PRISM2 demonstrates how language-supervised pretraining provides a scalable, clinically grounded signal for generalizable pathology representations, bridging human diagnostic reasoning and foundation model performance.
Abstract
Recent rapid progress in the field of computational pathology has been enabled by foundation models. These models are beginning to move beyond encoding image patches toward whole-slide understanding, but their clinical utility remains limited. Here we present PRISM2, a multimodal slide-level foundation model trained on 2.3 million whole-slide images and 14 million question-answer pairs derived from 700,000 pathology reports. Through clinical dialogue supervision, PRISM2 aligns histomorphology with diagnostic reasoning, yielding representations that support both prompt-based inference and transferable embeddings for downstream tasks. With prompt-based inference, PRISM2 achieves or exceeds (P < 0.05) the balanced accuracy of clinical-grade products calibrated for cancer detection in the prostate, breast and breast lymph node. Additionally, across comprehensive diagnostic, biomarker and survival benchmarks, PRISM2 embeddings never statistically underperform previous foundation models via linear probing (P < 0.05). Furthermore, task-specific fine-tuning on survival prediction outperforms training from scratch on the same large survival dataset. PRISM2 demonstrates how language-supervised pretraining provides a scalable, clinically grounded signal for generalizable pathology representations, bridging human diagnostic reasoning and foundation model performance.
Multimodal large language models (MLLMs) are increasingly used to analyze pathology images. However, dominant multimodal benchmarks in pathology mainly score final diagnostic answers, captions, or reports. These evaluations provide limited insight into whether a model understands the multiscale visual content needed for pathology reasoning and decision-making. We introduce PathVU, a vision-anchored benchmark for fine-grained and multiscale visual understanding in computational pathology. Built from 23 public pathology imaging datasets with human-supervised labels and spatial annotations, PathVU evaluates MLLM understanding in two fields of view: Region FOV for high-resolution local regions and Slide FOV for macro whole-slide views. By converting raw annotations into deterministic task targets, PathVU enables programmatic scoring of region localization, visual recognition, quantity estimation, spatial reasoning, and insufficient-context judgment. The benchmark contains 14 VQA-style tasks, 61,673 images, and 308,070 samples across 28 organs and 7,253,526 annotations. Evaluating 18 representative general-purpose, medical-domain, and pathology-oriented MLLMs, we observe substantial limitations even in advanced models on fine-grained visual tasks across multiscale pathology images. PathVU provides a reproducible basis for developing and evaluating pathology MLLMs with explicit multiscale visual understanding.
Zongyi Chen, Yuping Liang, Jie Lin et al.· 2 citations
Foundation models are reshaping computational pathology, yet their capabilities remain shaped by pretraining objectives, data sources, and spatial scales, fragmenting complementary expertise across separate backbones. Here we present ALICE, a unified foundation model trained through multi-stage agglomerative distillation that sequentially distills eight vision-only, vision-language, and slide-level teacher models into dedicated modules of a single backbone. ALICE is pretrained on 24,985,184 tile-level pathology images and 155,604 high-resolution images, and evaluated across 21 task scenarios, 96 downstream tasks, and 48 data sources, spanning region-of-interest tissue analysis, vision-language multimodal evaluation, and whole-slide clinical assessment. In all three evaluation settings, ALICE achieved the best average rank among task-matched pathology foundation models. These results demonstrate that agglomerative distillation can consolidate complementary capabilities from specialized models into a unified backbone for broad computational pathology applications. The model is available at https://github.com/WonderLandxD/ALICE.
Jiawen Li, Tian Guan, Huijuan Shi et al.· 0 citations
Pathology foundation models (PFMs) provide strong tile-level representations via self-supervised pre-training on large-scale pathology images. Yet, PFMs are developed under diverse and often opaque data, architecture, and objective choices, inducing latent representational biases that limit robustness and obscure what each model specialises in. We present AdaFusion, a lightweight adaptive fusion framework that integrates complementary signals from multiple frozen PFMs through (1) low-dimensional feature compression and (2) a sample-conditioned gating module that reweights model-wise (and optionally channel-wise) contributions. Beyond improving predictive accuracy, AdaFusion provides contribution-driven interpretation that offers evidence consistent with model-specific preferences and synergistic interactions across tissue phenotypes. We evaluate AdaFusion on three public benchmarks spanning treatment response prediction, prostate cancer grading, and spatial gene expression inference. AdaFusion consistently outperforms individual PFMs and other fusion baselines, while providing interpretable tissue visualisation which aligns model preferences with morphological patterns. Code is available at: https://github.com/xyx-98/PathoOracle.
A definitive taxonomy of the medical VLM landscape is provided, tracing the evolution from early Contrastive Alignment and Generative MLLMs to the cutting-edge frontiers of Dense Pixel-Grounding, Sparse Mixture-of-Experts (MoE), and Reasoning-Incentivized (RL) architectures.
Taha Razzaq, Murtaza Taj, Asim Iqbal· Journal of Biomedical Inform...· 0 citations
ABSTRACT
The adoption of whole-slide imaging is establishing a new paradigm in digital pathology. However, the translation of artificial intelligence (AI) from research to clinical practice faces significant hurdles, largely due to a misalignment between algorithmic advances and the practical demands of pathological diagnosis and prognosis. In this review, we propose a dual-perspective framework to systematically bridge this gap by linking core clinical tasks with cutting-edge deep learning methodologies. We present a comprehensive overview of the field from 2020 to 2025, analyzing how architectures such as convolutional neural networks, vision transformers, and graph neural networks are being adapted for diagnostic classification, tissue segmentation, and prognostic prediction. A key contribution is our novel algorithm-clinical task mapping framework, which offers practical guidance for selecting and designing AI solutions tailored to specific clinical goals. We also highlight emerging trends that minimize reliance on costly annotations-including weakly supervised and self-supervised learning-as well as advances in predicting immunohistochemistry results directly from hematoxylin and eosin-stained slides. Finally, we address critical challenges related to model interpretability, regulatory approval, and multicenter generalization, and outline a future pathway focused on developing integrated, trustworthy, and equitable AI systems that enhance, rather than replace, the expertise of pathologists.
Yun-qiu Gao, Teng Ma, Lisha Li et al.· Chinese Medical Journal· 0 citations
Pathology Foundation Models (PFMs) offer powerful Whole Slide Image (WSI) representations but suffer from massive computational costs. While Knowledge Distillation (KD) can create efficient student models, existing multi-teacher methods often use suboptimal uniform weighting that ignores tissue heterogeneity. We propose LaGuadia (Language-Guided Adaptive DistillAtion), a framework that develops a compact pathology image encoder by dynamically integrating expertise from multiple PFMs under clinical linguistic guidance. Our approach utilizes a multi-stage pipeline: first, extracting visually observable clinical keywords from pathology reports; second, aligning visual features with these keywords via a Vision-Language meta-teacher (MedSigLIP) to provide dense semantic guidance; and finally, performing adaptive KD where teacher contributions are weighted based on their semantic alignment with the clinical narrative. Experiments on WSI captioning, visual question answering, and slide-level classification tasks demonstrate that an 87M parameter LaGuadia student model matches or exceeds foundation-scale models such as GigaPath and UNI, achieving strong factual consistency and robust generalization. These results highlight clinical language as an effective semantic anchor for building efficient and reliable digital pathology systems. Code is available at https://github.com/hvcl/LaGuadia.