Research Survey: Applications and Citations of the ISET Software Suite

The Image Systems Engineering Toolbox (ISET) suite—including ISETCam, ISETBio, ISET3d, and related packages like ISETHDR and ISETAuto—serves as an open-source, physics-based computational framework across several major academic disciplines.

Below is an organized summary of key publication domains, representative papers with links, and positioning insights for future community outreach.

1. Visual Neuroscience & Computational Psychophysics

The primary academic adoption of ISETBio and ISET3d is centered on modeling early visual encoding (corneal optics, wavefront aberrations, macular transmission, cone mosaic photopigment isomerizations, and phototransduction current).

Representative Papers & Research Areas

• Spatial Contrast Sensitivity & Computational Observers:

  • A computational-observer model of spatial contrast sensitivity: Effects of wavefront-based optics, cone-mosaic structure, and inference engine (Journal of Vision)
  • Focus: Replaces traditional analytical ideal-observer approximations with an image-computable, end-to-end biological pipeline.

• Visual Field Asymmetries & Eye Movements:

  • Modeling visual performance differences 'around' the visual field: A computational observer approach (PLOS Computational Biology)
  • Focus: Using ISETBio to evaluate how cone density variations, optical quality, and fixational drift account for performance differences across polar angles.

• Chromatic Aberration & Retinal Neural Mechanisms:

  • Neuronal Mechanism for Compensation of Longitudinal Chromatic Aberration in the Human Visual System (Frontiers in Bioengineering and Biotechnology)
  • Focus: Simulating wavelength-dependent blur on multispectral HDR natural scenes using ISETBio's wavefront optics models.

• Cone Mosaic Sampling & Retinal Pathologies:

  • The Relationship Between Visual Sensitivity and Eccentricity, Cone Density, and Mosaic Regularity (Investigative Ophthalmology & Visual Science)
  • Focus: Linking anatomical cone distributions and clinical perimetry to simulated ideal observer limits.

2. 3D Scene Ray Tracing & Physiological Optics

The integration of ISET3d with Physically Based Ray Tracing (PBRT) allows researchers to trace complex spectral scenes directly through anatomically accurate schematic eye models to calculate retinal irradiance.

Representative Papers & Research Areas

• 3D Retinal Image Formation:
  • Ray tracing 3D spectral scenes through human optics models (Journal of Vision)
  • Focus: Simulating depth-dependent defocus, chromatic aberration, and ocular transmittance from complex 3D environments onto a curved retinal surface.
• Foundational Architecture:
  • Visual encoding: Principles and software (Brainard Lab / ISETBio Overview)
  • Focus: Comprehensive documentation bridging calibrated display models, 3D scenes, and retinal ganglion cell responses.

3. Autonomous Driving, Sensor Prototyping & Machine Learning

In engineering and computer vision, ISETCam and ISET3d are used for sensor co-design and soft-prototyping—generating physically accurate synthetic sensor training data where real-world collection is unsafe or prohibitive.

Representative Papers & Research Areas

• Lens Modeling for 3D Camera Simulations:
  • Ray-transfer functions for camera simulation of 3D scenes (arXiv / Optics Express)
  • Focus: Generating polynomial ray-transfer functions from proprietary Zemax black-box lenses to enable physically accurate rendering inside PBRT/ISET without exposing lens intellectual property.
• High Dynamic Range (HDR) Night Driving Sensors:
  • ISETHDR: A Physically Accurate Synthetic Radiance Dataset for High Dynamic Range Driving Scenes (GitHub / ISET Dataset & Toolchain)
  • Focus: Evaluating multi-exposure and split-pixel (RGBW vs. multi-capture) HDR image sensors for low-light automotive vision and object detection networks.

4. Medical Imaging & Fluorescence Diagnostics

A growing area of soft-prototyping involves applying radiometric simulation to diagnostic biomedical hardware.

Representative Papers & Research Areas

• Endoscopic & Autofluorescence Systems:
  • Simulations of fluorescence imaging in the oral cavity (Biomedical Optics Express)
  • Focus: End-to-end radiometric modeling incorporating 3D tissue geometry, excitation-emission matrices of endogenous fluorophores, optical filters, and sensor architectures for early mucosal lesion screening.

Summary Matrix

Field / Community	Core Tool(s)	Primary Use Case
Computational Vision Science	ISETBio, ISET3d	Ideal observers, cone mosaic sampling, CSF modeling, spatial/color psychophysics
Physiological & Ophthalmic Optics	ISETBio, ISET3d	Wavefront aberrations, lens aging/transmittance, intraocular lens (IOL) simulation
Automotive & Computer Vision	ISETCam, ISET3d, ISETHDR	Sensor evaluation, split-pixel HDR benchmarking, synthetic ML training pipelines
Biomedical & Diagnostic Devices	ISETCam, ISET3d	Fluorescence modeling, filter/LED spectral tuning, soft-prototyping medical hardware
Display Engineering & AR/VR	ISETCam, ISETBio	Chromatic fringe compensation, foveated display metrics, perceptual colorimetry

Outreach & Promotion Takeaways

1. Vision Science Community: Emphasize ISETBio as the benchmark open-source computational observer platform that eliminates the need for arbitrary optical approximations.

2. Optics & Medical Device Engineers: Position ISETCam/ISET3d as a "virtual optical bench" capable of testing sensor and illumination designs prior to manufacturing physical prototypes.

3. Computer Vision & Sensor Architects: Frame ISET as an end-to-end radiometric bridge linking physics-based rendering (PBRT) directly to pixel-level sensor simulation and ISP pipelines.