# Q Bio

Ruslan Vasylev: Senior Visualization Front-End Engineer, Oct 2021–Jun 2024.

Help doctors explain changes in members’ organs over time using MRI-derived views.

At Q Bio, I built the viewer and geometry-processing path for doctor-led conversations with members. Synchronized prior/current views, organ models, slices and relevant biomarkers helped doctors explain changes over time.

- Operator: Doctor leading the member conversation
- Audience: Member discussing changes in their own scans over time
- Ownership: Frontend viewer + geometry-processing pipeline
- Workflow: Segmentation masks → prepared organ models → aligned prior/current dashboard views → doctor–member discussion

## Projects

- **Medical digital twin visualization**: Whole-body MRI and biomarker data in the browser, with synchronized scan panels and geometry processing from NIfTI to optimized glTF. Facts: Organ selection + annotations, Prior ↔ current scans, MRI to browser geometry.

## What could doctors do?

### Organ selection + annotations

Ownership: Organ interaction, annotations and biomarker integration

Organ selection and annotations expose relevant biomarker information during the doctor’s explanation to the member, connecting the 3D view to the longitudinal dashboard discussion.

- React · TypeScript · Three.js
- Select an organ → relevant biomarkers
- 3D organs + 2D slices + volume rendering

Technologies: React, TypeScript, Three.js, WebGL

### Prior ↔ current scans

Ownership: Linked camera and scan-comparison UI

I introduced camera locking across image panels so doctors could keep prior and current scans aligned while explaining changes to members.

- Prior/current scan panels
- Linked camera movement · rotate / orbit
- Doctor explains changes over time to the member
- Doctor–member screen-sharing conversations

### Volumetric shader

Ownership: Frontend volume-rendering shader

I used volume rendering alongside the prepared organ models so the browser could show the source imaging context as well as the extracted surfaces.

- GLSL volume rendering
- MRI scan context alongside organ geometry
- Inspect source slices + derived models

Technologies: GLSL

## Organ geometry

### MRI to browser geometry

Ownership: Python geometry processing from imaging-team masks

I built the processing path from the imaging team's segmentation masks to browser-ready organ geometry, including cleanup, validation and multiple levels of detail.

- Image Recon supplies segmentation masks
- Python · Marching Cubes → rough meshes
- Cleanup + validation → GLB levels of detail
- Prepared outputs → frontend viewer

Technologies: Python, GLB

### Conversion failures

Ownership: Geometry-conversion checks and failure alerts

I added checks and alerts for conversion failures and investigated the bad outputs. I also profiled the resulting models so they could render efficiently in the browser.

- Checks + alerts for failed outputs
- Debug conversion problems
- Profile geometry for browser rendering

## How did you work across teams?

### Team handoffs

Ownership: Geometry-processing pipeline and browser viewer

I worked primarily with Image Recon to understand the source data and make the derived geometry useful. Their team owned segmentation AI; the API team owned the application APIs.

- Image Recon · segmentation AI + masks
- My work · geometry pipeline + viewer
- Separate API team · application APIs

Collaboration: The imaging team owned segmentation AI; a separate team owned application APIs.

### Member access and doctor use

Ownership: Viewer/dashboard workflow used in member conversations

About three thousand members had dashboard access. Doctors used the viewer daily to communicate with members about changes in their organs over time and the relevant biomarkers.

- Doctors · daily member conversations
- Explain organ changes over time + relevant biomarkers