The Wadsworth Center at the New York State Department of Health and its Advanced Light Microscopy and Image Analysis core facility recently hosted its annual Imaging Workshop, welcoming researchers, imaging specialists, and graduate students eager to expand their microscopy and image analysis skills.
This intensive five-day course blends foundational theory with practical application. Lectures introduce key microscopy concepts, while lab sessions give participants the opportunity to apply what they've learned in real time using a range of imaging techniques and tools. The workshop also provided an excellent opportunity to engage with emerging scientists as they build the strong foundational skills needed for their research careers.
Media Cybernetics had the honor of presenting two lectures at the event: one on deconvolution, delivered by Senior Software Engineer Andrew Molnar, and another on Convolutional Neural Networks (CNNs), presented by CEO Nick Beavers.
Workshop topics included:
- The microscope: Light paths through the microscope, image formation, cameras, and LASERS
- Image processing: AI, image visualization, ethical considerations, and deconvolution
- Principles & practices of specialized microscopy: fluorescence microscopy, confocal microscopy, super-resolution (nanoscopy), light sheet microscopy, Fluorescence Recovery After Photobleaching (FRAP), and Total Internal Reflection Fluorescence (TIRF)
Figure 1. Andrew Molnar presenting on deconvolution at the NYSDOH Wadsworth Center Imaging Workshop.
Restoring Image Fidelity with Deconvolution
Andrew Molnar, Senior Software Engineer at Media Cybernetics, delivered a lecture on deconvolution, an advanced computational technique that enhances image clarity by reversing the optical blurring introduced during image acquisition.
What is deconvolution?
Deconvolution is a mathematical process used to sharpen blurred microscopy images by compensating for distortions caused by the microscope’s optics. This technique can dramatically improve resolution and contrast, making it easier to detect fine structural details that might otherwise be lost.
Molnar discussed the underlying principles behind deconvolution and demonstrated how proper application of the technique can reveal subtle image features, enhance resolution, and increase the accuracy of quantitative analysis.
Clear, high-quality images are essential for producing reliable scientific results. By unlocking hidden detail and improving measurement precision, deconvolution has become a critical tool in the modern microscopy workflow.
Figure 2. A collage of microscopy images showcasing a range of imaging techniques. (Image courtesy of NYSDOH Wadsworth Center, 2025)
Convolutional Neural Networks (CNNs): The Engine Behind Modern AI Segmentation
Nick Beavers, CEO of Media Cybernetics, presented a lecture on Convolutional Neural Networks (CNNs)—a foundational deep learning approach used in modern image analysis.
What are Convolutional Neural Networks (CNNs)?
Convolutional Neural Networks (CNNs) are a type of deep learning model that excels at recognizing patterns in microscopy images. They can automatically identify features like cell structures, organelles, or particles—eliminating the need for manual thresholding or filtering. CNNs enable faster, more accurate image segmentation and classification, improving both efficiency and reproducibility in research workflows.
His session covered how CNNs are trained to recognize complex visual patterns, how they differ from classical algorithms, and how they’re increasingly being used in microscopy for tasks like segmentation, classification, and feature extraction.
By breaking down how CNNs process image data, Beavers gave attendees a clear, practical introduction to the role of AI in transforming scientific imaging workflows.
Figure 3. Schematic of a deep neural network with stacked layers and dense links for hierarchical image learning.
Figure 4. Nick Beavers of Media Cybernetics captures a selfie with the company’s educational posters on display at the NYSDOH Imaging Workshop.
Building the Next Generation of Imaging Experts
The combination of lectures and hands-on labs helps early-career scientists build confidence and competence in microscopy techniques. Events like this workshop are critical for training researchers who will advance scientific discovery through better imaging.
Figure 5. Participants and instructors of the 2025 NYSDOH Wadsworth Center Imaging Workshop gather for a group photo. Not pictured: Nick Beavers. (Photo courtesy of NYSDOH Wadsworth Center, 2025)
![[Webinar Replay] Particle Analysis Errors Costing You? Protect QC with Deep Learning Automation (in 1 Hour!)](https://mediacy.com/wp-content/uploads/2025/06/Particle-Analysis-Webinar-2025-06-Webinar-Replay-Blog-5-1140x740.jpg)
Media Cybernetics
sales@mediacy.com
Related Links