Histopathologists Research Articles

The role of progesterone and estrogen receptors in treatment choice after endometriosis surgery: A cross-sectional study.

The lack of improvement in some endometriotic people's pain after surgery even while using hormone treatment may suggest an inappropriate response to routine hormonal therapies. This study aimed to determine a cut-off point for selecting the most appropriate treatment based on the hormone receptors of endometriotic lesions. In this cross-sectional study, by reviewing the medical records of participants and testing their archive samples and phone interviews (if needed), 86 symptomatic women after endometriosis surgery who were operated into governmental hospitals, Shahid Faghihi and Hazrate Zeinab Shiraz Iran were enrolled between March 2017 and March 2019. Women were divided into 2 groups: responsiveness (n = 73 for dysmenorrhea, n = 60 for dyspareunia) to medical treatment and surgery, and unresponsiveness (n = 13, n = 7). We examined the pathological slides of 86 women to determine the amount of hormone receptors and the relationship between the type of medical treatment and the level of hormone receptors on pain relief within 1 yr after surgery. Based on the receiver operating characteristic curve, dysmenorrhea in the presence of tissue estrogen receptors 60% (p = 0.1065), and dyspareunia in the presence of tissue progesterone receptors 80% (p = 0.0001) responded well to medical treatment after surgery. In the presence of endometrioma-dysmenorrhea showed the best response to oral contraceptive pills (69.4%), while in deep infiltrative endometriosis-dyspareunia showed the best response to progesterone treatment (75%). Prescribing an appropriate hormone therapy based on a specific immunohistochemistry staining pattern can improve the life quality of postoperative endometriosis individuals.

Weakly supervised deep learning image analysis can differentiate melanoma from naevi on haematoxylin and eosin-stained histopathology slides.

The broad histomorphological spectrum of melanocytic pathologies requires large data sets to develop accurate and generalisable deep learning (DL)-based diagnostic pathology classifiers. Weakly supervised DL promotes utilisation of larger training data sets compared to fully supervised (patch annotation) approaches. To evaluate weakly supervised DL image classifiers for discriminating melanomas from naevi on haematoxylin and eosin (H&E)-stained pathology slides. A representative H&E slide for 260 naevi and 260 melanomas from mucocutaneous sites at one tertiary institution was digitized. Clinicopathological features were recorded for each case including thickness and histological subtype. Whole-slide or whole-tissue section labels were applied. The ground truth was established by consensus diagnosis from two pathologists. Multiple-instance learning models, Trans-MIL, CLAM and DTFD-MIL were evaluated at 10×, 20× and 40× magnifications using stratified fivefold Monte Carlo cross-validation, with 80/10/10 splits for training/validation/test groups, to predict melanoma from naevus. Heatmaps were generated to understand model performance. Naevi cases were younger (median age: 51 years; melanoma median age: 71.5 years), with more balanced sex distribution (males: 48.8%, melanoma male subgroup: 64.2%). The most frequent histological subtypes of naevi and melanomas were dysplastic compound (n = 99, 38.1%) and superficial spreading (n = 124, 47.7%), respectively. Average AUC (±1 SD) for Trans-MIL, CLAM and DTFD-MIL across test groups were 0.9952 ± 0.006, 0.9925 ± 0.0052 and 0.9708 ± 0.0328, at 20× magnification, respectively. Performance of the models varied according to the magnification used. Heatmaps from the two best performing models, Trans-MIL and CLAM, generally indicated attention on appropriate tissue regions for interpretation. Weakly supervised DL on pathological slides of common mucocutaneous melanocytic tumours provides highly accurate diagnostic value for discrimination of melanomas and naevi. External validation and further assessment on less frequently occurring histologic subtypes and borderline cases using this method is required.

Mapping Tumor Habitats in IDH-Wild Type Glioblastoma: Integrating MR Imaging, Pathologic, and RNA Data from Ivy Glioblastoma Atlas Project.

To spatially validate intratumoral subregions (tumor habitat) using physiologic MRI on pathology of the isocitrate dehydrogenase (IDH)-wildtype whole-glioblastoma sample. Data of 20 patients (168 slides) were obtained from the Ivy Glioblastoma Atlas Project. On MRI, tumor habitats were defined using voxel-wise clustering of apparent diffusion coefficient (ADC) and cerebral blood volume (CBV) maps for contrast-enhancing lesion (CEL) and non-enhancing lesion (NEL). On pathology slides, normalized areas of leading edge (LE), infiltrating tumor (IT), cellular tumor (CT), hypervascular lesion (CThypervascular), and perinecrotic lesion (CTperinecrotic) were obtained. Gross specimen was co-registered on MRI and correlation between pathology-MRI habitats was calculated. RNA sequencing of 67 samples was assessed using 4 Neftel subtypes and further correlated with pathology. Six tumor habitats were identified: hypervascular, hypovascular cellular, and hypovascular hypocellular habitats for CEL and NEL. CT was correlated with hypovascular cellular habitat in CEL (r= 0.238, p =.005). IT was correlated with hypovascular cellular habitat in NEL (r= 0.294, p =.017). CThypervascular was correlated with hypervascular habitat in NEL (r= 0.195, p = .023). CTperinecrotic was correlated with imaging necrosis (r= 0.199, p =.005). Astrocyte-like subtypes were correlated with IT (r= 0.256, p <.001), while mesenchymal-like subtypes were correlated with CTperinecrotic area (r= 0.246, p <.001). Pathologically matched tumor subregions were cellular tumor with hypovascular cellular habitat in CEL and infiltrative tumor with hypovascular cellular habitat in NEL. Identification of the most aggressive as well as infiltrative tumor portion can be achieved using non-invasive MRI tumor habitats.

Towards Digital Quantification of Ploidy from Pan-Cancer Digital Pathology Slides using Deep Learning.

Abnormal DNA ploidy, found in numerous cancers, is increasingly being recognized as a contributor in driving chromosomal instability, genome evolution, and the heterogeneity that fuels cancer cell progression. Furthermore, it has been linked with poor prognosis of cancer patients. While next-generation sequencing can be used to approximate tumor ploidy, it has a high error rate for near-euploid states, a high cost and is time consuming, motivating alternative rapid quantification methods. We introduce PloiViT, a transformer-based model for tumor ploidy quantification that outperforms traditional machine learning models, enabling rapid and cost-effective quantification directly from pathology slides. We trained PloiViT on a dataset of fifteen cancer types from The Cancer Genome Atlas and validated its performance in multiple independent cohorts. Additionally, we explored the impact of self-supervised feature extraction on performance. PloiViT, using self-supervised features, achieved the lowest prediction error in multiple independent cohorts, exhibiting better generalization capabilities. Our findings demonstrate that PloiViT predicts higher ploidy values in aggressive cancer groups and patients with specific mutations, validating PloiViT potential as complementary for ploidy assessment to next-generation sequencing data. To further promote its use, we release our models as a user-friendly inference application and a Python package for easy adoption and use.

Virtual multiplexed immunofluorescence staining from non-antibody-stained fluorescence imaging for gastric cancer prognosis

Multiplexed immunofluorescence (mIF) staining, such as CODEX and MIBI, holds significant clinical value for various fields, such as disease diagnosis, biological research, and drug development. However, these techniques are often hindered by high time and cost requirements. Here we present a Multimodal-Attention-based virtual mIF Staining (MAS) system that utilises a deep learning model to extract potential antibody-related features from dual-modal non-antibody-stained fluorescence imaging, specifically autofluorescence (AF) and DAPI imaging. The MAS system simultaneously generates predictions of mIF with multiple survival-associated biomarkers in gastric cancer using self- and multi-attention learning mechanisms. Experimental results with 180 pathological slides from 94 patients with gastric cancer demonstrate the efficiency and consistent performance of the MAS system in both cancer and noncancer gastric tissues. Furthermore, we showcase the prognostic accuracy of the virtual mIF images of seven gastric cancer related biomarkers, including CD3, CD20, FOXP3, PD1, CD8, CD163, and PD-L1, which is comparable to those obtained from the standard mIF staining. The MAS system rapidly generates reliable multiplexed staining, greatly reducing the cost of mIF and improving clinical workflow. Stanford 2022 HAI Seed Grant; National Institutes of Health 1R01CA256890.

A novel model for predicting postoperative liver metastasis in R0 resected pancreatic neuroendocrine tumors: integrating computational pathology and deep learning-radiomics

BackgroundPostoperative liver metastasis significantly impacts the prognosis of pancreatic neuroendocrine tumor (panNET) patients after R0 resection. Combining computational pathology and deep learning radiomics can enhance the detection of postoperative liver metastasis in panNET patients.MethodsClinical data, pathology slides, and radiographic images were collected from 163 panNET patients post-R0 resection at Fudan University Shanghai Cancer Center (FUSCC) and FUSCC Pathology Consultation Center. Digital image analysis and deep learning identified liver metastasis-related features in Ki67-stained whole slide images (WSIs) and enhanced CT scans to create a nomogram. The model’s performance was validated in both internal and external test cohorts.ResultsMultivariate logistic regression identified nerve infiltration as an independent risk factor for liver metastasis (p < 0.05). The Pathomics score, which was based on a hotspot and the heterogeneous distribution of Ki67 staining, showed improved predictive accuracy for liver metastasis (AUC = 0.799). The deep learning-radiomics (DLR) score achieved an AUC of 0.875. The integrated nomogram, which combines clinical, pathological, and imaging features, demonstrated outstanding performance, with an AUC of 0.985 in the training cohort and 0.961 in the validation cohort. High-risk group had a median recurrence-free survival of 28.5 months compared to 34.7 months for the low-risk group, showing significant correlation with prognosis (p < 0.05).ConclusionA new predictive model that integrates computational pathologic scores and deep learning-radiomics can better predict postoperative liver metastasis in panNET patients, aiding clinicians in developing personalized treatments.

Evaluation of prostate cancer detection using micro-ultrasound versus MRI through co-registration to whole-mount pathology

Micro-ultrasound has recently been introduced as a low-cost alternative to multi-parametric MRI for imaging prostate cancer. Early clinical studies have demonstrated promising results; however, robust validation via comparison with whole-mount pathology has yet to be achieved. Due to micro-ultrasound probe design and tissue deformation during scanning, it is difficult to accurately correlate micro-ultrasound imaging planes with ground truth whole-mount pathology slides. In this study, we developed a multi-step methodology to co-register micro-ultrasound and MRI to whole-mount pathology. The three-step process had a registration error of 3.90 ± 0.11 mm and consists of: (1) micro-ultrasound image reconstruction, (2) 3D landmark registration of micro-ultrasound to MRI, and (3) 2D capsule registration of MRI to whole-mount pathology. This process was then used in a preliminary reader study to compare the diagnostic accuracy of micro-ultrasound and MRI in 15 patients who underwent radical prostatectomy for prostate cancer. Micro-ultrasound was found to have equivalent performance to retrospective MRI review for index lesion detection (91.7% vs. 80%), while demonstrating an increased detection of tumor extent (52.5% vs. 36.7%) with similar false positive regions-of-interest (38.3% vs. 40.8%). Prospective MRI review had reduced detection of index lesions (73.3%) and tumor extent (18.9%) but improved false positive regions-of-interest (22.7%) relative to micro-ultrasound and retrospective MRI. Further evaluation is needed with a larger sample size.

AnNoBrainer, An Automated Annotation of Mouse Brain Images using Deep Learning.

Annotation of multiple regions of interest across the whole mouse brain is an indispensable process for quantitative evaluation of a multitude of study endpointsin neuroscience digital pathology. Prior experience and domain expert knowledge are the key aspects for image annotation quality and consistency. At present, image annotation is often achieved manually by certified pathologists or trained technicians, limiting the total throughput of studies performed at neuroscience digital pathology labs. It may also mean that simpler and quicker methods of examining tissue samples are used by non-pathologists, especially in the early stages of research and preclinical studies. To address these limitations and to meet the growing demand for image analysis in a pharmaceutical setting, we developed AnNoBrainer, an open-source software tool that leverages deep learning, image registration, and standard cortical brain templates to automatically annotate individual brain regions on 2D pathology slides. Application of AnNoBrainer to a published set of pathology slides from transgenic mice models of synucleinopathy revealed comparable accuracy, increased reproducibility, and a significant reduction (~ 50%) in time spent on brain annotation, quality control and labelling compared to trained scientists in pathology. Taken together, AnNoBrainer offers a rapid, accurate, and reproducible automated annotation of mouse brain images that largely meets the experts' histopathological assessment standards (> 85% of cases) and enables high-throughput image analysis workflows in digital pathology labs.

AB003. Clinical analysis of central nervous system (CNS) World Health Organization (WHO) grade 4 gliomas with IDH-mutant versus IDH-wildtype focused on CDKN2A homozygous deletion.

We are primarily investigating the prognostic role of cell-cycle-dependent kinase inhibitor (CDKN)-2A homozygous deletion in central nervous system (CNS) World Health Organization (WHO) grade 4 gliomas. Additionally, traditional prognostic factors for grade 4 gliomas will be examined, and our results will be validated. We conducted a retrospective analysis of glioma cohorts in our institute. Medical records were reviewed for 142 glioblastoma patients for 15 years, and pathological slides were examined again for the updated diagnosis according to the 2021 WHO classification of CNS tumors. The isocitrate dehydrase (IDH) mutation and CDKN2A deletion were examined by next generation sequencing (NGS) analysis using ONCO accuPanel®. Traditional prognostic factors including age, WHO performance status, extent of resection, and O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation were examined. After the exclusion of 6 patients with poor status of pathologic samples, 136 glioblastoma that were diagnosed by previous WHO criteria were changed into 29 (21.3%) astrocytoma, IDH-mutant, CNS WHO grade 4 and 107 (78.7%) glioblastoma, IDH-wildtype, CNS WHO grade 4. Among them, 61 patients (56.0%) had CDKN2A deletion. Group A with IDH-wildtype and CDKN2A deletion had a mean overall survival (OS) of 15.70 months [95% confident interval (CI): 13.86-17.54], group B with IDH-mutant and CDKN2A deletion had a mean OS of 19.37 months (95% CI: 13.43-25.30), group C with IDH-wildtype and intact CDKN2A had a mean OS of 22.63 months (95% CI: 20.10-25.17), and group D with IDH-mutant and intact CDKN2A had a mean OS of 33.38 months (95% CI: 29.35-37.40). Multifactor analysis showed following factors were independently associated with OS: age [≥50 vs. <50 years; hazard ratio (HR) 4.642], extent of resection (gross total resection vs. others; HR 5.523), WHO performance (0, 1 vs. 2; HR 5.007), MGMT promoter methylation, (methylated vs. unmethylated; HR 5.075), IDH mutation (mutant vs. wildtype; HR 6.358), and CDKN2A deletion (absence vs. presence; HR 13.452). The presenting study suggests that CDKN2A deletion should play a powerful prognostic role in CNS WHO grade 4 gliomas as well as low-grade glioma. Even if CNS WHO grade 4 gliomas had mutant IDH, they can have poor clinical outcomes due to CDKN2A deletion.

Predicting tumor mutation burden and VHL mutation from renal cancer pathology slides with self-supervised deep learning.

Tumor mutation burden (TMB) and VHL mutation play a crucial role in the management of patients with clear cell renal cell carcinoma (ccRCC), such as guiding adjuvant chemotherapy and improving clinical outcomes. However, the time-consuming and expensive high-throughput sequencing methods severely limit their clinical applicability. Predicting intratumoral heterogeneity poses significant challenges in biology and clinical settings. Our aimed to develop a self-supervised attention-based multiple instance learning (SSL-ABMIL) model to predict TMB and VHL mutation status from hematoxylin and eosin-stained histopathological images. We obtained whole slide images (WSIs) and somatic mutation data of 350 ccRCC patients from The Cancer Genome Atlas for developing SSL-ABMIL model. In parallel, 163 ccRCC patients from Clinical Proteomic Tumor Analysis Consortium cohort was used as independent external validation set. We systematically compared three different models (Wang-ABMIL, Ciga-ABMIL, and ImageNet-MIL) for their ability to predict TMB and VHL alterations. We first identified two groups of populations with high- and low-TMB (cut-off point = 0.9). In two independent cohorts, the Wang-ABMIL model achieved the highest performance with decent generalization performance (AUROC = 0.83 ± 0.02 and 0.8 ± 0.04 in predicting TMB and VHL, respectively). Attention heatmaps revealed that the Wang-ABMIL model paid the highest attention to tumor regions in high-TMB patients, while in VHL mutation prediction, non-tumor regions were also assigned high attention, particularly the stromal regions infiltrated by lymphocytes. Our results indicated that SSL-ABMIL can effectively extract histological features for predicting TMB and VHL mutation, demonstrating promising results in linking tumor morphology and molecular biology.

A Deep Learning Approach for the Identification of the Molecular Subtypes of Pancreatic Ductal Adenocarcinoma Based on Whole Slide Pathology Images

Delayed diagnosis and treatment resistance make pancreatic ductal adenocarcinoma (PDAC) mortality rates high. Identifying molecular subtypes can improve treatment, but current methods are costly and time-consuming. In this study, deep learning models were utilized to identify histological features that classify PDAC molecular subtypes based on routine hematoxylin-eosin (H&E)-stained histopathological slides. 97 histopathology slides associated with resectable PDAC from the Cancer Genome Atlas (TCGA) project were utilized to train a deep learning model and tested the performance on 44 needle biopsy material (110 slides) from a local annotated patient cohort. The model achieved balanced accuracy of 96.19% and 83.03% in identifying the classical and basal subtypes of PDAC in the TCGA and the local cohort, respectively. This study provides a promising method to cost-effectively and rapidly classifying PDAC molecular subtypes based on routine H&E slides, potentially leading to more effective clinical management of this disease.

PathoDuet: Foundation models for pathological slide analysis of H&E and IHC stains

Large amounts of digitized histopathological data display a promising future for developing pathological foundation models via self-supervised learning methods. Foundation models pretrained with these methods serve as a good basis for downstream tasks. However, the gap between natural and histopathological images hinders the direct application of existing methods. In this work, we present PathoDuet, a series of pretrained models on histopathological images, and a new self-supervised learning framework in histopathology. The framework is featured by a newly-introduced pretext token and later task raisers to explicitly utilize certain relations between images, like multiple magnifications and multiple stains. Based on this, two pretext tasks, cross-scale positioning and cross-stain transferring, are designed to pretrain the model on Hematoxylin and Eosin (H&E) images and transfer the model to immunohistochemistry (IHC) images, respectively. To validate the efficacy of our models, we evaluate the performance over a wide variety of downstream tasks, including patch-level colorectal cancer subtyping and whole slide image (WSI)-level classification in H&E field, together with expression level prediction of IHC marker, tumor identification and slide-level qualitative analysis in IHC field. The experimental results show the superiority of our models over most tasks and the efficacy of proposed pretext tasks. The codes and models are available at https://github.com/openmedlab/PathoDuet.

An analysis on efficacy of applying β-elemene intervention on chemically -induced tongue lesions using SAM algorithm.

An artificial intelligence (AI) model was designed to assist pathologists in diagnosing and quantifying structural changes in tongue lesions induced by chemical carcinogens. Using a tongue cancer model induced by 4-nitroquinoline-N-oxide and treated with β-elemene, a total of 183 digital pathology slides were processed. The Segment Anything Model (SAM) was employed for initial segmentation, followed by conventional algorithms for more detailed segmentation. The epithelial contour area was computed using OpenCV's findcontour method, and the skeletonize method was used to calculate the distance map and skeletonized representation. The AI model demonstrated high accuracy in measuring tongue epithelial thickness and the number of papilla-like protrusions. Results indicated that the model group had significantly higher epithelial thickness and fewer papillae compared with the blank group. Furthermore, the treatment group exhibited reduced epithelial thickness and fewer papilla-like protrusions compared with the model group, though these differences were less pronounced. Overall, the SAM framework algorithm proved effective in quantifying tongue epithelial thickness and the number of papilla-like protrusions, thereby assisting healthcare professionals in understanding pathological changes and assessing treatment outcomes.

Histopathology of Allograft Nephrectomies – A Ten Year Observational Study

Background Though infrequent, allograft nephrectomies are performed for early and late graft loss. The study aims to analyze the histopathologic characteristics of allograft nephrectomy specimens. Materials and Methods We conducted an observational study of 103 cases of allograft nephrectomies from 21 centers from 2013 to 2023. All the pathology slides, including hematoxylin and eosin-stained sections, masson trichrome, jones methenamine silver, PAS, GMS, AFB, and immunohistochemistry (C4d, SV40) were reviewed. Pathologic findings were analyzed based on the transplant to nephrectomy interval (0–3 months, &gt; 3 months) and type of donor (deceased, live donor). Results Of the total 103 cases, 77 were male. The mean age at the time of nephrectomy was 36.4 (range 5–64) years. The allografts were obtained from deceased (57) donors and live related (46) donors. Graft tenderness, oliguria/anuria, and fever were common clinical presentations. The majority (71.8%) of the nephrectomies were performed within the first 3 months of renal transplant. Renal vessel thrombosis (32.03%) was the most common pathologic finding. Infections were more common in the first 3 months after the transplant. Fungal infection had a significant association with deceased donor transplantation (p = 0.029). Conclusion Histopathological study of allograft nephrectomy specimens aids understanding of graft loss causes. The study also provides opportunities to prevent complications and implement measures to prolong graft survival in a subsequent transplant.

AI-Based multimodal Multi-tasks analysis reveals tumor molecular heterogeneity, predicts preoperative lymph node metastasis and prognosis in papillary thyroid carcinoma: A retrospective study.

Papillary thyroid carcinoma (PTC) is the predominant form of thyroid cancer globally, especially when lymph node metastasis (LNM) occurs. Molecular heterogeneity, driven by genetic alterations and tumor microenvironment components, contributes to the complexity of PTC. Understanding these complexities is essential for precise risk stratification and therapeutic decisions. This study involved a comprehensive analysis of 521 patients with PTC from our hospital and 499 patients from The Cancer Genome Atlas (TCGA). The real-world cohort 1 comprised 256 patients with stage I-III PTC. Tissues from 252 patients were analyzed by DNA-based next-generation sequencing, and tissues from four patients were analyzed by single-cell RNA sequencing (scRNA-seq). Additionally, 586 PTC pathological sections were collected from TCGA, and 275 PTC pathological sections were collected from the real-world cohort 2. A deep learning multimodal model was developed using matched histopathology images, genomic, transcriptomic, and immune cell data to predict LNM and disease-free survival (DFS). This study included a total of 1,011 PTC patients, comprising 256 patients from cohort 1, 275 patients from cohort 2, and 499 patients from TCGA. In cohort 1, we categorized PTC into four molecular subtypes based on BRAF, RAS, RET, and other mutations. BRAF mutations were significantly associated with LNM and impacted DFS. ScRNA-seq identified distinct T cell subtypes and reduced B cell diversity in BRAF-mutated PTC with LNM. The study also explored cancer-associated fibroblasts and macrophages, highlighting their associations with LNM. The deep learning model was trained using 405 pathology slides and RNA sequences from 328 PTC patients and validated with 181 slides and RNA sequences from 140 PTC patients in the TCGA cohort. It achieved high accuracy, with an AUC of 0.86 in the training cohort, 0.84 in the validation cohort, and 0.83 in the real-world cohort 2. High-risk patients in the training cohort had significantly lower DFS rates (P<0.001). Model AUCs were 0.91 at 1 year, 0.93 at 3 years, and 0.87 at 5 years. In the validation cohort, high-risk patients also had lower DFS (P<0.001); the AUCs were 0.89, 0.87, and 0.80 at 1, 3, and 5 years. We utilized the GradCAM algorithm to generate heatmaps from pathology-based deep learning models, which visually highlighted high-risk tumor areas in PTC patients. This enhanced clinicians' understanding of the model's predictions and improved diagnostic accuracy, especially in cases with lymph node metastasis. The AI-based analysis uncovered vital insights into PTC molecular heterogeneity, emphasizing BRAF mutations' impact. The integrated deep learning model shows promise in predicting metastasis, offering valuable contributions to improved diagnostic and therapeutic strategies.

Global contrast-masked autoencoders are powerful pathological representation learners

Using digital pathology slide scanning technology, artificial intelligence algorithms, particularly deep learning, have achieved significant results in the field of computational pathology. Compared to other medical images, pathology images are more difficult to annotate, and thus, there is an extreme lack of available datasets for conducting supervised learning to train robust deep learning models. In this paper, we introduce a self-supervised learning (SSL) model, the Global Contrast-masked Autoencoder (GCMAE), designed to train encoders to capture both local and global features of pathological images and significantly enhance the performance of transfer learning across datasets. Our study demonstrates the capability of the GCMAE to learn transferable representations through extensive experiments on three distinct disease-specific hematoxylin and eosin (H&E)-stained pathology datasets: Camelyon16, NCT-CRC, and BreakHis. Moreover, we propose an effective automated pathology diagnosis process based on the GCMAE for clinical applications. The source code of this paper is publicly available at https://github.com/StarUniversus/gcmae.

P-320 Ovarian tissue attached to stripped endometriomas: transforming the collateral damage to a potential source for oocyte preservation?

Abstract Study question To determine (antral) follicle density in the ovarian cortex tissue attached to stripped endometriomas and patient- and surgical characteristics affecting this density. Summary answer Ovarian tissue attached to stripped endometriomas offers a non-invasive source for antral follicles. Research with fresh tissue is necessary to further asses the quality. What is known already Ovarian endometriosis is associated with diminished ovarian reserve due to the disease itself as well as damage caused by excisional surgery, which may result in subfertility. When resecting endometriomas with the stripping technique, in the majority of cases, a thin line of adjacent ovarian cortex is attached to the endometrioma. Data regarding density and maturation of the follicles in this tissue are conflicting in previous publications. This study aims to quantify and characterise follicles and in particular antral follicles in ovarian tissue surrounding resected endometriomas as these may serve as a potential oocyte source for in vitro maturation (IVM). Study design, size, duration A retrospective cohort study was conducted at the Leiden University Medical Centre and the Endometriosis in Balance endometriosis expertise centre in Haaglanden Medical Centre (The Netherlands). Pathology slides of patients (≤50 years) that underwent uni- or bilateral endometrioma resection between 2011 and 2021 were analysed for the presence of follicles. Participants/materials, setting, methods Histological slides of previously removed endometriomas were assessed according to maturation and follicular density. Gougeon’s original follicle classification was modified by adding additional intermediate stages. Immunofluorescent staining of antral follicles in a subset of endometriomas was also performed. Main results and the role of chance In 90 out of 96 included endometriomas (93.7%), ovarian tissue attached to the cyst wall was observed. 1944 follicles at different maturation stages were identified (3 follicles/mm3). Follicle density was negatively associated with age (p &amp;lt; 0.001). Antral follicles (&amp;lt;7mm diameter) were present in the ovarian tissue attached to 35 endometriomas (36.5%) derived from younger patients compared to endometriomas where none were detected (30 versus 35 years, p = 0.003). Antral follicle density was 1 follicles/mm3. Based on immunofluorescence, healthy antral follicles were identified in 2 out of 4 examined endometriomas. Limitations, reasons for caution Tissue analysis was constrained by number of available pathology slides per patient, not fully representing the follicle distribution in the removed ovarian tissue. Immunofluorescence was performed on paraffin embedded tissue derived from a small number of endometriomas. Future research with fresh tissue is necessary to further assess viability. Wider implications of the findings Ovarian tissue attached to stripped endometriomas holds potential as a non-invasive source for antral follicles. In theory, application of IVM could be an interesting alternative fertility preservation option in young patients with endometriomas who undergo cystectomy in order to transform the surgical collateral damage to a potential oocyte source. Trial registration number not applicable

Treatment Patterns and Surgical Outcomes of Stage IA1 Cervical Carcinoma

OBJECTIVE The primary objective was to determine the 5-year overall survival rate and 5-year disease-free survival rate of patients with CACx 1A1 treated at Chiang Mai University Hospital and the secondary objective was to describe treatment modalities as well as their associated complications. METHODS Patients with stage IA1 cervical cancer diagnosed from January 2001 to June 2018 at Chiang Mai University Hospital were retrospectively reviewed. Inclusion criteria included previously untreated patients diagnosed with stage IA1 cervical cancer (2018 FIGO staging system). Exclusion criteria were patients with recurrent cervical cancer, other gynecologic malignancies, and those in a pregnant state. The analysis included treatment patterns, surgical types, and clinicopathologic variables, i.e., nodal metastasis, parametrial involvement, positive surgical margins, deep stromal invasion (DSI), lymph-vascular space invasion (LVSI), adjuvant treatment, 5-year disease-free survival, and 5-year overall survival. All pathologic slides were reviewed by gynecologic pathologists. The Kruskal-Wallis test, Fisher’s exact test, the Kaplan-Meier method and log-rank test were used for statistical analysis. RESULTS Of the 184 patients included in this study, simple hysterectomy was the most common treatment (57.3%), followed by modified radical hysterectomy (MRH) and radical hysterectomy (RH) (27.6% and 9.7% respectively). Conization and radiation were chosen in a few cases. At the median follow up time of 40.8 months, the 5-year disease free survival rate was 99.1% and the 5-year overall survival rate was 95.0%. Pelvic lymph node dissection was done in 62 cases (33.7%), but only one case (0.54%) had pelvic lymph node metastasis. CONCLUSIONS The surgical and survival outcomes for women with stage IA1 cervical cancer are excellent. They can be effectively treated with less radical interventions such as simple hysterectomy and conization. Lymph node metastasis is rare at this stage (0.54%); therefore, lympha-denectomy may possibly be omitted. KEYWORDS cervical cancer, IA1, treatment outcome

A Self-Supervised Learning Based Framework for Eyelid Malignant Melanoma Diagnosis in Whole Slide Images.

Eyelid malignant melanoma (MM) is a rare disease with high mortality. Accurate diagnosis of such disease is important but challenging. In clinical practice, the diagnosis of MM is currently performed manually by pathologists, which is subjective and biased. Since the heavy manual annotation workload, most pathological whole slide image (WSI) datasets are only partially labeled (without region annotations), which cannot be directly used in supervised deep learning. For these reasons, it is of great practical significance to design a laborsaving and high data utilization diagnosis method. In this paper, a self-supervised learning (SSL) based framework for automatically detecting eyelid MM is proposed. The framework consists of a self-supervised model for detecting MM areas at the patch-level and a second model for classifying lesion types at the slide level. A squeeze-excitation (SE) attention structure and a feature-projection (FP) structure are integrated to boost learning on details of pathological images and improve model performance. In addition, this framework also provides visual heatmaps with high quality and reliability to highlight the likely areas of the lesion to assist the evaluation and diagnosis of the eyelid MM. Extensive experimental results on different datasets show that our proposed method outperforms other state-of-the-art SSL and fully supervised methods at both patch and slide levels when only a subset of WSIs are annotated. It should be noted that our method is even comparable to supervised methods when all WSIs are fully annotated. To the best of our knowledge, our work is the first SSL method for automatic diagnosis of MM at the eyelid and has a great potential impact on reducing the workload of human annotations in clinical practice.

PhiHER2: phenotype-informed weakly supervised model for HER2 status prediction from pathological images.

Human epidermal growth factor receptor 2 (HER2) status identification enables physicians to assess the prognosis risk and determine the treatment schedule for patients. In clinical practice, pathological slides serve as the gold standard, offering morphological information on cellular structure and tumoral regions. Computational analysis of pathological images has the potential to discover morphological patterns associated with HER2 molecular targets and achieve precise status prediction. However, pathological images are typically equipped with high-resolution attributes, and HER2 expression in breast cancer (BC) images often manifests the intratumoral heterogeneity. We present a phenotype-informed weakly supervised multiple instance learning architecture (PhiHER2) for the prediction of the HER2 status from pathological images of BC. Specifically, a hierarchical prototype clustering module is designed to identify representative phenotypes across whole slide images. These phenotype embeddings are then integrated into a cross-attention module, enhancing feature interaction and aggregation on instances. This yields a phenotype-based feature space that leverages the intratumoral morphological heterogeneity for HER2 status prediction. Extensive results demonstrate that PhiHER2 captures a better WSI-level representation by the typical phenotype guidance and significantly outperforms existing methods on real-world datasets. Additionally, interpretability analyses of both phenotypes and WSIs provide explicit insights into the heterogeneity of morphological patterns associated with molecular HER2 status. Our model is available at https://github.com/lyotvincent/PhiHER2.