Skill UI

A comprehensive guide for academic researchers on how to structure and support claims in AI papers, particularly for top-tier conferences like ECAI. It details how to match the type of claim (e.g., completeness, efficiency, generalization) to the appropriate evidence—whether it requires a formal proof, a controlled empirical comparison, or a real-world deployment demonstration. The guide covers best practices for ML (seeds, ablations, baselines), multi-agent systems, and ensuring full provenance and academic honesty.

This guide provides comprehensive instructions for preparing and submitting supplementary material for computer vision conferences like ECCV. It details how the supplement should support the main paper's claims without adding new results, covering topics such as qualitative result visualization, detailed ablation studies, code archive scrubbing, and ensuring full anonymity across all uploaded assets.

Cirq is a powerful, open-source Python framework developed by Google Quantum AI. It allows users to design, simulate, and execute complex quantum circuits. It supports advanced features including parameterized circuits, noise modeling, circuit transformation, and seamless integration with multiple quantum hardware backends (e.g., IonQ, Google Quantum, Pasqal).

A robust, on-device solution for processing large-scale clinical text corpora. Supports batch NER, PII extraction, and de-identification over thousands of notes. Features include sharding, checkpointing, resumable functionality, and streaming results to JSONL, ensuring PHI safety and constant memory usage for massive datasets.

This skill deploys the OpenMed Model Context Protocol (MCP) server, exposing critical clinical NLP tools such as NER, PII extraction, and de-identification. It enables AI coding agents and chat clients to interact with OpenMed's capabilities via simple tool calls, running locally on-device to ensure user data privacy and security. Supports both stdio and Streamable HTTP deployment modes.

This module converts diverse clinical inputs—including scanned images, faxes, CSV/TSV exports, and C-CDA XML—into a normalized, clean text format (`ExtractedDocument`). It supports on-device OCR (Tesseract/PaddleOCR) and structured redaction for tables. It is the crucial first stage for feeding de-identification and Named Entity Recognition (NER) pipelines.

This skill manages the loading and efficient caching of OpenMed clinical/biomedical Named Entity Recognition (NER) models. Models are downloaded once from the Hugging Face Hub, enabling full offline operation afterward. It supports model reuse across multiple calls, configuring cache directories, specifying devices (CPU/GPU), and handling various model identifiers (registry keys, full HF IDs, or local paths).

This skill parses complex, free-text clinical trial eligibility criteria from sources like ClinicalTrials.gov into structured, machine-readable rules. It then evaluates these criteria against a comprehensive patient profile (derived from OpenMed data) to determine potential eligibility. The output provides an explainable verdict (eligible, ineligible, or unknown) for every criterion, serving as a critical decision support tool for clinicians and researchers.

This skill enhances raw clinical entity recognition by assigning three critical axes: Negation, Temporality, and Uncertainty (ConText). It determines the clinical status of a mention (e.g., affirmed, negated, historical, or uncertain), ensuring that basic NER results are transformed into clinically faithful assertions before grounding to standards like FHIR. Essential for accurate problem list generation and differential diagnosis support.

Provides a hardened FastAPI endpoint for OpenMed's core functionalities: clinical Named Entity Recognition (NER), Personally Identifiable Information (PII) extraction, and de-identification. Designed for self-hosted microservices, it handles inference requests, model lifecycle management, and supports dynamic batching, ensuring secure, on-device processing of sensitive clinical data.

This skill converts free-text radiology narratives (CT, MRI, X-ray, etc.) into structured, machine-readable data. It automatically identifies and segments key report sections (Technique, Findings, Impression) and extracts critical clinical details such as anatomy, measurements, laterality, BI-RADS/Lung-RADS scores, and follow-up recommendations. It is crucial for building structured medical records and decision support systems.

A comprehensive skill designed to transform contested questions or complex decisions into a structured, auditable evidence map. It systematically tracks all supporting, contradicting, qualifying, and missing evidence, ensuring that reasoning relationships and source regions are preserved for rigorous review.