The research, led by B. V. Divyashree of the Mysore University School of Engineering together with Shamair Khushba and G. Hemantha Kumar of the University of Mysore, deliberately charts a middle course between two dominant paradigms in digital pathology. On one side stand handcrafted, interpretable feature-based methods; on the other, data-hungry deep learning architectures such as convolutional neural networks, ResNet, DenseNet, and EfficientNet. Deep models have produced impressive results, but they typically demand large annotated datasets, substantial GPU memory, and careful hyperparameter tuning. By contrast, the pipeline proposed by the Mysore team relies on classical image processing and machine learning, offering transparency and computational efficiency that make it attractive for laboratories and clinics where data or hardware are limited.

The pipeline unfolds in three hierarchical stages: preprocessing, nuclei segmentation, and classification. Preprocessing itself begins with enhancement. The authors tested two complementary techniques on the raw images. The first is Gaussian filtering, a noise-reduction method that convolves the image with a kernel whose weights follow a Gaussian distribution, assigning the greatest weight to pixels near the kernel’s center. Because histopathological micrographs frequently contain Gaussian noise, this filter suppresses high-frequency artifacts while preserving the global intensity variations needed for later segmentation. The second technique, decorrelation stretch, addresses a different problem: in most histopathology images the red, green, and blue color channels are highly correlated and carry overlapping information, muting the color contrast that carries diagnostic meaning. The method rotates and rescales the pixel data into a new space where the channels become less correlated, then applies an inverse transformation back to the original space, producing brighter, more spectrally distinct images.

To decide objectively which enhancement method deserved a place in the pipeline, the researchers evaluated both using two quantitative metrics. Peak Signal-to-Noise Ratio, or PSNR, expresses the relationship between the maximum power of a signal and the power of the corrupting noise, while Root Mean Square Error, or RMSE, quantifies the pixel-wise difference between the enhanced image and the original reference. Higher PSNR and lower RMSE indicate better quality. Across images captured at magnifications of 40X, 100X, 200X, and 400X, and for both benign and malignant samples, the Gaussian filter delivered the highest PSNR and the lowest RMSE. Although decorrelation stretch produced visually stronger color separation, it could not match the Gaussian filter’s fidelity, so Gaussian-filtered images were carried forward into the segmentation stage.

The next challenge was isolating cell nuclei, the structures whose shape, size, and texture betray whether tissue is healthy or cancerous. In normal tissue, nuclei tend to be round and uniform; in malignant tissue they appear irregular and vary widely in size. Poor segmentation that bleeds in surrounding cytoplasm or stroma, or that merges overlapping nuclei, can distort these cues and doom the classifier downstream. The authors’ approach begins by converting the enhanced RGB image into the YCbCr color space, which decouples luminance, the Y channel representing brightness, from chrominance, the Cb and Cr channels encoding blue-difference and red-difference respectively. This separation makes segmentation robust to the staining variations and lighting inconsistencies that plague routine slides. Contrast Limited Adaptive Histogram Equalization, or CLAHE, is then applied only to the Y component. Unlike conventional histogram equalization, CLAHE operates on small local tiles and applies a contrast-limiting function, sharpening local detail without amplifying noise or over-enhancing the image, while the untouched chrominance channels preserve color fidelity.

With contrast enhanced, the pipeline applies K-means clustering to partition each pixel into one of three clusters, a value chosen empirically. The algorithm iteratively assigns pixels to the nearest cluster centroid and recomputes centroids until convergence, effectively dividing the image into background, normal tissue, and suspicious regions. One cluster corresponds to the nuclei, which are thereby separated from stroma, cytoplasm, and other cellular components. The authors report that the segmentation shows consistent uniformity and strong correlation with the underlying tissue structures, though regions where nuclei overlap remain a recognized challenge. Notably, because the BreakHis dataset used for validation provides ground-truth labels only for classification and not for segmentation, the segmentation module was assessed qualitatively rather than against annotated masks.

From the segmented nuclei, the team extracted a rich set of handcrafted features spanning three families. Texture features were computed from the Gray-Level Co-Occurrence Matrix, or GLCM, using two matrices that capture horizontal and vertical pixel relationships; these yield quantitative descriptors such as homogeneity, energy, correlation, contrast, and entropy that summarize the spatial arrangement of cells within tissue structures. Geometric features, including area, diameter, radius, perimeter, eccentricity, elongation, and solidity, characterize the shape variability that distinguishes cancerous nuclei from their benign counterparts. Finally, color features were drawn from the Cb and Cr components of the enhanced YCbCr images. The authors emphasize that these interpretable, handcrafted features were chosen deliberately over learned representations because they explicitly characterize tissue properties, enhance clinical explainability, and remain effective in limited-data scenarios where deep networks would struggle.

Classification was performed with two of machine learning’s most dependable workhorses: Support Vector Machine and K-Nearest Neighbor. The SVM, equipped with a radial basis function kernel selected empirically for its ability to capture non-linear relationships in the feature space, achieved the stronger results, reaching approximately 87.5 percent accuracy on the BreakHis dataset and outperforming both the K-NN baseline and many prior methods in the literature. For K-NN, the number of neighbors was varied from 1 to 15, with optimal performance at k equal to 5 using Euclidean distance, a setting that balances the low bias but high variance of small k against the increased bias of large k. Critically, the researchers adopted a patient-level split strategy: because BreakHis contains multiple images at different magnifications from the same patient, all images from a single patient were assigned exclusively to either the training or testing set, in an 80:20 ratio, preventing data leakage that can inflate reported accuracy. Performance was assessed with accuracy, sensitivity, specificity, precision, recall, and F-score.

The implications reach beyond a single benchmark score. In head-to-head comparison with representative prior work on BreakHis, the proposed pipeline achieved the highest reported accuracy while offering improved interpretability, lower computational complexity, and suitability for small-scale datasets without extensive GPU resources. An ablation study varying filtering techniques, color spaces, and clustering parameters confirmed that the integration of Gaussian enhancement, YCbCr conversion with CLAHE, and three-cluster K-means segmentation justifies each design choice. The authors acknowledge the approach’s limits: handcrafted features may miss highly complex patterns, and some nuclei remain over-segmented, leaving room for error reduction. Future work will integrate deep learning-based feature extraction for end-to-end learning and test robustness on additional datasets. For now, the study offers a compelling demonstration that careful, interpretable engineering, rather than sheer model size, can push computer-aided diagnosis of breast cancer meaningfully forward, with the potential to speed up pathology workflows and reduce diagnostic variability where it matters most.

Subject of Research: Machine learning classification of breast histopathology images into benign and malignant using a hierarchical preprocessing and segmentation pipeline

Article Title: A hierarchical pipeline for preprocessing and classification of breast histopathology images into benign and malignant

Article References: Divyashree, B. V., Khushba, S., & Kumar, G. H. (2026). A hierarchical pipeline for preprocessing and classification of breast histopathology images into benign and malignant. Discover Informatics, 1(1), Article 11. https://doi.org/10.1007/s44564-026-00013-2

Image Credits: AI Generated

DOI: 10.1007/s44564-026-00013-2

Keywords: breast cancer, histopathology, machine learning, SVM, K-means clustering, image segmentation, CLAHE, YCbCr, Gaussian filter, BreakHis dataset, computer-aided diagnosis, medical imaging

Unlike previous studies, the PRISM Trial aims to serve thousands of women across multiple states, including California, Florida, Massachusetts, Washington, and Wisconsin. By analyzing hundreds of thousands of mammograms from a network of academic medical centers and breast imaging facilities, this large-scale investigation is designed to generate robust data that can inform clinical practices and policies surrounding breast cancer screening. The AI tool under examination is Transpara, developed by ScreenPoint Medical, and its integration into clinical workflows will be facilitated by the Aidoc aiOS platform, setting the stage for a comprehensive evaluation of its capabilities.

A primary innovation of the PRISM trial is its commitment to patient-centered research. This initiative is not merely a collaboration among researchers and clinicians; it has been developed in close partnership with patient advocates, health system leaders, and policymakers. The input and perspectives from these diverse stakeholders will be invaluable in shaping the design and impact of the study. Patient experience remains paramount, as mammograms will continue to be conducted in accordance with existing clinical protocols, ensuring that there is no disruption to routine screening practices.

Within the framework of the trial, mammograms will be assigned randomly to one of two groups: those interpreted solely by radiologists and those analyzed with AI support. Notably, in every instance, a qualified radiologist will retain the ultimate authority, reading the exam and rendering the final decision regarding diagnosis. This collaborative approach underscores the intention to evaluate how AI may serve as an adjunctive tool that enhances, rather than replaces, human expertise.

Breast cancer remains a prominent health concern, constituting one of the leading causes of cancer-related mortality among women in the United States. Despite the benefits of routine mammography in reducing mortality through early detection, several challenges persist. Among these challenges are the potential for false-positive results, which can lead to unnecessary additional testing, heightened anxiety, and increased healthcare costs. Furthermore, there are instances where significant cancers may be missed altogether, resulting in devastating consequences for patients. The PRISM trial endeavors to assess whether AI can play an integral role in mitigating these challenges, from improving cancer detection rates to reducing the frequency of false positives.

Dr. Joann G. Elmore, Dual Principal Investigator and professor of medicine at UCLA, emphasizes the trailblazing nature of this study: “This is the first large-scale randomized trial of AI in breast cancer screening in the United States.” Her remarks highlight the importance of careful and objective examination to determine whether AI genuinely assists in augmenting radiologist performance or merely increases the number of flagged exams that turn out to be normal. The findings from this study have the potential to significantly influence clinical decision-making and the future integration of AI technology in routine breast cancer screening.

In addition to evaluating the accuracy of cancer detection and recalls, this trial will also utilize focus groups and surveys to delve into the perceptions and trust levels of both patients and radiologists regarding AI-assisted care. Understanding the human experiences associated with AI technology will be crucial in determining how best to implement these advanced systems within healthcare environments.

The collaborative effort in the PRISM trial spans seven leading academic medical centers, showcasing the broad commitment to advancing cancer care across the country. The diverse institutions involved, from UCLA and UC Davis to medical centers in Florida, Massachusetts, and Washington, exemplify a united front in addressing the challenges of breast cancer screening. This multi-institutional approach not only fosters a rich exchange of ideas but also enhances the robustness of the data being collected, ensuring that the findings will be meaningful and applicable across a range of healthcare contexts.

The integration of AI into healthcare raises both opportunities and challenges. While AI holds considerable promise for improving patient outcomes, it also raises essential questions about its real-world effects and implications. Dr. Elmore stresses that rigorous trials like the PRISM initiative are essential for generating trustworthy evidence regarding AI’s utility in clinical practice. The potential of AI to streamline workflows, reduce human error, and enhance diagnostic precision is exhilarating, yet it requires careful validation to ensure that these technologies genuinely provide value without inadvertently compromising patient safety or care quality.

Ultimately, the PRISM trial aims to strike the right balance between human expertise and AI support. The goal is not to supplant the invaluable role of expert radiologists but rather to explore how AI can function as an effective co-pilot in the interpretation of mammograms. Dr. Elmore describes the initiative as an effort to provide expert radiologists with additional resources while maintaining their critical responsibilities in the diagnostic process. This partnership between AI and human judgment is likely to yield new insights into the future of breast cancer screening and may redefine best practices on national and international levels.

Through these groundbreaking efforts, the UCLA and UC Davis-led PRISM Trial aspires to inform future policy decisions and integrate emerging technologies into patient care effectively. The results will likely resonate far beyond the realm of breast cancer screening, paving the way for innovative applications of AI technology in various medical specialties. By prioritizing patient-centered design and maintaining extensive collaboration, this research is poised to make meaningful contributions to the field of oncology, ultimately enhancing care for patients facing the challenges of breast cancer diagnosis and treatment.

The PRISM Trial exemplifies a meticulous approach to intertwining technology with compassionate healthcare delivery, with the primary aim of enhancing patient outcomes while also alleviating the pressures faced by healthcare providers. As the study unfolds over the coming months and years, the insights gained will not only illuminate critical aspects of AI’s role in mammography but could also serve as a model for future integrative studies exploring the intersection of technology and patient care across all domains of medicine.

This is a momentous endeavor, and its findings could resonate broadly, offering hope to many and possibly establishing a new standard for breast cancer screening and care. The collaboration highlights an important juncture in medical history, where technology and the pursuit of improved patient outcomes converge, potentially leading to transformative changes in breast cancer screening practices.

Subject of Research: The role of artificial intelligence in improving mammogram interpretation
Article Title: UCLA and UC Davis Launch PRISM Trial to Enhance Breast Cancer Screening with AI
News Publication Date: October 2023
Web References:
References:
Image Credits:

Keywords