organization, naming conventions, annotation provenance schemas, and quality documentation within a single framework, along with 21 machine-verifiable rules. Built around the NIfTI working format while preserving DICOM metadata, VIDS provides an open-source validator installable via PyPI and supports export to mainstream frameworks such as nnU-Net, MONAI, and COCO. Evaluation reveals that four widely used datasets comply with only 20–39% of VIDS dimensions. We also release LIDC-Hybrid-100, a fully compliant reference dataset comprising 100 CT scans annotated by consensus among four radiologists (mean Dice: 0.7765), which passes all 21 validation checks.

Medical imaging data—along with associated structured data such as segmentation maps, radiotherapy dose distributions, and structured reports—pose significant re-identification risks when shared publicly across jurisdictions. Method: We propose the first systematic de-identification technical specification tailored to public release scenarios, integrating precise DICOM metadata redaction, pixel-level anonymization of sensitive anatomical regions, semantic consistency preservation for structured objects, and a quantitative privacy risk assessment framework. Contribution/Results: This work establishes the first formally defined technical boundary for medical image de-identification and introduces a standardized, modality-agnostic risk control framework encompassing both primary imaging modalities and derived objects. The resulting guidelines constitute an internationally recognized best-practice standard for de-identification, demonstrably reducing re-identification risk while enabling compliant, open-scientific sharing of imaging data for AI training and research.

This study addresses the challenge that heterogeneous CT acquisition protocols and diverse downstream tasks hinder the generalizability of fixed preprocessing pipelines, as existing approaches rely on handcrafted rules and lack adaptability. To overcome this limitation, the authors introduce, for the first time, a large language model (LLM) agent that constructs structured data-task profiles and employs a bounded decision policy to select optimal DICOM sequences or preprocessing configurations. A controlled deterministic execution mechanism is further designed to enable decision verification, fault-tolerant recovery, and secure isolation. The proposed method achieves state-of-the-art macro-averaged Dice scores across three public CT segmentation benchmarks and improves the output rate from 61.7% to 70.0% on two private raw DICOM cohorts while maintaining stable registration metrics.

本文解决了病理图像衍生数据共享不足的问题，通过将这些数据转换为DICOM标准格式并在NCI IDC平台上公开分享来解决。

研究通过开放的PACS-AI平台在六家医院部署影像AI，发现主要问题在于基础设施而非模型准确性，并强调了发布模型准备状态的重要性。

本文提出一种基于技能的方法，通过结合文献引导、自动代码生成和假设驱动实验来自动化医学影像模型开发流程，以减少构建竞争性基线所需的工程努力。

This work addresses the stringent demands of medical imaging systems for reliability, real-time performance, and deterministic timing in front-end electronics. To meet these requirements, the authors design and implement an FPGA-based miniature X-ray detector front-end prototype that integrates sensor data acquisition, real-time image processing, error detection, and result communication. By efficiently mapping complex imaging front-end functionalities onto a compact FPGA architecture, the system achieves low latency, high throughput, and deterministic response while significantly reducing overall complexity. The prototype demonstrates the feasibility of satisfying critical medical imaging performance criteria under strict resource constraints, offering a scalable hardware implementation paradigm for high-reliability embedded medical devices.