The study, published in Discover Artificial Intelligence by Harish Kumar, Anuradha Taluja, N. Partheeban, Shashi Mehrotra and colleagues, introduces a Confidentiality-Preserving Spiking Neural Transformer, or CP-SNT, designed to detect breast cancer from histopathology images while keeping sensitive medical records locked away from prying eyes. The motivation is straightforward: most AI diagnostic systems chase accuracy alone, while noise, staining variation, structural heterogeneity, and class imbalance quietly erode their reliability, and centralized cloud training or inference exposes raw images to breach and unauthorized access. Regulatory frameworks increasingly demand strict confidentiality for medical images, and the authors argue that privacy is no longer an optional add-on but a core requirement for any medical AI system that hopes to reach the clinic. CP-SNT was built to answer accuracy, robustness, uncertainty modeling, and data security in a single end-to-end pipeline rather than treating them as separate engineering problems.

The journey of an image through the framework begins with cleanup. Histopathology slides are notoriously corrupted by uneven illumination, staining artifacts, and sensor noise, and conventional smoothing filters tend to blur the very cellular boundaries that pathologists rely on. The researchers therefore apply a Robust Double-Weighted Guided Image Filtration module, an edge-preserving denoising method that computes local linear coefficients through a regularized least-squares minimization. In practical terms, the filter assumes a local affinity between a guidance image and the input, allowing homogeneous regions to be smoothed while fine morphological features such as nuclear boundaries, glandular structures, and tissue architectures are preserved. The team also explored content-aware seam carving to strip away low-information background regions, but their experiments showed that structural fidelity, measured by the SSIM metric, degrades noticeably once more than ten seams are removed, so aggressive seam removal was rejected as unsuitable for diagnostic preprocessing.

Once the image is clean, the framework moves into feature extraction using a Discrete Quaternion Quadratic Phase Fourier Transform, a mouthful of a name that hides an elegant idea. Instead of processing the red, green, and blue channels of a stained slide independently, as classical Fourier methods do, the quaternion representation encodes all three channels as a single hypercomplex value, R i plus G j plus B k. This allows the transform to capture inter-channel color dependencies and phase correlations simultaneously, which is crucial because staining intensity varies from lab to lab and slide to slide. The resulting frequency- and phase-domain coefficients preserve magnitude and phase information together, yielding compact discriminative features that are more robust to staining variability and texture heterogeneity than raw pixel intensities. In effect, the transform converts the visual complexity of tissue into a mathematical fingerprint that downstream networks can learn from more reliably.

The heart of the system is a hybrid classifier that fuses two very different computational philosophies. The first component is a Probabilistic Spiking Neural Network, a biologically inspired model in which neurons communicate through discrete spikes rather than continuous activation values. Each neuron maintains a membrane potential governed by a decay dynamic, and instead of firing deterministically when a threshold is crossed, it fires stochastically with a probability given by a sigmoid function of its membrane potential. In a worked example from the paper, a feature vector encoding texture intensity, edge sharpness, and color variation is weighted and summed to a potential of 1.34, producing a spike probability of roughly 0.79; a random draw below that value triggers a spike, above it the neuron stays silent. This probabilistic, sparse encoding explicitly models uncertainty, resists noisy feature variation, reduces overfitting, and emphasizes salient discriminative patterns while suppressing redundant activations, which is precisely the kind of behavior medical diagnosis demands when ambiguity is routine.

The second component, a Causality-Aware Transformer Network, takes the spike-encoded embeddings and applies causal self-attention, computing query, key, and value matrices scaled by the square root of the feature dimension. Where the spiking network excels at local, uncertainty-aware encoding, the transformer captures long-range contextual dependencies across the tissue patch and models causal interactions between features, helping the model avoid spurious correlations that plague purely correlational attention. The authors also incorporate metaheuristic optimization, drawing on the Farmer Ants Optimization Algorithm for feature selection and hyperparameter tuning, balancing exploration and exploitation to speed convergence and avoid the fragility of manual tuning. Ablation-style analysis attributes gains to each stage: the guided filtration contributes about 1.28 percent, the quaternion transform about 0.83 percent, the spiking network a substantial 1.45 percent, and the causal attention module about 0.32 percent, while the encryption layer preserves privacy without measurable loss of accuracy.

The privacy layer is where the framework earns its confidentiality-preserving name. A module called PrivTuner applies homomorphic encryption, a family of cryptographic techniques that allow arithmetic to be performed directly on ciphertext, so that model parameters and inference data can be processed without ever being decrypted. This means training gradients and diagnostic predictions can be computed on encrypted representations of patient images, shielding records from cloud operators, network intermediaries, and potential attackers alike. The team measured encryption and decryption time, inference latency, memory usage, and communication overhead, and reported that the added cost is modest relative to the security gained, though they candidly note that a comprehensive trade-off analysis of encryption overhead, computational complexity, and scalability remains future work. This distinguishes the approach from most prior privacy-conscious designs, which rely on federated learning or blockchain architectures rather than embedding encrypted-domain computation inside a modern hybrid deep learning model.

The experimental evidence is compelling. Evaluated on two public benchmarks, BreakHis and the Breast Histopathology Images dataset, together totaling 16,116 images spanning magnifications from 40x to 400x and patch-level samples of benign and malignant tissue, CP-SNT achieved accuracies of 98.60 percent and 98.37 percent respectively, with precision above 98.1 percent and F1-scores near 98.2 percent. Five-fold cross-validation yielded a mean accuracy of 98.62 percent with a standard deviation of only 0.23, indicating strong stability and resistance to overfitting, and paired t-tests confirmed that improvements over CNN-based, transformer-based, and hybrid baselines were statistically significant at P less than 0.05. Crucially, the model was also tested on an external Invasive Ductal Carcinoma dataset that played no role in development or tuning, where it reached a validation accuracy of 95.73 percent, test accuracy of 95.41 percent, precision of 98.83 percent, recall of 93.89 percent, an F1-score of 96.30 percent, and an ROC-AUC of 98.45 percent, demonstrating genuine generalization beyond the datasets it was built on.

The authors are appropriately candid about limitations. Validation so far rests on public datasets rather than large multicenter clinical cohorts, the multi-module architecture is computationally heavier than simpler deep learning methods, and the system currently performs binary benign-versus-malignant classification without tumor subtyping or built-in explainability. Borderline tissue patterns and complex morphologies remain sources of error, and the encryption overhead analysis is still preliminary. Future work, they write, will target heterogeneous clinical data, deeper privacy-performance trade-off studies, explainable AI techniques, multi-class subtype classification, and real-time clinical deployment. Even so, the study marks a notable convergence of two research frontiers that have largely progressed in parallel: neuromorphic, spike-based computing that mimics the brain’s event-driven efficiency, and cryptographic computation that treats patient privacy as a first-class architectural constraint. If the framework’s accuracy holds up in clinical settings, it suggests a future in which the most sensitive medical images can be analyzed by powerful AI without ever being exposed in the clear, a combination that could reshape how hospitals, laboratories, and cloud providers collaborate on the next generation of diagnostic intelligence.

Subject of Research: A privacy-preserving spiking neural transformer framework for breast cancer detection in histopathology images using homomorphic encryption

Article Title: Spiking neural transformer based confidentiality preserving framework for breast cancer detection in histopathology images

Article References: Spiking neural transformer based confidentiality preserving framework for breast cancer detection in histopathology images. (n.d.). https://doi.org/10.1007/s44163-026-02179-y

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02179-y

Keywords: breast cancer detection, histopathology, spiking neural network, transformer, homomorphic encryption, privacy-preserving AI, medical imaging, deep learning, uncertainty modeling, BreakHis dataset, causal attention, secure healthcare analytics

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

The work, led by Nikita Grewal and Yash Arora together with S. K. Gupta and Sanjeev Kumar, builds on the twin support vector machine, a widely recognized classification model first introduced as a faster, leaner cousin of the classical support vector machine. Rather than searching for a single separating hyperplane, the twin version constructs two nonparallel hyperplanes, one for each class, and assigns each point to whichever hyperplane it sits closer to. This structural difference typically makes training faster than the classical formulation, since each optimization problem involves the samples of only one class at a time. But the twin model has well-documented weaknesses: it is sensitive to noise and outliers, and when class sizes are unequal it can produce systematically biased decision boundaries that favor the dominant class.

The Roorkee team’s central insight is that not all samples deserve equal trust. In real datasets, some points sit comfortably in the heart of their class while others linger near the boundary or belong to scattered secondary clusters, and a few are outright errors. To capture this heterogeneity, CPLTSVM assigns each training sample a fuzzy membership weight derived from its proximity to the nearest class centers. Instead of assuming every class forms a single compact blob, the model adopts a polycentric view: each class may have several centers, reflecting the multi-cluster structure that naturally arises in high-dimensional data. Samples close to any center of their own class receive high membership values, meaning they exert strong influence on the learned hyperplane, while distant or structurally isolated points receive lower weights and therefore contribute less to the final decision rule. This mechanism embeds global structural information about the dataset directly into the training objective, something conventional support vector formulations ignore entirely.

The fuzzy weighting scheme draws on a long intellectual lineage. Fuzzy sets, introduced by Lotfi Zadeh in 1965, allow data points to belong to categories to intermediate degrees rather than in a binary fashion, and researchers have spent decades folding such gradations into kernel-based classifiers. Earlier fuzzy twin support vector machines have used information entropy, affinity measures and intuitionistic fuzzy memberships to dampen the effect of noisy points. What distinguishes the new approach is the combination of two complementary signals. The polycentric membership captures how representative a point is of its class structure, while a separate class probability estimate captures how confidently a point belongs to its assigned class at all.

That probability estimate comes from the k-nearest-neighbor technique, one of the oldest and most intuitive tools in pattern recognition. For each training sample, the model examines its closest neighbors and computes the fraction that share its label, yielding a local estimate of class membership probability. Points surrounded by neighbors of the same class earn high probability scores; points in mixed neighborhoods earn low ones. When these probabilities are folded into the fuzzy membership function, noisy samples and borderline outliers are automatically down-weighted, reducing their power to drag the decision boundary away from the true underlying structure. The result is a classifier that is robust by construction rather than by post-hoc correction.

Computational efficiency was a second design priority. Standard twin support vector machines solve two quadratic programming problems, which can be expensive on large datasets. By integrating the fuzzy membership weights into the least squares twin support vector machine formulation, which replaces the inequality constraints of the original with equality constraints and a squared loss, the authors convert training into the solution of systems of linear equations. This dramatically lowers the computational cost while preserving the benefits of the fuzzy weighting scheme, making the approach practical for the kinds of large, messy datasets where imbalance problems are most acute.

To evaluate the framework, the team ran experiments on twenty benchmark imbalanced datasets, testing both linear and nonlinear kernel versions of CPLTSVM against several state-of-the-art competitors. Performance was measured with three metrics suited to imbalanced settings: the area under the receiver operating characteristic curve, known as AUC, the geometric mean of class sensitivities, known as G-mean, and the F-measure, which balances precision and recall. The numbers are striking. The model achieved an average AUC of 89.71 percent with the linear kernel and 94.87 percent with the nonlinear kernel, and it consistently surpassed the existing techniques across the benchmark suite. The authors supplemented raw scores with comprehensive statistical analyses to confirm that the observed gains were significant rather than artifacts of dataset selection.

The researchers also demonstrated the model’s practical value on a task with real clinical stakes: breast cancer diagnosis from histopathological images using the BreakHis dataset. On this medical imaging benchmark, CPLTSVM delivered superior performance compared with baseline methods, suggesting the technique could support real-world classification pipelines where false negatives carry severe consequences. Because histopathological datasets are naturally imbalanced and noisy, with subtle visual differences between benign and malignant tissue, this application exercises exactly the weaknesses the new membership scheme was designed to address. The authors have made the source code publicly available on GitHub, allowing other researchers to reproduce the results and extend the framework.

The significance of the work extends beyond a single benchmark victory. Machine learning systems are increasingly deployed in settings where rare events dominate the consequences: detecting defective components in manufacturing lines, flagging fraudulent transactions, spotting early signals of disease, and identifying anomalies in wind turbine gearboxes, a domain where twin support vector machines have previously found application. Every one of these settings shares the same structural pathology that the Roorkee team targeted, and a classifier that resists both noise and imbalance without requiring delicate resampling or costly retraining could simplify deployment considerably. Compared with popular alternatives such as the SMOTE synthetic oversampling technique or cost-sensitive learning frameworks, which modify the data or the loss rather than the classifier’s view of structure, the fuzzy membership approach works within the learning algorithm itself, adjusting how much each sample matters based on where it sits in the geometry of the data.

There are, as with any method, avenues for future work. The polycentric membership depends on identifying class centers, and the k-nearest-neighbor probability estimation introduces a hyperparameter in the choice of k, questions of sensitivity that practitioners will want to probe. The evaluation here focuses on binary classification, the native territory of twin support vector machines, and extending the framework to multicategory problems, as earlier fuzzy proximal approaches have done, would broaden its reach. Still, the combination of strong empirical results, computational efficiency and demonstrated medical utility marks CPLTSVM as a notable advance in the ongoing effort to make classifiers trustworthy when the world refuses to hand us balanced data. As algorithms increasingly inform decisions about health, security and safety, techniques that give quiet, underrepresented samples their due weight are not merely academic refinements; they are prerequisites for reliable artificial intelligence.

Subject of Research: A fuzzy least squares twin support vector machine using class probability and polycentric memberships to improve classification of imbalanced datasets

Article Title: Class probability and polycentric fuzzy least squares twin support vector machine for imbalanced data

Article References: Grewal, N., Arora, Y., Gupta, S. K., & Kumar, S. (2026). Class probability and polycentric fuzzy least squares twin support vector machine for imbalanced data. Knowledge and Information Systems, 68(1), Article 259. https://doi.org/10.1007/s10115-026-02862-7

Image Credits: AI Generated

DOI: 10.1007/s10115-026-02862-7

Keywords: imbalanced datasets, support vector machine, twin support vector machine, fuzzy membership, polycentric membership, class probability, least squares, machine learning, medical diagnosis, breast cancer, BreakHis dataset, classification