Glossary & Terminology¶
Entheory.AI System Terms¶
| Term | Definition |
|---|---|
| Canonical Bundle | The single source of truth for a patient's data within Entheory.AI. A JSON document that aggregates and normalizes data from all sources (EMR, LIS, PACS) into a unified structure. |
| Longitudinal Record | A patient's medical history organized chronologically over time, linking events across different facilities and encounters to tell a complete story of their care journey. |
| Ingestion Pipeline | The automated process of receiving data (HL7, JSON, Files), validating it, and updating the patient bundle. |
| Processing Worker | A background service responsible for heavy lifting tasks like OCR (Optical Character Recognition) or ASR (Automatic Speech Recognition). |
| Provenance | Metadata tracking the origin of every data point (e.g., "Source: LIS-HOSPITALX, Received: 2024-12-01"). Crucial for clinical trust. |
| Ambient Scribe | An AI-powered system that listens to patient-doctor conversations in real-time and automatically generates clinical notes, reducing documentation burden. |
| Human-in-the-Loop (HITL) | A workflow where a human expert (scribe or doctor) reviews and validates AI-generated outputs before they are finalized in the EMR. |
| Dead Letter Queue (DLQ) | A specialized queue where messages that fail processing (due to errors or malformed data) are sent for manual inspection, ensuring no data is ever lost. |
Medical & Oncology Terms¶
| Term | Definition |
|---|---|
| Adjuvant Therapy | Additional cancer treatment given after the primary treatment (usually surgery) to lower the risk that the cancer will come back. |
| Biomarker | A biological molecule found in blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or of a condition or disease (e.g., HER2, EGFR). |
| CR (Complete Response) | Disappearance of all signs of cancer in response to treatment. |
| ECOG | Eastern Cooperative Oncology Group Performance Status. A scale from 0 (fully active) to 5 (dead) used to assess a patient's disease progression and daily living abilities. |
| Histology | The study of the microscopic structure of tissues. In oncology, refers to the type of cell the cancer originated from (e.g., "Invasive Ductal Carcinoma"). |
| Line of Therapy | The sequence of treatments given to a patient. 1st line is the initial treatment; 2nd line is given if the disease progresses, etc. |
| Metastasis | The spread of cancer cells from the place where they first formed to another part of the body. |
| Neoadjuvant Therapy | Treatment given as a first step to shrink a tumor before the main treatment, which is usually surgery. |
| Palliative Care | Care given to improve the quality of life of patients who have a serious or life-threatening disease. The goal is to prevent or treat as early as possible the symptoms of a disease, side effects caused by treatment of a disease, and psychological, social, and spiritual problems related to a disease or its treatment. |
| PR (Partial Response) | A decrease in the size of a tumor, or in the extent of cancer in the body, in response to treatment. |
| Progression-Free Survival (PFS) | The length of time during and after the treatment of a disease, such as cancer, that a patient lives with the disease but it does not get worse. |
| RECIST | Response Evaluation Criteria in Solid Tumors. A standard way to measure how well a cancer patient responds to treatment (CR=Complete Response, PR=Partial Response, SD=Stable Disease, PD=Progressive Disease). |
| SOAP Note | Subjective, Objective, Assessment, Plan. A standard format for documenting patient encounters. AI Scribe generates notes in this structure. |
| TNM Staging | A system to describe the amount and spread of cancer in a patient's body. T describes the size of the tumor and any spread of cancer into nearby tissue; N describes spread of cancer to nearby lymph nodes; M describes metastasis. |
AI & Machine Learning Terms¶
| Term | Definition |
|---|---|
| ASR | Automatic Speech Recognition. Technology that converts spoken language into text. We use it for transcribing doctor dictations and patient consultations. |
| Diarization | The process of partitioning an audio stream into homogeneous segments according to the speaker identity (e.g., "Who spoke when?"). Crucial for distinguishing Doctor vs. Patient. |
| Hallucination | A phenomenon where an AI model generates incorrect or nonsensical information confidently. We mitigate this by grounding all AI outputs in the patient's actual data. |
| LLM | Large Language Model. A type of AI model trained on vast amounts of text data to understand and generate human-like text. |
| OCR | Optical Character Recognition. Technology that converts different types of documents, such as scanned paper documents, PDF files, or images captured by a digital camera, into editable and searchable data. |
| RAG | Retrieval-Augmented Generation. A technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources (in our case, the patient's medical record). |
| Transformer | A deep learning architecture that relies on the mechanism of self-attention, weighing the significance of each part of the input data. It is the foundation of modern LLMs. |
Technical Standards & Formats¶
| Term | Definition |
|---|---|
| DICOM | Digital Imaging and Communications in Medicine. The standard for handling, storing, printing, and transmitting information in medical imaging. |
| FHIR | Fast Healthcare Interoperability Resources. A standard for exchanging healthcare information electronically. Entheory.AI uses FHIR R4. |
| HL7 v2 | Health Level Seven. An older but widely used standard for exchanging information between medical applications (e.g., ADT for admissions, ORU for results). |
| JSON | JavaScript Object Notation. A lightweight data-interchange format. Used for our internal canonical bundles and API responses. |
| ICD-10 | International Classification of Diseases, 10th Revision. A system used by physicians and other healthcare providers to classify and code all diagnoses, symptoms, and procedures. |
| LOINC | Logical Observation Identifiers Names and Codes. A universal code system for identifying laboratory and clinical observations. |
| RxNorm | A standardized nomenclature for clinical drugs, produced by the National Library of Medicine (NLM). Used to normalize medication names and dosages. |
| MLLP | Minimal Lower Layer Protocol. A standard for transmitting HL7 messages over TCP/IP. |
| SNOMED CT | Systematized Nomenclature of Medicine -- Clinical Terms. A comprehensive, multilingual clinical healthcare terminology. |
Indian Healthcare Ecosystem¶
| Term | Definition |
|---|---|
| ABDM | Ayushman Bharat Digital Mission. The Indian government's initiative to develop the backbone necessary to support the integrated digital health infrastructure of the country. |
| ABHA | Ayushman Bharat Health Account (formerly Health ID). A 14-digit number used to uniquely identify persons, authenticate them, and thread their health records (only with their informed consent) across multiple systems and stakeholders. |
| DPDP Act | Digital Personal Data Protection Act, 2023. India's law governing the processing of digital personal data, emphasizing consent, purpose limitation, and data minimization. |
| HIP | Health Information Provider. An entity (like a hospital or lab) that creates and stores health data. |
| HIU | Health Information User. An entity that requests access to health data (with consent) to provide a service (like Entheory.AI). |
| NHA | National Health Authority. The apex body responsible for implementing India's flagship public health insurance/assurance scheme called Ayushman Bharat Pradhan Mantri Jan Arogya Yojana. |
| PHR | Personal Health Record. An electronic application used by patients to maintain and manage their health information in a private, secure, and confidential environment. |
| PM-JAY | Pradhan Mantri Jan Arogya Yojana. The world's largest health assurance scheme, which aims to provide a health cover of Rs. 5 lakhs per family per year for secondary and tertiary care hospitalization to over 12 crore poor and vulnerable families. |
Data Privacy & Security¶
| Term | Definition |
|---|---|
| Anonymization | The process of removing personally identifiable information from data sets, so that the people whom the data describe remain anonymous. |
| Encryption at Rest | Protecting data that is stored physically in any digital form (e.g., database, hard drive) using cryptographic methods. |
| Encryption in Transit | Protecting data while it is moving between locations (e.g., from browser to server) using protocols like TLS/SSL. |
| Pseudonymization | A data management and de-identification procedure by which personally identifiable information fields within a data record are replaced by one or more artificial identifiers, or pseudonyms. |
| Consent Artifact | A digital record (often signed/timestamped) that proves a patient explicitly agreed to a specific data action (e.g., "Allow Audio Recording"). |
| RBAC | Role-Based Access Control. A method of restricting network access based on the roles of individual users within an enterprise. |
Acronyms¶
- ADT: Admission, Discharge, Transfer
- API: Application Programming Interface
- ASR: Automatic Speech Recognition
- DLQ: Dead Letter Queue
- EMR: Electronic Medical Record
- HIMS: Hospital Information Management System
- LIS: Laboratory Information System
- OCR: Optical Character Recognition
- PACS: Picture Archiving and Communication System
- RBAC: Role-Based Access Control
- SaaS: Software as a Service
- VPC: Virtual Private Cloud