AI-enhanced endoscopic surgery

Say Goodbye to Blurry Endoscopies: The AI Tech Sharpening the Future of Surgery

Samir D’Costa in Tech & Innovation December 2025 • 5 min read.

"Discover how a novel image enhancement technique is poised to revolutionize minimally invasive surgery, offering clearer visuals and safer procedures."

Minimally invasive surgery has been revolutionized through stereoscopic endoscopes, which provide surgeons with the ability to visualize organ surfaces and use surgical tools with increasing accuracy. A persistent obstacle in endoscopic surgery is the issue of insufficient and irregular light sources. This issue significantly affects image quality and occasionally causes surgical tools to be hardly apparent in low-light environments.

Images taken in low-light conditions are more likely to have a low signal-to-noise ratio and metrication artifacts brought on by quantization errors. Consequently, conventional image enhancement techniques frequently worsen noise in low-light areas, which compromises image clarity and surgical precision. Current methods frequently amplify existing noise during this process, reducing their efficacy.

This article explores a groundbreaking method designed to improve endoscopic image quality by identifying distinct illumination zones and tailoring enhancement strategies according to desired visual results. This novel technique dramatically reduces noise amplification during image processing, especially in previously challenging low-light circumstances, in contrast to existing enhancement approaches. This could be especially useful in preventing damage to sensitive organs like the liver and spleen.

How Does AI Enhance Endoscopic Images?

The key lies in a refined approach to image processing, which addresses typical challenges such artifacts, noise, and uneven illumination. The method begins by evaluating the source and nature of the image's lighting environment, and then tailors processes to enhance image clarity, reduce artifacts, and limit noise. This process relies on several steps:

The new method improves on current image enhancement methods by improving visibility in endoscopies, maintaining natural picture quality, and minimizing the amplification of noise and artifacts. The outcomes are enhanced image quality that can function as a crucial first step for other image processing techniques by addressing common limitations.

Illumination Region Identification: The algorithm begins by segmenting the endoscopic image into three distinct regions based on lighting conditions: well-lit, low-light, and lossy (extremely low light with significant detail loss). This is achieved by analyzing the V-space element from the HSV (Hue, Saturation, Value) transformation of the image, using predefined thresholds to categorize each pixel.
Base and Detail Layer Decomposition: After identifying the illumination regions, the image is divided into a base layer, representing smooth luminance variations, and a detail layer, capturing local contrasts and fine details. This decomposition is performed using an edge-preserving smoothing filter, such as the tree filter, which efficiently separates the image components while preserving important structural details.
Enhancement Factor Application: Enhancement factors are then applied to both the base and detail layers, tailored to each illumination region. The well-lit regions are largely preserved to maintain their natural appearance, while the low-light regions undergo a gamma correction-like enhancement to improve luminance. In the lossy regions, a linear gain is applied to suppress noise amplification, ensuring that only genuine details are enhanced.
Adaptive Smoothing: To further refine the enhancement process, the pseudo enhancement factors for both base and detail layers are smoothed using Gaussian kernels. This step helps to enforce a piece-wise linear assumption of the illumination, reducing artifacts and ensuring a seamless transition between different enhancement levels.
Image Reconstruction: Finally, the enhanced base and detail layers are combined to produce the final enhanced image. This reconstructed image exhibits improved visibility in low-light areas, reduced noise and artifacts, and an overall natural appearance.

In practical terms, AI algorithms analyze endoscopic images, classifying areas into well-lit, low-lit, and lossy regions. For areas determined to be adequately lit, the algorithm works to maintain existing visual features and refine the comparison of existing image characteristics. In low-lit regions, the algorithm boosts brightness and contrast to bring hidden details into view. Special measures are applied to the base and detail layers of lossy regions to suppress noise amplification while preserving underlying details.

The Future of Surgical Imaging

The innovative method not only improves image quality but also lays the groundwork for more sophisticated image processing applications in surgery. The results show that this method improves endoscopic images by preserving contrast and color in well-lit areas while improving low-light visibility. This leads to more natural-looking images with less noise. This method outperforms existing technologies, increasing its effectiveness in endoscopic procedures.

About this Article -

This article was crafted using a human-AI hybrid and collaborative approach. AI assisted our team with initial drafting, research insights, identifying key questions, and image generation. Our human editors guided topic selection, defined the angle, structured the content, ensured factual accuracy and relevance, refined the tone, and conducted thorough editing to deliver helpful, high-quality information.See our About page for more information.

This article is based on research published under:

DOI-LINK: 10.1049/htl.2018.5067, Alternate LINK

Title: Endoscopic Image Enhancement With Noise Suppression

Subject: Health Information Management

Journal: Healthcare Technology Letters

Publisher: Institution of Engineering and Technology (IET)

Authors: Wenyao Xia, Elvis C.S. Chen, Terry Peters

Published: 2018-09-14

Everything You Need To Know

What is the primary challenge in endoscopic surgery that this AI-powered method addresses?

The primary challenge addressed is the issue of insufficient and irregular light sources within endoscopic procedures. This issue significantly affects image quality, leading to low signal-to-noise ratios, metrication artifacts, and making surgical tools difficult to see in low-light environments. This method aims to overcome these limitations to enhance image clarity and improve surgical precision.

How does the AI-powered image enhancement method improve upon existing techniques in endoscopic surgery?

The AI method distinguishes itself by tailoring enhancement strategies based on identified illumination zones within the endoscopic image. Unlike existing techniques that often worsen noise, this method dramatically reduces noise amplification, particularly in low-light areas. This is achieved through a multi-step process that includes Illumination Region Identification, Base and Detail Layer Decomposition, Enhancement Factor Application, Adaptive Smoothing, and Image Reconstruction. This leads to enhanced image quality while preserving the natural appearance of well-lit areas.

Describe the 'Illumination Region Identification' step in the AI image enhancement process. How does it work, and why is it important?

The Illumination Region Identification step is crucial because it allows the AI to understand the light environment within the endoscopic image. The algorithm segments the image into three distinct regions: well-lit, low-light, and lossy. This is achieved by analyzing the V-space element from the HSV (Hue, Saturation, Value) transformation of the image, using predefined thresholds to categorize each pixel. Based on this identification, the AI applies targeted enhancement strategies. This is important because it avoids applying uniform enhancements that can amplify noise or distort well-lit areas.

Explain the purpose of 'Base and Detail Layer Decomposition' and 'Enhancement Factor Application' in the AI image enhancement process. How do these steps contribute to the overall goal?

The 'Base and Detail Layer Decomposition' separates the image into two layers: a base layer representing smooth luminance variations, and a detail layer capturing local contrasts and fine details. This is done using an edge-preserving smoothing filter, which preserves structural details. Following this, 'Enhancement Factor Application' applies tailored enhancements to each illumination region. Well-lit areas are largely preserved, low-light regions get a gamma correction-like boost, and lossy regions receive a linear gain to suppress noise. These steps together contribute to the overall goal by allowing the AI to refine and enhance specific image characteristics. It ensures that improvements are applied judiciously, maximizing clarity and minimizing artifacts, while preserving natural image characteristics.

What are the potential benefits of this AI-powered image enhancement technique for surgical outcomes and other applications?

This AI-powered method enhances image quality by improving visibility in low-light areas, reducing noise, and maintaining the natural appearance of images, which translates to a more accurate surgical experience. The ability to clearly visualize organ surfaces and surgical tools is a significant advantage. By addressing common limitations of current methods, it enhances the effectiveness of endoscopic procedures and potentially reduces the risk of damaging sensitive organs. This enhancement also sets the foundation for more advanced image processing applications, such as real-time surgical guidance systems and automated analysis tools, which can lead to safer and more precise surgical interventions.