Modern Pathology最新文献

{"title":"Spatial Mapping of Gene Signatures in Hematoxylin and Eosin-Stained Images: A Proof of Concept for Interpretable Predictions Using Additive Multiple Instance Learning","authors":"Miles Markey,&nbsp;Juhyun Kim,&nbsp;Zvi Goldstein,&nbsp;Ylaine Gerardin,&nbsp;Jacqueline Brosnan-Cashman,&nbsp;Syed Ashar Javed,&nbsp;Dinkar Juyal,&nbsp;Harshith Pagidela,&nbsp;Limin Yu,&nbsp;Bahar Rahsepar,&nbsp;John Abel,&nbsp;Stephanie Hennek,&nbsp;Archit Khosla,&nbsp;Amaro Taylor-Weiner,&nbsp;Chintan Parmar","doi":"10.1016/j.modpat.2025.100772","DOIUrl":"10.1016/j.modpat.2025.100772","url":null,"abstract":"<div><div>The relative abundance of cancer-associated fibroblast (CAF) subtypes influences a tumor’s response to treatment, especially immunotherapy. However, the gene expression signatures associated with these CAF subtypes have yet to realize their potential as clinical biomarkers. Here, we describe an interpretable machine learning approach, additive multiple instance learning (aMIL), to predict bulk gene expression signatures from hematoxylin and eosin-stained whole-slide images, focusing on an immunosuppressive LRRC15+ CAF-enriched TGFβ-CAF signature. aMIL models accurately predicted TGFβ-CAF across various cancer types. Tissue regions contributing most highly to slide-level predictions of TGFβ-CAF were evaluated by machine learning models characterizing spatial distributions of diverse cell and tissue types, stromal subtypes, and nuclear morphology. In breast cancer, regions contributing most to TGFβ-CAF-high predictions (“excitatory”) were localized to cancer stroma with high fibroblast density and mature collagen fibers. Regions contributing most to TGFβ-CAF-low predictions (“inhibitory”) were localized to cancer epithelium and densely inflamed stroma. Fibroblast and lymphocyte nuclear morphology also differed between excitatory and inhibitory regions. Thus, aMIL enables a data-driven link between histologic features and transcription, offering biological interpretability beyond typical black-box models.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 8","pages":"Article 100772"},"PeriodicalIF":7.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143941833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence–Based Single-Cell Analysis as a Next-Generation Histologic Grading Approach in Colorectal Cancer: Prognostic Role and Tumor Biology Assessment","authors":"Vincenzo Mitchell Barroso ,&nbsp;Zhilong Weng ,&nbsp;Lennert Glamann ,&nbsp;Marcus Bauer ,&nbsp;Claudia Wickenhauser ,&nbsp;Thomas Zander ,&nbsp;Reinhard Büttner ,&nbsp;Alexander Quaas ,&nbsp;Yuri Tolkach","doi":"10.1016/j.modpat.2025.100771","DOIUrl":"10.1016/j.modpat.2025.100771","url":null,"abstract":"<div><div>The management of colorectal carcinoma (CRC) relies on pathological interpretation. Digital pathology approaches allow for development of new potent artificial intelligence–based prognostic parameters. The study aimed to develop an artificial intelligence–based image analysis platform allowing fully automatized, quantitative, and explainable tumor microenvironment analysis and extraction of prognostic information from hematoxylin and eosin–stained whole-slide images of CRC patients. Three well--characterized, multi-institutional patient cohorts were included (patient n = 1438, whole-slide image n &gt; 2400). The developed image analysis platform implements quality control and established algorithms to segment tissue and detect cell types. It enabled systematic analysis of immune infiltrate, assessing its prognostic relevance, intratumoral heterogeneity, and biological concepts across multiple survival end points. Analyzing single-cell types and their combinations reveals independent, prognostic parameters, highlighting significant intratumoral heterogeneity, especially in the biopsy setting, which must be accounted for. A key morphologic concept related to tumor control by the immune system is described, resulting in a capable, independent prognostic parameter (tumor “out of control”). Our findings have direct clinical implications and can be used as a foundation for updating the existing CRC grading systems.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 7","pages":"Article 100771"},"PeriodicalIF":7.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143904171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Undifferentiated Pleomorphic Sarcoma in Children and Young Adults: A Comprehensive Clinicopathologic, Genomic, and Epigenetic Comparison With Adult Counterparts","authors":"Carla Saoud ,&nbsp;Gunes Gundem ,&nbsp;Chad M. Vanderbilt ,&nbsp;Leonard H. Wexler ,&nbsp;Damon R. Reed ,&nbsp;William Tap ,&nbsp;Samuel Singer ,&nbsp;Liliana B. Villafania ,&nbsp;Elli Papaemmanouil ,&nbsp;Jamal Benhamida ,&nbsp;Tejus A. Bale ,&nbsp;Cristina R. Antonescu","doi":"10.1016/j.modpat.2025.100769","DOIUrl":"10.1016/j.modpat.2025.100769","url":null,"abstract":"<div><div>Undifferentiated pleomorphic sarcoma (UPS) occurs primarily in older adults and remains a diagnosis of exclusion due to its lack of differentiation and specific molecular alterations. Its occurrence in children is rare and controversial, with an unclear relationship to its adult counterpart. In this study, we aimed to investigate a cohort of 6 pediatric undifferentiated pleomorphic sarcoma (P-UPS, mean 10 years old) and 19 young-adult undifferentiated pleomorphic sarcoma (YA-UPS, mean 30 years old) cases by conducting a comprehensive comparative analysis of their clinicopathologic, genomic, and epigenetic features relative to their adult undifferentiated pleomorphic sarcoma counterparts (A-UPS, n = 100). Histologically, P-UPS and YA-UPS exhibited broad morphologic spectrum. The most frequent alterations across all groups were <em>TP53</em>, <em>CDKN2A/B</em>, and <em>ATRX,</em> with no significant differences among subsets. Notably, <em>RB1</em> alterations were absent in P-UPS, although representing the second most common alteration in YA-UPS (32%) and A-UPS (41%). <em>PTEN</em> alterations were significantly more prevalent in YA-UPS (26%) compared with that in P-UPS (0%) and A-UPS (6%). Deletions in chromosomes 10, 16q, and 13q, along with amplification of 20q, were the most common across all groups. Except for a higher frequency of 17q amplification in P-UPS (33%) and YA-UPS (26%) compared with that in A-UPS (6%), no other arm-level differences were observed. P-UPS showed a lower mean fraction genome altered compared with YA-UPS and A-UPS, whereas all UPS age groups showed a low tumor mutational burden (mean &lt;10 mut/MB). Pathogenic germline variants of high clinical significance (<em>TP53</em>, <em>NF1</em>, <em>MLH1</em>, <em>CHEK2</em>, and <em>BARD1</em>) were observed only in YA-UPS (31%) and A-UPS (12%) cases. By T-distributed stochastic neighborhood embedding and hierarchical clustering of DNA methylation, the majority of P-UPS and a small subset of YA-UPS grouped in a distinct cluster, characterized by a lower genomic index compared to A-UPS. In contrast, most UPS occurring in young adults genomically parallel their older adults’ counterparts. P-UPS and YA-UPS cases exhibited a better disease-specific and progression-free survival, compared with A-UPS cases.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 8","pages":"Article 100769"},"PeriodicalIF":7.1,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143917344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pathology Asks for Global Regulations in Artificial Intelligence Employment","authors":"Massimo Rugge ,&nbsp;Matteo Fraschini ,&nbsp;Alessandro D’Amuri ,&nbsp;Gavino Faa","doi":"10.1016/j.modpat.2025.100754","DOIUrl":"10.1016/j.modpat.2025.100754","url":null,"abstract":"","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 5","pages":"Article 100754"},"PeriodicalIF":7.1,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143807594","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatially Resolved Single-Cell Morphometry of Benign Breast Disease Biopsy Images Uncovers Quantitative Cytomorphometric Features Predictive of Subsequent Invasive Breast Cancer Risk","authors":"Mustapha Abubakar ,&nbsp;Shaoqi Fan ,&nbsp;Alyssa Klein ,&nbsp;Ruth M. Pfeiffer ,&nbsp;Scott Lawrence ,&nbsp;Karun Mutreja ,&nbsp;Teresa M. Kimes ,&nbsp;Kathryn Richert-Boe ,&nbsp;Jonine D. Figueroa ,&nbsp;Gretchen L. Gierach ,&nbsp;Maire A. Duggan ,&nbsp;Thomas E. Rohan","doi":"10.1016/j.modpat.2025.100767","DOIUrl":"10.1016/j.modpat.2025.100767","url":null,"abstract":"<div><div>Currently, benign breast disease (BBD) pathologic classification and invasive breast cancer (BC) risk assessment are based on qualitative epithelial changes, with limited utility for BC risk stratification for women with lower-risk category BBD (ie, nonproliferative disease [NPD] and proliferative disease without atypia [PDWA]). Here, machine learning–based single-cell morphometry was used to characterize quantitative changes in epithelial nuclear morphology that reflect functional/structural decline (ie, increasing nuclear size, assessed as epithelial nuclear area and nuclear perimeter), altered DNA chromatin content (ie, increasing nuclear chromasia), and increased cellular crowding/proliferation (ie, increasing nuclear contour irregularity). Cytomorphologic changes reflecting chronic stromal inflammation were assessed using stromal cellular density. Data and pathology materials were obtained from a case-control study (n = 972) nested within a cohort of 15,395 women diagnosed with BBD at Kaiser Permanente Northwest (1971-2012). Odds ratios (ORs) and 95% confidence intervals (CIs) for associations of cytomorphometric features with risk of subsequent BC were assessed using multivariable logistic regression. More than 55 million epithelial and 37 million stromal cells were profiled across 972 BBD images. Cytomorphometric features were individually predictive of subsequent BC risk, independently of BBD histologic classification. However, cytomorphometric features of epithelial functional/structural decline were statistically significantly predictive of low-grade but not high-grade BC following PDWA (OR for low-grade BC per 1-SD increase in nuclear area and nuclear perimeter, 2.10; 95% CI, 1.26-3.49, and 2.22; 95% CI, 1.30-3.78, respectively), whereas stromal inflammation was predictive of high-grade but not low-grade BC following NPD (OR for high-grade BC per 1-SD increase in stromal cellular density, 1.53; 95% CI, 1.13-2.08). Associations of nuclear chromasia and nuclear contour irregularity with subsequent tumor grade were context specific, with both features predicting low-grade BC risk following PDWA (OR per 1-SD, 1.58; 95% CI, 1.06-2.35, and 2.21; 95% CI, 1.25-3.91, for nuclear chromasia and nuclear contour irregularity, respectively) and high-grade BC following NPD (OR per 1-SD, 1.47; 95% CI, 1.11-1.96, and 1.29; 95% CI, 1.00-1.70, for nuclear chromasia and nuclear contour irregularity, respectively). The results indicate that cytomorphometric features on BBD hematoxylin-eosin–stained images might help to refine BC risk estimation and potentially inform BC risk reduction strategies for BBD patients, particularly those currently designated as low risk.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 7","pages":"Article 100767"},"PeriodicalIF":7.1,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143882593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Polychromatic Polarization Microscopy Differentiates Collagen Fiber Signatures in Benign Pancreatic Tissue and Pancreatic Ductal Adenocarcinoma","authors":"Mahsa Chitsaz ,&nbsp;Linlin Yang ,&nbsp;Rania Rayes-Danan ,&nbsp;Omid Savari ,&nbsp;Bin Li ,&nbsp;Michael Shribak ,&nbsp;Kevin Eliceiri ,&nbsp;Agnes Loeffler","doi":"10.1016/j.modpat.2025.100768","DOIUrl":"10.1016/j.modpat.2025.100768","url":null,"abstract":"<div><div>The orientation of collagen fibers in relation to malignant epithelium is known to carry prognostic information in a variety of tissues. The data are the strongest for breast and pancreatic ductal adenocarcinoma. However, information inherent in collagen fiber topology in malignant tissues remains untapped in daily surgical pathology practice, largely because collagen fibers within areas of desmoplasia cannot be resolved with standard diagnostic microscopy. The methodologies used to visualize collagen fiber orientation are either of insufficient resolution to consistently capture collagen fiber topology or require resources in time and money that do not fit into the daily surgical pathology workflow. Polychromatic polarization microscopy has the potential to bring collagen topology to the attention of pathologists during their routine work. It has been demonstrated to be equivalent to the gold standard methodology used to research collagen, second harmonic generation. We use polychromatic polarization microscopy to visualize and describe the differences in collagen topology in normal pancreas, chronic pancreatitis, and pancreatic ductal adenocarcinoma with a standard microscope, using hematoxylin and eosin-stained sections. In the process, we propose a lexicon with which to describe the morphologic characteristics of collagen in benign and malignant pancreatic tissues.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 8","pages":"Article 100768"},"PeriodicalIF":7.1,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143917342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward Optimizing the Impact of Digital Pathology and Augmented Intelligence on Issues of Diagnosis, Grading, Staging, and Classification","authors":"Lewis A. Hassell ,&nbsp;Marika L. Forsythe ,&nbsp;Ami Bhalodia ,&nbsp;Thanh Lan ,&nbsp;Tasnuva Rashid ,&nbsp;Astin Powers,&nbsp;Marilyn M. Bui ,&nbsp;Arlen Brickman ,&nbsp;Qiangqiang Gu ,&nbsp;Andrey Bychkov ,&nbsp;Ian Cree ,&nbsp;Liron Pantanowitz","doi":"10.1016/j.modpat.2025.100765","DOIUrl":"10.1016/j.modpat.2025.100765","url":null,"abstract":"<div><div>The introduction of new diagnostic information in pathology requires effective dissemination and adoption strategies. Although traditional methods like journals, meetings, and atlases have been used, they pose challenges in accessibility, interactivity, and performance validation. Digital pathology (DP) and artificial or augmented intelligence (AI) offer promising solutions to address these limitations. This paper advocates the use of DP and AI tools to facilitate the introduction of new diagnostic information in pathology. It highlights the importance of standardized training and validation sets, digital slide libraries, and AI-enhanced diagnostic tools. Although AI can improve efficiency and accuracy, it is crucial to address potential pitfalls such as over-reliance on AI, bias, and the need for human oversight. By leveraging DP and AI, the efficiency and accuracy of diagnosis, grading, staging, and classification can be augmented, ultimately improving patient care.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 7","pages":"Article 100765"},"PeriodicalIF":7.1,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144023790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cytoplasmic p53 Immunostaining in Salivary Duct Carcinoma: A Poor Prognostic Factor Associated With Characteristic TP53 Variants","authors":"Yoshitaka Utsumi ,&nbsp;Masato Nakaguro ,&nbsp;Daisuke Kawakita ,&nbsp;Hideaki Hirai ,&nbsp;Aoi Sukeda ,&nbsp;Shinji Kohsaka ,&nbsp;Kiyoaki Tsukahara ,&nbsp;Toyoyuki Hanazawa ,&nbsp;Satoshi Kano ,&nbsp;Keisuke Yamazaki ,&nbsp;Yushi Ueki ,&nbsp;Kenji Okami ,&nbsp;Yuki Saito ,&nbsp;Hiroyuki Ozawa ,&nbsp;Yoshitaka Honma ,&nbsp;Akira Shimizu ,&nbsp;Kenji Hanyu ,&nbsp;Shota Fujii ,&nbsp;Tomoyuki Arai ,&nbsp;Sho Iwaki ,&nbsp;Toshitaka Nagao","doi":"10.1016/j.modpat.2025.100766","DOIUrl":"10.1016/j.modpat.2025.100766","url":null,"abstract":"<div><div>Salivary duct carcinoma (SDC) is an uncommon, high-grade malignancy. Identifying suitable prognostic factors is crucial for developing effective treatment strategies for SDC. p53 Immunohistochemistry (IHC) is a potential prognostic marker for SDC. Traditionally, only the nuclear expression has been considered when evaluating aberrant p53 IHC patterns. However, recent studies on other organ cancers have highlighted the significance of the cytoplasmic p53 expression. We aimed to investigate the prognostic implications of cytoplasmic p53 positivity and its association with <em>TP53</em> variants in a large cohort of patients with SDC. p53 IHC was performed in 327 patients with SDC who had undergone primary curative resection. Based on the immunostaining patterns, patients were classified into 4 groups: wild-type (WT), overexpression (OE), complete absence (CA), and cytoplasmic (CY). Additionally, the <em>TP53</em> gene status was analyzed in 239 cases by Sanger and/or next-generation sequencing. The p53 IHC patterns of 327 cases were as follows: WT (n = 125; 38.2%), OE (n = 100; 30.6%), CA (n = 75; 22.9%), and CY (n = 27; 8.3%). A <em>TP53</em> genetic analysis of 239 cases revealed the following: WT status (n = 80; 33.5%), missense/inframe variants (n = 86; 36.0%), and truncating variants (n = 73; 30.5%). Notably, 24 of the 25 CY cases (96%) harbored <em>TP53</em> variants, which were predominantly located in the domains responsible for nuclear translocation. Of these, 22 exhibited truncating variants. In a multivariate analysis, CY cases demonstrated significantly shorter disease-free survival (DFS) than WT cases (<em>P</em> = .01). Furthermore, patients with aberrant p53 expression patterns (OE+CA+CY) had significantly worse DFS and overall survival than those with WT (<em>P</em> = .003 and .002, respectively). The presence of <em>TP53</em> variants was also associated with poorer DFS and overall survival (<em>P</em> = .003 and .02, respectively). Our findings suggest that the cytoplasmic expression of p53 in SDC represents a distinct aberrant pattern underlying characteristic genetic abnormalities and has significant prognostic implications.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 7","pages":"Article 100766"},"PeriodicalIF":7.1,"publicationDate":"2025-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143882592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Labor-Efficient Pathological Auxiliary Diagnostic Model for Primary and Metastatic Tumor Tissue Detection in Pancreatic Ductal Adenocarcinoma","authors":"Xinyi Ke ,&nbsp;Moxuan Yang ,&nbsp;Jingci Chen ,&nbsp;Ruping Hong ,&nbsp;Zheng Wang ,&nbsp;Shuhao Wang ,&nbsp;Hui Zhang ,&nbsp;Junliang Lu ,&nbsp;Boju Pan ,&nbsp;Yike Gao ,&nbsp;Xiaoding Liu ,&nbsp;Xiaoyu Li ,&nbsp;Yang Zhang ,&nbsp;Si Su ,&nbsp;Huanwen Wu ,&nbsp;Zhiyong Liang","doi":"10.1016/j.modpat.2025.100764","DOIUrl":"10.1016/j.modpat.2025.100764","url":null,"abstract":"<div><div>Accurate histopathological evaluation of pancreatic ductal adenocarcinoma (PDAC), including primary tumor lesions and lymph node metastases, is critical for prognostic evaluation and personalized therapeutic strategies. Distinct from other solid tumors, PDAC presents unique diagnostic challenges owing to its extensive desmoplasia, unclear tumor boundary, and difficulty in differentiating from chronic pancreatitis. These characteristics not only complicate pathological diagnosis but also hinder the acquisition of pixel-level annotations required for training computational pathology models. In this study, we present PANseg, a multiscale weakly supervised deep learning framework for PDAC segmentation, trained and tested on 368 whole-slide images (WSIs) from 208 patients across 2 independent centers. Using only image-level labels (2048 × 2048 pixels), PANseg achieved comparable performance with fully supervised baseline (FSB) across the internal test set 1 (17 patients/58 WSIs; PANseg area under the receiver operating characteristic curve [AUROC]: 0.969 vs FSB AUROC: 0.968), internal test set 2 (40 patients/44 WSIs; PANseg AUROC: 0.991 vs FSB AUROC: 0.980), and external test set (20 patients/20 WSIs; PANseg AUROC: 0.950 vs FSB AUROC: 0.958). Moreover, the model demonstrated considerable generalizability with previously unseen sample types, attaining AUROCs of 0.878 on fresh-frozen specimens (20 patients/20 WSIs) and 0.821 on biopsy sections (20 patients/20 WSIs). In lymph node metastasis detection, PANseg augmented the diagnostic accuracy of 6 pathologists from 0.888 to 0.961, while reducing the average diagnostic time by 32.6% (72.0 vs 48.5 minutes). This study demonstrates that our weakly supervised model can achieve expert-level segmentation performance and substantially reduce annotation burden. The clinical implementation of PANseg holds great potential in enhancing diagnostic precision and workflow efficiency in the routine histopathological assessment of PDAC.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 7","pages":"Article 100764"},"PeriodicalIF":7.1,"publicationDate":"2025-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143811767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cover","authors":"","doi":"10.1016/S0893-3952(25)00056-0","DOIUrl":"10.1016/S0893-3952(25)00056-0","url":null,"abstract":"","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 4","pages":"Article 100760"},"PeriodicalIF":7.1,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143834470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}