Overview
Biomedical data encompasses an exceptionally broad spectrum of multi-modal, high-dimensional information generated across healthcare ecosystems, clinical diagnostic workflows, consumer wearables, and biological research laboratories. Effective organization, classification, and retrieval of these heterogeneous data streams form the foundation of modern medical informatics, clinical decision support systems ($\text{CDSS}$), computational biology, and AI-driven precision medicine.
Biomedical data structures can be broadly categorized into several primary modalities:
Clinical Data (EHR/EMR): Structured and unstructured records detailing patient demographics, vital sign histories, diagnostic coding ($\text{ICD-10/11}$), and prescription orders.
Medical Imaging Data: High-resolution spatial arrays capturing anatomical structures ($\text{X-ray, CT, MRI, Ultrasound}$) and functional metabolic activity ($\text{PET}$), standardized under the $\text{DICOM}$ protocol.
Genomic & Molecular Data: High-throughput nucleotide sequences ($\text{DNA/RNA}$), single-cell transcriptomics, proteomics, and epigenomic modifications representing cellular state spaces.
Biomedical Signal Data: High-frequency continuous time-series waveforms generated by electrophysiological currents ($\text{EEG, ECG, EMG, MEG}$) and photoplethysmographic sensors ($\text{PPG}$).
Standardizing these data structures enables interoperability between clinical health networks and research repositories. This interactive explorer provides a comprehensive taxonomy of the primary data modalities utilized in medical data science, illustrating their underlying biophysical origins, typical storage formats, and analytical applications.
How to Use
This interactive explorer allows you to navigate, query, and filter the master biomedical data hierarchy in real-time. Use the interface features below:
Real-Time Search & Filter Controls
- Keyword Search: Type any term (e.g. "spectrogram", "glucose", "DICOM", "cortex") into the "Real-Time Keyword Query" input bar. The explorer dynamically highlights matching strings in neon gold while instantly filtering out non-matching subcategories.
- Category Pill Selection: Click any of the category pill buttons (e.g. "Clinical Data", "Medical Imaging", "Genomic & Molecular") above the tree to isolate specific domain categories.
- Accordion Toggle: Click on any category header row to expand or collapse its sub-items. Click "Expand All Folders" in the right console to open all categories simultaneously.
- JSON Export: Click "Export JSON Payload" to copy or download the structured category array directly into your clipboard for downstream processing.
Audio Feedback & Interactive Inspection Demo
- Interface Sonification: Toggle the top-right "Audio" button to enable audio feedback. As you search or navigate categories, the explorer synthesizes subtle sine tones proportional to search result density.
- Automated Inspection Demo: Click "Play Demo" to run an automated tour through the category tree. Any manual mouse movement, keystroke, or click will immediately interrupt demo mode.
Technical Details
The core classification engine uses client-side regular expression filtering wrapped inside a debounced input event loop ($300\text{ms}$ throttle), preventing main-thread blocking during rapid typing. The string highlighting engine uses regular expression boundary escaping ($\text{escapeRegExp}$) to safely wrap matching substrings inside styled HTML spans without breaking DOM structure:
\text{RegExp}_{\text{match}} = \text{new RegExp}\Big(\text{string}.\text{replace}(/[.*+?^${}()|[\]\\]/g, \, \text{'\\$&'}), \, \text{'gi'}\Big)
Safe Regular Expression Substring Matching and DOM Injection Pattern
To eliminate Cumulative Layout Shift (CLS) on mobile devices, the visualizer panel defines a fixed minimum layout footprint (`min-height: 320px` with fluid overflow rules). Navigation headers contain explicit inline dimensions (`width="150" height="40"` on the logo image), and fluid AdSense containers are wrapped inside `min-height: 250px` layout elements to prevent ad insertion shifts.
Future Directions
Future development goals for this biomedical data repository include:
- FHIR & OpenEHR Schema Integration: Adding interactive schema validators to inspect live Fast Healthcare Interoperability Resources ($\text{FHIR}$) JSON resource bundles.
- Interactive 3D Knowledge Graph: Converting the flat accordion layout into an interactive 3D force-directed node graph leveraging D3.js and WebGL.
- Vector Semantic Search: Implementing local cosine similarity embeddings to allow natural language semantic queries beyond exact keyword matches.