The history of screening for cervical cancer is rich with implementing cutting-edge ideas and technologies. From the very first “Pap smear” to the semi-automated and computerized systems of today, the way we screen for cervical cancer has changed dramatically in the past 75 years. With the advent of new techniques and more advanced machine learning algorithms, we sought to understand the current and future applications of artificial intelligence in clinical pathology around cervical cancer screening, diagnosis, and treatment.

A structured narrative review was conducted to examine the historical evolution and contemporary advances in cervical cancer screening, diagnosis, excisional management, and artificial intelligence applications. Peer-reviewed articles, consensus guidelines, and global policy documents published between January 1990 and March 2025 were identified through targeted searches of PubMed and review of reference lists from relevant publications. Search terms included combinations of “cervical cancer screening,” “Papanicolaou test,” “liquid-based cytology,” “HPV testing,” “colposcopy,” “loop electrosurgical excision procedure,” “digital pathology,” “deep learning,” and “artificial intelligence.” Emphasis was placed on multi-center validation studies, systematic reviews, regulatory and implementation guidance, and global health frameworks. Publications lacking methodological transparency or direct relevance to clinical or translational practice were excluded.

Through a review of the literature, we describe how innovations in conventional and liquid-based cytology, human papillomavirus testing, and organized screening programs established the current prevention framework. Building on this foundation, recent studies demonstrate promising performance of deep learning algorithms applied to conventionally prepared cervical cytology slides, with systems capable of binary normal versus abnormal classification as well as more granular diagnostic categorization. Artificial intelligence-assisted colposcopy and computer-vision approaches have also shown improved diagnostic concordance, workflow efficiency, and potential to expand screening capacity in resource-limited environments.

There has been much work done in the past several years surrounding the implementation of deep learning algorithms in regard to cervical cancer screening. The work in this field shows promise in enhancing diagnostic accuracy, streamlining diagnostic workflow, and decreasing turnaround times from specimen collection to rendering a diagnosis. However, there are still many technical, legal, and ethical questions that must be answered prior to widespread adoption of these algorithms for patient care.

Further decompensation in cirrhosis is associated with increased mortality. However, reliable tools to predict further decompensation after transjugular intrahepatic portosystemic shunt (TIPS) are currently limited. This study aimed to investigate the incidence and risk factors of further decompensation within one year post-TIPS in patients with cirrhosis and to develop a predictive model for identifying high-risk individuals.

This retrospective cohort study enrolled 152 patients with cirrhosis undergoing TIPS for variceal bleeding and/or refractory ascites (January 2018–January 2024). Patients were stratified according to one-year decompensation outcomes. LASSO regression and multivariable logistic analysis were used to identify predictors, and a nomogram was constructed and internally validated using bootstrapping (1,000 replicates).

Among the 152 patients (median age 57.5 years [IQR 50.0–66.0]; 58.6% male; 58.6% viral/alcohol-associated etiology), 65.8% (100/152) achieved clinical stability at one year post-TIPS, while 34.2% (52/152) developed further decompensation. LASSO regression identified right hepatic lobe volume, spleen volume, and portal pressure gradient (PPG) reduction as key predictors, all independently associated with further decompensation risk in multivariable analysis (OR [95% CI]: 0.683 [0.535–0.873], 1.435 [1.240–1.661], and 0.961 [0.927–0.996], respectively). The nomogram demonstrated superior discrimination compared with PPG reduction alone and benchmark prognostic scores (AUC 0.854 [0.792–0.915] vs. 0.619–0.652; ΔAUC +0.201–+0.235, p < 0.001) with 92.3% sensitivity. High-risk patients (score > 86) had a 10.7-fold higher risk of further decompensation than low-risk patients (60.0% vs. 5.6%; p < 0.0001).

This validated model, combining hepatosplenic volumetry and PPG reduction, accurately stratifies further decompensation risk post-TIPS and may guide targeted surveillance and preventive interventions.

Portal vein thrombosis (PVT) frequently occurs in patients with porto-sinusoidal vascular disease (PSVD), but its clinical characteristics and outcomes remain poorly understood. This study aimed to investigate the clinical features and outcomes of PVT in PSVD.

A total of 169 patients with PSVD confirmed by hepatic histology were included. PVT was diagnosed using contrast-enhanced magnetic resonance imaging or computed tomography. Demographic, clinical, and laboratory data, portal hypertension-related complications, comorbidities, and mortality were collected and compared between patients with and without PVT. The primary outcomes were baseline clinical characteristics and liver-transplantation-free mortality; the secondary outcome was the dynamic changes of PVT during follow-up.

At baseline, 45 (26.6%) PSVD patients had PVT. Compared to those without PVT, patients with PVT had significantly higher rates of esophageal variceal bleeding (62.2% vs. 29.0%), ascites (73.3% vs. 35.5%), antithrombin III deficiency (78.1% vs. 38.4%) (all p < 0.001), and a history of hematological disorders (11.1% vs. 0.8%, p = 0.005). After a median follow-up of 40.1 (23.4–62.3) months, liver-transplantation-free mortality rates were 7.9% (3/38) and 1.8% (2/112) in patients with and without PVT, respectively (log-rank p = 0.110). Among 41 patients followed for a median of 17.1 (7.4–39.3) months, PVT resolved in 9.1% (1/11) of those with baseline PVT and developed in 13.3% (4/30) of those without PVT at baseline. The one- and two-year cumulative incidence rates of PVT were 3.3% and 6.7%, respectively.

PSVD patients with PVT experience more portal hypertension-related complications, complex coagulation profiles, hematological disorders, and a higher risk of death compared to those without PVT.

Diabetic cardiomyopathy (DCM), a diabetes-specific cardiovascular complication, is pathologically characterized by cardiomyocyte apoptosis, oxidative stress, inflammatory responses, and myocardial fibrosis, distinguishing it from other cardiac disorders, such as hypertension and coronary artery disease. Challenges in early diagnosis, coupled with the limited efficacy and adverse effects of current treatments, have made DCM a significant contributor to heart failure and mortality in patients with diabetes. Natural products, recognized for their diverse sources, structural variety, and multitarget therapeutic potential, have shown promise in preventing and treating DCM. Drawing on advances over the past five years, this review systematically summarizes the pharmacological effects and molecular mechanisms of natural products (e.g., flavonoids, terpenoids, phenylpropanoids, alkaloids, and polysaccharides) in the treatment of DCM, with the aim of providing a theoretical foundation for further research and drug development.

Febrile neutropenia (FN) is one of the acute and serious complications of chemotherapy-induced myelosuppression in tumor patients. Antibiotics and granulocyte colony-stimulating factor are the mainstays of its treatment. However, this therapy still faces many challenges and may trigger drug resistance, as well as adverse effects such as bone pain and vasculitis. How to minimize treatment-related toxicity while ensuring therapeutic efficacy has become a key issue to be addressed in current clinical practice. In recent years, traditional Chinese medicine (TCM) has demonstrated unique advantages in the prevention and treatment of FN. We conducted a comprehensive search of the PubMed, Web of Science, and CNKI databases using keywords such as TCM and FN, covering the period from their establishment to May 2025. Clinical studies have shown that the combination of TCM and modern medicine can significantly reduce the incidence of FN, while also enhancing the number of granulocytes, shortening the duration of fever, improving the quality of life of patients, and reducing other toxic effects of chemotherapy. These results suggest that TCM is a promising and safe complementary therapy. However, more high-quality trials are needed to verify its benefits. This review summarizes the latest progress in the treatment of FN with TCM and discusses future development directions.

Acute-on-chronic liver failure (ACLF) is a life-threatening syndrome characterized by systemic inflammation and immune dysregulation, in which macrophages play a key role in organ injury. This study aimed to investigate the role and mechanism of pyruvate dehydrogenase kinase 4 (PDK4) in ACLF to identify therapeutic targets that modulate macrophage function and mitigate ACLF progression.

Single-cell RNA sequencing data from healthy and ACLF liver tissues were analyzed from the Sequence Read Archive database. Transcriptomic data of peripheral blood mononuclear cells from ACLF patients (GSE168048) were also examined. In vitro experiments assessed PDK4 expression and macrophage polarization, and conditioned-medium studies evaluated effects on LO2 hepatocytes. In vivo validation was performed in ACLF mouse models treated with a PDK4 inhibitor.

Single-cell analysis revealed a predominance of M1-polarized hepatic macrophages in ACLF with marked upregulation of PDK4. Peripheral blood mononuclear cell transcriptomics showed that higher PDK4 expression correlated with 28-day mortality. In vitro, PDK4 expression increased in M1 macrophages; PDK4 inhibition attenuated M1 polarization and reduced cytotoxic effects on LO2 cells. In vivo, pharmacologic inhibition of PDK4 suppressed M1 polarization in macrophages, alleviated liver inflammation, and reduced tissue injury. Mechanistically, PDK4 promoted M1 polarization via activation of signal transducer and activator of transcription 1 signaling.

PDK4 is a key pro-inflammatory regulator in ACLF by promoting M1 macrophage polarization. Targeting PDK4 may be a promising strategy to attenuate inflammation and improve clinical outcomes in ACLF.

In the past decade, immune checkpoint inhibitors (ICIs) have dramatically changed cancer treatment, significantly improving outcomes for patients with various malignancies. Nonetheless, their widespread application has resulted in a rise in immune-related adverse events due to excessive immune activation, including immune-mediated hepatotoxicity (IMH). IMH can cause serious complications and even death, underscoring the need for early prediction and intervention. This review outlines the current understanding of risk factors and predictive biomarkers for IMH in cancer patients undergoing ICI therapy, with risk factors divided into patient-associated, tumor-associated, and agent-associated categories. Higher IMH risk is related to female sex, younger age, extreme BMI, Asian ethnicity, and chronic liver disease. Cancer type, prior ICI treatment, dual ICI combination therapy, and the concurrent use of chemotherapy, targeted agents, or other hepatotoxic drugs (e.g., acetaminophen, statins) also increase the risk of IMH. Potential predictive biomarkers encompass circulating blood cells, serum proteins, autoantibodies, cytokines, gene profiles, and the gut microbiome. Despite promising findings, the predictive value of these biomarkers remains inconsistent, and no definitive biomarker has been established for routine clinical use. Large-scale prospective studies are essential to verify the predictive value of these biomarkers and facilitate their integration into clinical practice, thereby providing deeper insights into the early identification and individualized management of IMH during ICI therapy.

Chimeric antigen receptor (CAR)-T cell therapy faces significant challenges in treating solid tumors, including immune evasion, suppressive tumor microenvironments, and on-target/off-tumor toxicity, which limit its clinical efficacy. Although it has revolutionized treatment for hematological malignancies, these obstacles hinder its broader application in solid tumors. Nanotechnology offers innovative strategies to address these limitations through enhanced delivery, localization, and control. This review summarizes recent advances in nanotechnology-assisted CAR-T cell therapies for gynecologic cancers, with a particular focus on messenger RNA (mRNA)-based delivery systems, lipid nanoparticles, hydrogels, and external activation techniques such as photothermal and acoustogenetic modulation. The integration of nanotechnology, especially mRNA-based delivery systems, holds transformative potential for overcoming these barriers. mRNA enables transient, non-integrating expression of CARs, meaning the genetic modifications are temporary. This improves safety and allows flexible control over treatment intensity, while rational sequence optimization (e.g., codon usage, guanine-cytosine content, secondary structure) enhances mRNA stability and protein translation efficiency. Lipid nanoparticles, the leading delivery platform, can be engineered for cell-type specificity and tissue targeting through modulation of their components and surface functionalization. Recent innovations, including siloxane-modified lipid nanoparticles, injectable hydrogels, and photothermal or acoustogenetic activation strategies, enable precise spatiotemporal control of CAR-T cell function in vivo. In ovarian cancer, preclinical studies targeting nfP2X7 and employing multifunctional nanoparticles have demonstrated synergistic efficacy and tumor-specific delivery. This review highlights how nanotechnology platforms can be integrated with CAR-T cell therapies to enhance safety, precision, and therapeutic outcomes in ovarian cancer.

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.

Sterol regulatory element-binding protein 1 (SREBP1), a key regulator of lipogenesis, is highly expressed in tumors, but the mechanisms sustaining its elevated levels remain unclear. The role of UFMylation, a posttranslational modification, in modulating SREBP1 stability and tumor progression has not been explored. This study aimed to investigate the role of UFMylation in the progression of liver cancer.

Liquid chromatography-tandem mass spectrometry was employed to investigate the interacting proteins of ubiquitin-fold modifier 1-specific ligase 1 (UFL1). Knockdown of UFL1 and DDRGK domain-containing protein 1 (DDRGK1) was performed to assess SREBP1 stability. In vitro and in vivo models of hepatocellular carcinoma (HCC) were used to evaluate tumor progression. Clinical correlations between UFL1/DDRGK1 and SREBP1 levels were analyzed in HCC patient samples.

SREBP1 undergoes UFMylation, which synergizes with ubiquitination to reduce its stability. Depletion of UFL1 or DDRGK1 increased SREBP1 stability, driving HCC progression. Clinically, UFL1 and DDRGK1 levels were reduced in HCC tissues and inversely correlated with SREBP1 expression. Fatostatin (an SREBP1 inhibitor) enhanced the therapeutic effect of Lenvatinib in HCC models with low UFL1 expression.

UFMylation is a critical posttranslational modification that destabilizes SREBP1, and its dysregulation contributes to HCC progression. Targeting the UFMylation-SREBP1 axis, particularly through Fatostatin and Lenvatinib combination therapy, represents a novel therapeutic strategy for HCC.