Medical Imaging Stack

The Challenge of Medical Imaging

Medical imaging is fundamental to modern healthcare. X-rays, CT scans, MRIs, ultrasounds - these generate massive amounts of visual data that clinicians rely on for diagnosis and treatment. But managing medical images is complex.

Images aren't just pictures. Each study contains dozens or hundreds of individual images plus metadata - patient information, acquisition parameters, measurements, and annotations. Storage requirements are enormous - a single CT scan can be several gigabytes.

Commercial PACS (Picture Archiving and Communication System) solutions exist but are expensive, often costing hundreds of thousands of dollars for mid-sized facilities. They're also proprietary, creating vendor lock-in and integration challenges.

The open-source stack of Orthanc and OHIF offers an alternative - a complete medical imaging solution that's free, extensible, and standards-compliant. Let's explore what each component does and how they work together.

Understanding the Components

What is Orthanc?

Orthanc is an open-source DICOM server - the backend that stores and manages medical images. Think of it as the database and file system for medical imaging, but specifically designed for healthcare's unique requirements.

Developed since 2012, Orthanc is lightweight, fast, and remarkably easy to deploy. A single executable runs the entire server - no complex installation, no dependencies to manage. Yet it's powerful enough to handle production workloads.

What is OHIF?

OHIF (Open Health Imaging Foundation) Viewer is a web-based image viewer - the frontend that displays images for clinicians. It runs in any modern web browser, no special software required.

OHIF provides medical-grade viewing capabilities: windowing and leveling, measurements, annotations, multi-plane reconstruction, 3D rendering, and more. It's designed by radiologists for clinical use, not just image display.

How They Work Together

The stack architecture is straightforward:

1. Imaging equipment (CT scanner, X-ray machine) sends images to Orthanc via DICOM protocol
2. Orthanc stores images in its database and filesystem, making them available via DICOMweb APIs
3. OHIF Viewer queries Orthanc via DICOMweb, retrieves images, and displays them to clinicians
4. Clinicians view, manipulate, and annotate images in their web browser

Each component is modular. Don't like OHIF? Use another viewer. Need different storage? Orthanc supports plugins. This flexibility is a key advantage over monolithic commercial PACS.

Orthanc Deep Dive

Core Capabilities

• DICOM Storage: Accepts images from any DICOM-compliant device
• DICOM Query/Retrieve: Other systems can search and fetch images
• DICOMweb: Modern REST APIs (WADO-RS, QIDO-RS, STOW-RS)
• Worklist Management: Modality worklist support for scheduling
• Image Processing: Convert formats, anonymize, compress
• Routing: Automatically forward images to other PACS systems
• Web Interface: Built-in browser interface for administration

Storage Architecture

Orthanc stores images in two places:

• Database: Metadata (patient name, study date, series info) stored in SQLite (default) or PostgreSQL/MySQL
• Filesystem: Actual image data stored as files, organized by internal IDs

This separation allows flexibility. You can query metadata quickly without loading massive image files. Images can be stored on high-capacity disks while the database sits on faster SSDs.

Plugin Ecosystem

Orthanc's functionality extends through plugins:

• Database plugins: Use PostgreSQL or MySQL instead of SQLite for production
• Storage plugins: Store images in cloud object storage (AWS S3, Azure Blob)
• DICOMweb plugin: Enable modern web-based image access
• Authentication plugin: Add user logins and access controls
• Transfer syntax plugins: Support additional image compression formats
• Python plugin: Write custom logic in Python for automation

Performance

Orthanc is surprisingly performant. A modest server can handle:

• Hundreds of concurrent users
• Millions of stored images
• Dozens of imaging devices sending studies simultaneously

The lightweight design means low resource requirements. Small clinics run Orthanc on entry-level hardware. Larger facilities run it on standard servers or cloud VMs.

OHIF Viewer Deep Dive

Viewing Features

OHIF provides clinical-grade imaging tools:

• Windowing: Adjust brightness and contrast to visualize different tissue types
• Pan and zoom: Navigate large images
• Multi-plane reconstruction (MPR): View CT/MRI in axial, sagittal, and coronal planes
• Stack scrolling: Navigate through series of images
• Measurements: Length, area, angle, region of interest
• Annotations: Add text labels and arrows
• Cine mode: Play image series as video
• Hanging protocols: Automatically arrange images based on modality and body part
• Layout options: 1x1, 2x2, 3x3 viewports for comparing studies

Modern Web Technology

OHIF is built with modern JavaScript frameworks (React):

• Runs entirely in the browser - no installation
• Works on desktops, tablets, even phones
• Progressive Web App - can work offline once loaded
• Responsive design adapts to screen size

Extensibility

OHIF is designed for customization:

• Extensions: Add new tools and features
• Modes: Different workflows for different specialties (radiology vs cardiology)
• Themes: Customize appearance to match organizational branding
• Panel plugins: Add custom panels for institutional workflows

Developers can build on OHIF rather than starting from scratch - a huge time saver for custom imaging applications.

Advanced Features

Beyond basic viewing, OHIF supports:

• Segmentation: Display AI-generated segmentations over images
• Report integration: Link to structured reports and findings
• 3D rendering: Volume rendering for CT and MRI
• Fusion: Overlay PET on CT, compare pre/post-treatment scans
• Microscopy: View digital pathology slides

Setting Up the Stack

Installation Options

Docker Compose (Recommended for Beginners)

The fastest way to get started. A single configuration file launches both Orthanc and OHIF with proper networking:

• Download Docker Compose file from OHIF documentation
• Run docker-compose up
• Access OHIF at http://localhost:3000
• Orthanc admin at http://localhost:8042

Manual Installation

For production deployments or custom configurations:

1. Install Orthanc: Download binary or build from source, configure database and storage
2. Enable DICOMweb: Install DICOMweb plugin, configure authentication
3. Deploy OHIF: Build OHIF viewer, configure to point to Orthanc, host on web server
4. Configure devices: Set imaging equipment to send DICOM to Orthanc's IP and port

Cloud Deployment

Many organizations run the stack in the cloud:

• Orthanc on a VM with attached block storage
• OHIF as a static site on CDN
• VPN or private networking to connect imaging devices securely
• Regular backups to object storage

Security Considerations

Medical images are PHI - security is critical:

• Authentication: Enable Orthanc's built-in auth or use plugin for SSO
• Encryption: TLS/HTTPS for all web access, VPN or encrypted DICOM for device communication
• Access control: Limit which users see which studies (requires plugin or external system)
• Audit logging: Track who accessed what images when
• Firewall: DICOM ports (typically 4242) should not be publicly exposed
• Backups: Regular encrypted backups, tested restore procedures

Integration with EMR/EHR

Images are most useful when linked to patient records:

• Simple: Link from EMR to OHIF with patient ID in URL
• Moderate: Use SMART on FHIR to launch OHIF from EMR with patient context
• Advanced: Bidirectional integration - EMR queries Orthanc for studies, Orthanc queries EMR for patient data

Use Cases Beyond Radiology

Cardiology

Store echocardiograms, cardiac CT, cardiac MRI. OHIF extensions support cardiac-specific measurements and ejection fraction calculations.

Pathology

Whole-slide imaging creates massive images (gigapixels). OHIF can display these digital pathology slides with pan and zoom at multiple magnifications.

Dentistry

Dental X-rays, panoramic images, and cone-beam CT all use DICOM. Orthanc and OHIF work just as well for dental practices as medical radiology.

Veterinary Medicine

Animal hospitals have the same imaging needs as human hospitals. The stack is imaging-agnostic - it doesn't care if the patient is human or animal.

Research

Clinical trials and research studies generate large imaging datasets. Orthanc's ability to anonymize images and export in bulk makes it ideal for research environments.

Scaling Considerations

Small Deployments (1-5 Modalities)

• Single server running both Orthanc and OHIF
• Local storage (RAID for redundancy)
• SQLite database
• Cost: $2,000-$5,000 hardware

Medium Deployments (5-20 Modalities)

• Dedicated server for Orthanc
• PostgreSQL database on separate server or managed service
• NAS or SAN for image storage
• Load balancer if high availability needed
• Cost: $10,000-$30,000 hardware

Large Deployments (20+ Modalities, Multiple Sites)

• Multiple Orthanc instances (one per site or region)
• Central Orthanc for long-term archive
• Cloud object storage for images
• Managed PostgreSQL database
• CDN for OHIF distribution
• Cost: Variable, often cloud-based with pay-per-use

Challenges and Limitations

Advanced PACS Features

Commercial PACS offer features Orthanc+OHIF don't provide out-of-the-box:

• Sophisticated access control (user roles, department restrictions)
• Built-in reporting with structured templates
• Radiology Information System (RIS) integration
• Advanced quality control workflows
• Enterprise-level support and SLAs

These can be built or integrated, but require development effort.

Support Model

Being open source, there's no phone number to call for support. You rely on:

• Community forums and mailing lists
• Documentation (generally good for both projects)
• Hiring consultants for implementation
• Internal IT staff for troubleshooting

Organizations accustomed to vendor support may find this adjustment challenging.

Regulatory Compliance

Neither Orthanc nor OHIF are FDA-cleared medical devices. This may matter for certain use cases:

• Most jurisdictions: Fine for viewing and storing images
• Advanced AI/diagnostic uses: May need cleared software
• Check local regulations and consult legal counsel

The Future of Open Source Imaging

Both Orthanc and OHIF continue evolving:

• AI integration: Displaying AI findings and predictions alongside images
• Cloud-native architectures: Better support for scalable cloud deployments
• Interoperability: FHIR ImagingStudy resources linking imaging to EMRs
• Collaboration tools: Real-time multi-user viewing and annotation
• Mobile optimization: Better experiences on tablets and phones

The open-source community continues expanding. More institutions are choosing Orthanc and OHIF, creating a positive feedback loop - more users mean more contributors, better features, and greater stability.

Getting Started

Ready to try the stack?

1. Demo: Visit OHIF's online demo to see the viewer in action
2. Local install: Use Docker Compose to run the stack locally
3. Upload sample images: Use Orthanc's web interface or command-line tools to add test DICOM files
4. View in OHIF: Access the viewer and explore the imaging tools
5. Connect a device: Configure an imaging modality to send to your Orthanc instance
6. Plan production: Consider security, backup, integration, and scalability needs

For production deployments, consider engaging a consultant experienced with medical imaging infrastructure. The software is free, but implementation expertise is valuable.