Clinical Report: A Glimpse into the Future of Analytical Instrument Design?

Overview

Optical engineer Michael John Fanous describes Scanimus, a portable pathology scanner that continuously captures images at high speed and uses AI-based computational reconstruction to reduce motion blur. The system is intended to make slide digitization more practical, while its slower conventional scanning mode allows comparison with the continuous approach.

Background

The article reports that more than 90 percent of routine biopsies are not digitized, including in well-resourced healthcare systems. Fanous describes constraints in pathology workspaces involving space, time, and cognitive load, and says the scanner was designed to be compact and straightforward to use. Its development grew from academic work at the University of Illinois Urbana-Champaign and UCLA, including the BlurryScope project published in NPJ Digital Medicine. The article presents the system as an exploration of computational imaging and machine learning applied to pathology instrumentation.

Data Highlights

Key Findings

• Scanimus records video while the slide stage moves continuously, stopping only at the edges of the scan; the reported scanning speed is 10–20 mm/s.
• At high speed, motion blur affects image quality, particularly horizontal spatial frequencies.
• For model training, fast blurred scans are paired with much slower reference scans, acquired at approximately 50 μm/s, to provide high-quality ground truth.
• An image-to-image translation model is used to reconstruct sharp micrographs from blurred scans; Fanous describes the reconstruction as reliable, while acknowledging concerns about AI-generated detail.
• The system also offers a slower stop-and-stare mode for direct comparison, although its hardware does not provide the throughput of high-end commercial scanners in that mode.
• Fanous says interactions with the system can be recorded and modelled, potentially supporting workflows adapted to individual users; the article does not report clinical validation or outcomes for this capability.

Clinical Implications

The article describes a potential approach to reducing practical barriers to pathology slide digitization through compact hardware and continuous scanning, but it does not provide clinical validation or comparative diagnostic-performance results. Fanous identifies the slower conventional mode as an option for users who want a traditional acquisition approach, while noting its throughput limitation.

Conclusion

Scanimus combines high-speed continuous acquisition with AI-based image reconstruction in a portable pathology scanner concept. The source describes its design and intended workflow, but does not establish clinical performance or adoption outcomes.

Related Resources & Content

1. The Analytical Scientist, A Glimpse into the Future of Analytical Instrument Design? -- Interview with Michael John Fanous, year not provided
2. The Analytical Scientist, Our Open Source Future: Hardware, 2013
3. The Analytical Scientist, The Analytical Scientist Innovation Awards 2019: The Shape of Things to Come, 2019
4. The Analytical Scientist, The Greatest Analytical Odyssey: Part 1, 2024
5. the analytical scientist — A Bleak Future?
6. Guidances with Digital Health Content | FDA
7. Radiation Dose Reduction in CT Exams with Iterative and Deep Learning Reconstruction: A Systematic Review
8. Improved image quality and reduced acquisition time in brain MRI using deep learning-based reconstruction: A quantitative and subjective assessment compared to standard MPRAGE in 0.55 T MRI.