The clinical motivation behind the work is straightforward and consequential. Colorectal cancer remains one of the most common and deadly malignancies worldwide, and most of these cancers arise from polyps, small growths protruding from the lining of the colon that can be removed during routine colonoscopy before they turn malignant. Detecting and delineating those polyps accurately is therefore a direct determinant of whether a lesion is excised in time. Dermoscopy, meanwhile, is the dermatologist’s principal tool for the early detection of melanoma and other skin cancers, and the precise boundary of a lesion carries decisive diagnostic weight, informing decisions about biopsy, excision margins and follow-up. In both settings, clinicians today rely on manual or semi-manual outlining of lesions in images, a process that is slow, subjective and prone to inter-observer variability. An algorithm that could reliably draw those contours automatically would not only save expert time but could also flag subtle lesions that tired human eyes might miss.

Getting a computer to draw those contours, however, has proven remarkably stubborn. Polyps in endoscopic images vary enormously in morphology, size and appearance; some are flat and barely distinguishable in color from the surrounding mucosa, others sit at the edge of the frame, partially obscured by specular highlights, bubbles, or motion blur. Skin lesions present a parallel problem, with irregular, sometimes fading boundaries that blend into healthy skin. Early deep learning solutions, most famously the U-Net family of convolutional networks, achieved strong results by repeatedly downsampling and upsampling images and stitching fine local texture onto coarse semantic context. But convolutional kernels see only a small window at a time, so these networks struggle to connect distant parts of an image into a coherent global understanding, and they often produce ragged or leaking boundaries around small or low-contrast objects. Transformers, which use self-attention to let every pixel consult every other pixel, solved the long-range dependency problem but introduced a new one: the computational cost of global attention grows quadratically with the number of image tokens, making naive transformer segmentation expensive and sometimes imprecise at fine scales.

NAHFormer’s answer begins at the front of the network. The framework employs a pyramid-structured Mix Transformer, or MiT, encoder, a hierarchical backbone in which the image is progressively broken into larger patches through successive stages, producing a ladder of feature maps that descend in spatial resolution while ascending in semantic richness. Early stages capture delicate texture and fine edges, essential for tracing the exact rim of a polyp or lesion, while later stages encode the broader context needed to say, with confidence, that a given blob is a lesion at all. This multi-scale representation, the authors argue, is what underpins the model’s generalization capability, allowing it to cope with the extraordinary diversity of lesion appearances it encounters across different patients, cameras, imaging conditions and anatomical sites rather than overfitting to the quirks of any single dataset.

The first bespoke innovation sits above that encoder: a Cross-Resolution Semantic Perception, or CRSP, module. Its job is to make different layers of the pyramid talk to each other in a structured way, integrating semantic information across multiple resolutions so that the coarse layers’ understanding of what a lesion is can guide the fine layers’ placement of its boundary. The module performs this cross-resolution conversation through neighborhood attention, an attention mechanism in which each token attends not to the entire image but only to a small window of its nearest neighbors. That locality restriction, borrowed from the neighborhood attention transformer introduced by computer vision researchers in 2023, dramatically reduces computation compared with global self-attention while retaining most of the discriminative power, because in dense prediction tasks the pixels most relevant to resolving a given boundary are typically its immediate surroundings. The practical payoff, according to the team, is precise delineation of lesion contours and boundaries, precisely the capability where earlier transformer models have tended to stumble.

The second innovation, a Hierarchical Feature Fusion, or HFF, module, tackles a quieter but equally corrosive problem: redundancy. In multi-scale architectures, as features from different levels are combined, the same information is often carried forward repeatedly, and low-level detail can be swamped or diluted by what are effectively duplicate signals from coarser scales. The HFF module progressively aggregates the multi-scale feature hierarchy in a stepwise fashion, deliberately suppressing redundant information as it fuses, so that the final segmentation head receives a cleaner, more informative summary of the image. This redundancy-aware aggregation, combined with the boundary-sensitive attention of the CRSP module, is what the authors describe as the framework’s distinctive pairing of redundancy-aware and boundary-aware capabilities, a combination intended to raise accuracy without inflating the network into an unwieldy behemoth.

The experimental evidence spans the field’s most widely used public benchmarks. On the polyp side, the team evaluated NAHFormer on five colonoscopy datasets: Kvasir, CVC-ClinicDB, CVC-ColonDB, EndoScene and ETIS. These sets differ substantially in acquisition conditions and difficulty; CVC-ClinicDB contains relatively clean and well-lit frames, while ETIS is widely regarded as the sternest test, composed of small, poorly contrasted polyps that routinely drag down the scores of models that excel elsewhere. On the dermatology side, the framework was tested on the ISIC 2017 and ISIC 2018 skin lesion archives, curated through the International Skin Imaging Collaboration’s annual challenges for melanoma detection. The pattern of results is telling: NAHFormer did not merely top the leaderboard on the easy datasets but held its advantage on the hard ones, achieving its reported mean Dice score of 0.821 on ETIS and 0.909 on ISIC 2018, indicating that its architectural choices translate into genuine robustness rather than dataset-specific tuning.

The Dice coefficient, the metric at the center of these comparisons, measures the overlap between the algorithm’s predicted segmentation and the ground truth annotated by experts, ranging from zero for no overlap to one for perfect agreement, so the difference between a middling model and a strong one often comes down to whether the predicted boundary clips a sliver off the lesion or lets background tissue bleed in. Success on ETIS, where polyps are small and faintly demarcated, suggests that the neighborhood attention scheme is doing exactly what it was designed to do, pinning down delicate boundaries without the blurring that coarse feature fusion can introduce. The strong ISIC result reinforces the point from the opposite direction, since skin lesion boundaries are often biologically diffuse, fading gradually into surrounding skin, and demand a different kind of perceptual finesse. A single architecture scoring highly across both regimes hints at a level of adaptability that previous task-specific models, whether convolutional designs like PraNet or hybrid transformer convolutional systems such as TransFuse, Swin-UNet, MissFormer and H2Former, have generally had to trade off against one another.

The broader significance of the work lies in what it suggests about the trajectory of clinical AI. Rather than choosing between the precision of convolutional networks and the global reasoning of transformers, the Chongqing team’s design stitches the two virtues together, using a hierarchical transformer backbone, locality-restricted attention and progressive fusion to keep both computation and accuracy in a clinically usable range. All of the datasets used in the study are publicly available, which means other groups can immediately stress-test the claims and build on the architecture, and the work was supported in part by the National Natural Science Foundation of China under Grant 61701060 and by a Chongqing Graduate Scientific Research Innovation Project under Grant CYS25470. The researchers frame the system as a step toward segmentation tools that could eventually assist physicians in real time, flagging and outlining lesions as images are captured. For now, the result stands as a demonstration that rethinking where an attention mechanism should look, and what it should ignore, can push the frontier of medical image analysis forward on the benchmarks that matter.

Subject of Research: A transformer-based deep learning framework for segmenting colonoscopic polyps and dermoscopic skin lesions in medical images.

Article Title: NAHFormer: Neighborhood self-attention and hierarchical feature fusion transformer for image medical segmentation

Article References: Yin, X., Li, H., Hou, T., & Tang, C. (2026). NAHFormer: Neighborhood self-attention and hierarchical feature fusion transformer for image medical segmentation. Applied Intelligence, 56(15), Article 430. https://doi.org/10.1007/s10489-026-07474-w

Image Credits: AI Generated

DOI: 10.1007/s10489-026-07474-w

Keywords: medical image segmentation, transformer, neighborhood attention, polyp segmentation, skin lesion segmentation, colonoscopy, dermoscopy, deep learning, computer vision, multi-scale feature fusion, Dice score, Applied Intelligence

Melanoma, one of the deadliest forms of skin cancer, often remains undetected until it reaches advanced stages where treatment becomes significantly more challenging. Early identification is foundational to improving patient prognosis and survival rates. As the prevalence of skin cancers rises globally, the necessity for efficient diagnostic solutions has never been more urgent. Traditional diagnostic methods heavily rely on the expertise of dermatologists, which can sometimes yield inconsistent results due to subjective interpretations. Thus, the integration of artificial intelligence into this field marks a transformative evolution.

The study leverages dermoscopy images, which are critical in the evaluation of skin lesions. These images provide intricate insights into skin features that are crucial for distinguishing between benign and malignant growths. However, manually analyzing dermoscopy images can be tedious and prone to error, underscoring the need for automated systems that can deliver accurate assessments.

By implementing a hybrid model that amalgamates the strengths of both CNNs and ANNs, the research team addressed the limitations often encountered in stand-alone systems. CNNs excel at extracting high-level features from images, leveraging deep learning architectures to recognize patterns that are not readily visible to the human eye. In contrast, ANNs contribute robust decision-making capabilities that utilize these features to enhance classification performance. The synergistic effect of combining these methodologies results in a powerful tool capable of discerning melanoma with improved precision.

This multifaceted approach begins at the preprocessing stage, where dermoscopy images are meticulously adjusted to ensure uniformity, thus optimizing the input for machine learning algorithms. Subsequent layers of CNN are designed to capture rich and complex features of skin lesions, progressively refining the image data to extract essential characteristics. The outputs from these convolutional layers are then funneled into the ANN, where sophisticated algorithms analyze the extracted features, culminating in a decisive classification of the images as benign or malignant.

In their experiments, the researchers utilized a comprehensive dataset comprising diverse dermoscopy images, ranging from common benign moles to various stages of melanoma. This diversity is crucial as it ensures that the model generalizes well across different skin types and conditions, a common challenge in dermatological diagnostics. The evaluation metrics used in the study reaffirmed the model’s effectiveness, showcasing notable improvements in accuracy, sensitivity, and specificity metrics over existing models.

Moreover, the study underscores the importance of explainability in AI-driven medical solutions. As healthcare professionals increasingly adopt AI tools, it becomes essential that these systems not only produce accurate results but also provide clear reasoning for their classifications. The architecture of the model designed in this study was enhanced to provide visual feedback on the decision-making process, allowing dermatologists to interpret AI findings more effectively and integrate them seamlessly into their clinical practices.

This research adds a significant layer of utility by presenting a robust framework that could potentially be integrated into current clinical systems, paving the way for real-time melanoma detection solutions in dermatology offices and hospitals across the globe. As AI technology evolves, its contributions to healthcare are destined to grow, transforming how medical professionals approach diagnostics and patient care.

The researchers have called for collaboration between technologists and healthcare practitioners to consistently refine these models further, making them even more tailored to specific populations. Cultural and geographical differences can influence the presentation of skin lesions, and thus the training datasets should reflect this diversity for broader applicability.

Additionally, the study opens doors for future explorations into integrating other forms of imaging technologies or data points, such as genetic markers, which could further enhance predictive capabilities. The potential for these AI-driven models to incorporate vast amounts of patient data creates a fertile ground for pioneering research that promises to redefine cancer care methodologies.

As this innovative modality permeates the medical landscape, it also brings important discussions about ethical considerations surrounding the deployment of AI in healthcare. Issues such as data privacy, algorithmic bias, and the need for regulatory frameworks are essential conversations as the technology matures. Ensuring that these systems function equitably and responsibly within society is paramount as we navigate the future of AI and medicine.

The team of Alshmrani, Alotaibi, and Alfakeeh is poised at the forefront of this transformative field, championing a model that not only enhances clinical accuracy but also bridges the gap between AI capabilities and practical applications in medicine. Their contributions underscore an exciting future in which technology and healthcare converge to enhance patient outcomes with unprecedented precision and reliability.

In conclusion, the fusion of multi CNN features with ANN represents an important advancement in the early classification of melanoma using dermoscopy images. By integrating cutting-edge machine learning techniques with rigorous medical analysis, this study not only showcases the potential of artificial intelligence but also highlights a pathway for improved diagnostic practices in dermatology, ultimately aiming to enhance patient care and outcomes in oncology.

Subject of Research: Early classification of melanoma using dermoscopy images through a hybrid model of CNN and ANN

Article Title: Fusion of multi CNN features with ANN for early classification of melanoma using dermoscopy images

Article References:

Alshmrani, A.S., Alotaibi, F.M. & Alfakeeh, A.S. Fusion of multi CNN features with ANN for early classification of melanoma using dermoscopy images.
Discov Sustain (2026). https://doi.org/10.1007/s43621-025-02556-0

Image Credits: AI Generated

DOI: 10.1007/s43621-025-02556-0

Keywords: melanoma, early classification, dermoscopy images, convolutional neural networks, artificial neural networks, machine learning, healthcare innovation, medical imaging, AI in dermatology, skin cancer detection.