Vision-language models (VLMs) have recently been integrated into multiple instance learning (MIL) frameworks to address the challenge of few-shot, weakly supervised classification of whole slide images (WSIs). A key trend involves leveraging multi-scale information to better represent hierarchical tissue structures. However, existing methods often face two key limitations: (1) insufficient modeling of interactions within the same modalities across scales (e.g., 5 and 20) and (2) inadequate alignment between visual and textual modalities on the same scale. To address these gaps, we propose HiVE-MIL, a hierarchical vision-language framework that constructs a unified graph consisting of (1) parent-child links between coarse (5) and fine (20) visual/textual nodes to capture hierarchical relationships, and (2) heterogeneous intra-scale edges linking visual and textual nodes on the same scale. To further enhance semantic consistency, HiVE-MIL incorporates a two-stage, text-guided dynamic filtering mechanism that removes weakly correlated patch-text pairs, and introduces a hierarchical contrastive loss to align textual semantics across scales. Extensive experiments on TCGA breast, lung, and kidney cancer datasets demonstrate that HiVE-MIL consistently outperforms both traditional MIL and recent VLM-based MIL approaches, achieving gains of up to 4.1% in macro F1 under 16-shot settings. Our results demonstrate the value of jointly modeling hierarchical structure and multimodal alignment for efficient and scalable learning from limited pathology data.

Explaining deep learning models is essential for clinical integration of medical image analysis systems. A good explanation highlights if a model depends on spurious features that undermines generalization and harms a subset of patients or, conversely, may present novel biological insights. Although techniques like GradCAM can identify influential features, they are measurement tools that do not themselves form an explanation. We propose a human-machine-VLM interaction system tailored to explaining classifiers in computational pathology, including multi-instance learning for whole-slide images. Our proof of concept comprises (1) an AI-integrated slide viewer to run sliding-window experiments to test claims of an explanation, and (2) quantification of an explanation's predictiveness using general-purpose vision-language models. The results demonstrate that this allows us to qualitatively test claims of explanations and can quantifiably distinguish competing explanations. This offers a practical path from explainable AI to explained AI in digital pathology and beyond. Code and prompts are available at https://github.com/nki-ai/x2x.

Circulating tumor cells (CTCs) are crucial biomarkers in liquid biopsy, offering a noninvasive tool for cancer patient management. However, their identification remains particularly challenging due to their limited number and heterogeneity. Labeling samples for contrast limits the generalization of fluorescence-based methods across different hospital datasets. Analyzing single-cell images enables detailed assessment of cell morphology, subcellular structures, and phenotypic variations, often hidden in clustered images. Developing a method based on bright-field single-cell analysis could overcome these limitations. CTCs can be isolated using an unbiased workflow combining Parsortix technology, which selects cells based on size and deformability, with DEPArray technology, enabling precise visualization and selection of single cells. Traditionally, DEPArray-acquired digital images are manually analyzed, making the process time-consuming and prone to variability. In this study, we present a Deep Learning-based classification pipeline designed to distinguish CTCs from leukocytes in blood samples, aimed to enhance diagnostic accuracy and optimize clinical workflows. Our approach employs images from the bright-field channel acquired through DEPArray technology leveraging a ResNet-based CNN. To improve model generalization, we applied three types of data augmentation techniques and incorporated fluorescence (DAPI) channel images into the training phase, allowing the network to learn additional CTC-specific features. Notably, only bright-field images have been used for testing, ensuring the model's ability to identify CTCs without relying on fluorescence markers. The proposed model achieved an F1-score of 0.798, demonstrating its capability to distinguish CTCs from leukocytes. These findings highlight the potential of DL in refining CTC analysis and advancing liquid biopsy applications.

The PD-L1 rate, the number of PD-L1 positive tumor cells over the total number of all tumor cells, is an important metric for immunotherapy. This metric is recorded as diagnostic information with pathological images. In this paper, we propose a proportion estimation method with a small amount of cell-level annotation and proportion annotation, which can be easily collected. Since the PD-L1 rate is calculated from only `tumor cells' and not using `non-tumor cells', we first detect tumor cells with a detection model. Then, we estimate the PD-L1 proportion by introducing a masking technique to `learning from label proportion.' In addition, we propose a weighted focal proportion loss to address data imbalance problems. Experiments using clinical data demonstrate the effectiveness of our method. Our method achieved the best performance in comparisons.

Stereotactic body radiation therapy (SBRT) refers to focusing high-energy rays in three-dimensional space on the tumor lesion area, reducing the dose received by surrounding normal tissues, which can effectively improve the local control rate of the tumor and reduce the probability of complications. With the comprehensive development of medical imaging, radiation biology and other disciplines, this less-fractional, high-dose radiotherapy method has been increasingly developed and applied in clinical practice. The background, radio-biological basis, key technologies and main equipment of SBRT are discussed, and its future development direction is prospected.

Surgical tumor removal remains one of the most common procedures during cancer treatment, with about 45% of cancer patients undergoing this surgery at some point. Thanks to recent progress in imaging and biochemical technologies, surgeons are now better able to tell tumors apart from healthy tissue. Specifically, this is enabled by a technique called "fluorescence-guided surgery" (FGS). In FGS, the patient's tissue is stained with a dye that emits infrared light when irradiated with a special light source. The dye preferentially binds to the surface of tumor cells, so that its light-wave emissions provide information on the location and extent of the tumor.

How do cancer cells differ from healthy cells? A new machine learning algorithm called "ikarus" knows the answer, reports a team led by MDC bioinformatician Altuna Akalin in the journal Genome Biology. The AI program has found a gene signature characteristic of tumors. When it comes to identifying patterns in mountains of data, human beings are no match for artificial intelligence (AI). In particular, a branch of AI called machine learning is often used to find regularities in data sets--be it for stock market analysis, image and speech recognition, or the classification of cells.

When it comes to identifying patterns in mountains of data, human beings are no match for artificial intelligence (AI). In particular, a branch of AI called machine learning is often used to find regularities in data sets – be it for stock market analysis, image and speech recognition, or the classification of cells. To reliably distinguish cancer cells from healthy cells, a team led by Dr. Altuna Akalin, head of the Bioinformatics and Omics Data Science Platform at the Max Delbrück Center for Molecular Medicine in the Helmholtz Association (MDC), has now developed a machine learning program called "ikarus." The program found a pattern in tumor cells that is common to different types of cancer, consisting of a characteristic combination of genes. According to the team's paper in the journal Genome Biology, the algorithm also detected types of genes in the pattern that had never been clearly linked to cancer before.

How do cancer cells differ from healthy cells? A new machine learning algorithm called "ikarus" knows the answer, reports a team led by MDC bioinformatician Altuna Akalin in the journal Genome Biology. The AI program has found a gene signature characteristic of tumors. When it comes to identifying patterns in mountains of data, human beings are no match for artificial intelligence (AI). In particular, a branch of AI called machine learning is often used to find regularities in data sets – be it for stock market analysis, image and speech recognition, or the classification of cells.

How do cancer cells differ from healthy cells? A new machine learning algorithm called "ikarus" knows the answer, reports a team led by MDC bioinformatician Altuna Akalin in the journal Genome Biology. The AI program has found a gene signature characteristic of tumors. When it comes to identifying patterns in mountains of data, human beings are no match for artificial intelligence (AI). In particular, a branch of AI called machine learning is often used to find regularities in data sets – be it for stock market analysis, image and speech recognition, or the classification of cells.