Mediastinal germ cell tumors (GCTs) are rare malignant neoplasms that occasionally develop somatic-type malignancies (SMs), such as sarcomas, carcinomas, and hematologic malignancies.

We report a unique case of a 16-year-old male patient with a mediastinal GCT that simultaneously developed two different SMs: well-differentiated angiosarcoma and acute megakaryoblastic leukemia (AML). The patient initially presented with left shoulder pain and intermittent shortness of breath. The imaging study demonstrated a 12.5 × 9.0 × 8.5 cm heterogeneous mass in the left anterior mediastinum. The mediastinal mass was resected and showed a cystic mature teratoma with somatic transformation into well-differentiated angiosarcoma and AML. A subsequent bone marrow biopsy confirmed the diagnosis of AML, and next-generation sequencing demonstrated the presence of PTEN and TP53 gene mutations in the AML. Despite aggressive chemotherapy and allogeneic stem cell transplantation, the patient died 10 months after diagnosis.

Our report demonstrates the unique capability of mediastinal GCTs to simultaneously develop two different SMs. The presence of two different SMs in mediastinal GCTs is associated with extremely aggressive behavior and a poor prognosis.

Normalization can standardize and improve machine learning (ML) performance on omics data. However, it is unclear whether normalization is associated with overfitting (i.e., worse cross-dataset performance than intra-dataset performance). Therefore, we aimed to examine associations of normalization and regularization with overfitting of ML on omics data.

Using three paired transcriptomic and clinical datasets (lung adenocarcinoma: the Cancer Genome Atlas (TCGA)/Oncology Singapore; melanoma: TCGA/Dana-Farber Cancer Institute; glioblastoma: TCGA/Clinical Proteomic Tumor Analysis Consortium), we applied ANOVA-based gene selection methods, six normalization methods, and six ML models to classify cancer patients’ deaths. Balanced accuracy (BA) and area under the curve (AUC) in intra- and cross-dataset settings were compared using inferential analyses.

Normalization consistently improved intra-dataset performance (median BA/AUC changes: 0.035–0.214/0.115–0.279) on all data, particularly with Z_Raw, but decreased or slightly increased cross-dataset performance (median BA/AUC changes: −0.029–0.079/0.029–0.064). Least Absolute Shrinkage and Selection Operator (LASSO) model without normalization consistently outperformed most of the ML models in cross-dataset testing across cancer types. ML models on all and molecular-alone data showed similar best performances.

Normalization increases ML’s intra-dataset performance and overfitting in three paired cancer transcriptomic and clinical datasets. Regularized models such as LASSO appear to mitigate overfitting and achieve robust cross-dataset performance. Therefore, cross-dataset evaluation and regularized models are recommended to assess and reduce overfitting, while normalization should be used cautiously. Adding clinical data seems to have little impact on ML models’ performance. However, future work on other diseases and datasets is warranted.

Colorectal cancer histopathological grading relies on the accurate segmentation of glandular structures. Current deep learning–based methods depend heavily on large-scale pixel-level annotations that are labor-intensive and not amenable to clinical practice. Weakly supervised semantic segmentation offers a promising alternative; yet, existing class activation map–based weakly supervised semantic segmentation approaches often produce incomplete, low-quality pseudo-masks that overemphasize discriminative regions and fail to provide reliable supervision for unannotated glandular structures, limiting their suitability for dense histopathology segmentation under sparse supervision. We propose a novel weakly supervised teacher–student framework that leverages sparse pathologists’ annotations and an Exponential Moving Average–stabilized teacher network to generate refined pseudo-masks.

Our framework integrates confidence-based filtering, adaptive fusion of teacher predictions with limited ground truth, and curriculum-guided refinement, enabling the student network to progressively delineate and accurately segment unannotated glandular regions. We validated our framework on an institutional colorectal cancer cohort from The Ohio State University Wexner Medical Center, consisting of 60 hematoxylin and eosin-stained whole-slide images from independent patients with varying degrees of gland differentiation, as well as on public benchmarks including the Gland Segmentation dataset (derived from stage T3–T4 colorectal adenocarcinomas), TCGA-COAD, TCGA-READ, and SPIDER.

The proposed framework achieved strong performance on the institutional dataset despite limited annotations. On the Gland Segmentation dataset, it demonstrated competitive performance compared to both weakly and fully supervised approaches, achieving a mean Intersection over Union of 80.10% ± 1.52 and a mean Dice coefficient of 89.10% ± 2.10. Moreover, cross-cohort evaluations showed robust generalization on TCGA-COAD and TCGA-READ without requiring additional annotations, while reduced performance on SPIDER reflected pronounced domain shift.

Our framework provides an annotation-efficient and generalizable paradigm for accurate gland segmentation in colorectal histopathology, offering a practical pathway toward significantly reducing annotation burdens while preserving high segmentation fidelity.

The International Academy of Cytology Yokohama System for Reporting Breast Fine-Needle Aspiration Biopsy Cytology was developed by a group of expert cytopathologists and clinicians in the breast field. Five categories are defined to stratify breast lesions by their risks of malignancy and managed accordingly. Clinical and radiologic information (triple test) are critical for further management. Ultrasound guidance, and rapid on-site evaluation are valuable for improving the rate of definitive diagnosis. Ancillary studies can be tested on cytologic samples to provide prognostic, predictive information as well as diagnostic clues. Based on many published studies in different institutions worldwide, the implementation of the system appears to have been successful. However, further studies are important for improvements and modifications to the current system.

The Bethesda System for Reporting Thyroid Cytopathology is the first reporting system established in non-gynecologic cytopathology after the Bethesda System for Reporting Cervical Cytopathology. It adopts the same concept of providing a uniform reporting system for all pathologists and clinicians to follow in thyroid cytology. The reporting system is composed of six diagnostic categories. There are defined diagnostic criteria, estimated risk of malignancy, and management recommendations for each category. The reporting system has undergone two revisions upon the emergence of a new entity, updated terminology in histology, and development and refinement of molecular testing. The third edition is soon to be published in 2023. This review will provide an updated summary of the reporting system. Potential diagnostic pitfalls and molecular testing are also discussed.

The Paris System for Reporting Urinary Cytology (TPS 1.0, published in 2016) is the first standardized, evidence-based reporting system established as an international reporting system for urine specimens after the International Congress of Cytology in Paris. Its primary purpose is to reduce the rate of unnecessary indeterminate diagnoses but maintain the excellent performance of urine cytology, in detecting high-grade urothelial carcinoma. The reporting system comprises six diagnostic categories, as well as each category’s diagnostic criteria, estimated risk of malignancy, and management recommendations. After six years, TPS 2.0 was applied in 2022 upon the unfolding of new data. TPS 2.0 clarifies the diagnosis categories and updates the risk of malignancy in each category and developments of molecular tests. This review provides an updated summary of TPS 2.0. Some diagnostic pitfalls and molecular tests were also discussed.