<![CDATA[This study developed a gel electrophoresis image analysis system using digital image processing and the Structural Similarity Index (SSIM) to improve interpretative accuracy and reduce analysis time. Traditional laboratory methods often rely on subjective human observation, which can lead to significant diagnostic errors. The system, implemented in Python, streamlines the research workflow from image import and enhancement to band-intensity pattern analysis and automated lane comparison. Performance was evaluated using 10 standardized gel images, with results compared against a group of 15 students and domain experts (gold standard). The system achieved an average lane-detection accuracy of 80.00%, closely matching expert performance and significantly outperforming the student group (36.01%). Furthermore, the average analysis time was 14.96 seconds per image, approximately 3–4 times faster than manual interpretation. However, certain limitations were observed during testing, particularly a reduction in performance when analyzing images with excessive noise or poor band clarity, which occasionally affected lane detection and similarity assessment accuracy. Beyond laboratory efficiency, the system serves as an effective educational tool for high school students in biology and introductory bioinformatics education, reducing the barrier to entry for image analysis and computational biology concepts through an intuitive, low-cost interface. These findings demonstrate that the system enhances accuracy, speed, and consistency, offering a robust, scalable, and practical solution for both professional biological laboratories and modern science education.]]>
Copyrights © 2026
