Accurate esophageal cancer staging relies on ¹⁸F fluorodeoxyglucose positron emission tomography (¹⁸F FDG-PET), but its interpretation is complex and time-intensive. This diagnostic burden is exacerbated by significant workforce shortages in both radiology and surgery, thus necessitating automated support systems. The emergence of advanced large language models (LLMs) has raised expectations for their potential to fulfill this role in complex medical tasks.

We evaluated the diagnostic accuracy of LLMs for staging esophageal cancer using ¹⁸F FDG-PET images, with a focus on their ability to assess lymph nodes (LNs; clinical N [cN]) and distant metastases (clinical M [cM]) for automated radiology reporting.

This retrospective study included 120 consecutive adult patients who were diagnosed with esophageal squamous cell carcinoma and underwent ¹⁸F FDG-PET/computed tomography at Tohoku University Hospital between January 2019 and December 2021. Patients with prior treatment, nonsquamous cell carcinoma histology, or blood glucose levels ≥200 mg/dL were excluded. Frontal maximum-intensity projection positron emission tomography images were extracted, standardized, and analyzed along with information regarding the tumor location. Six LLMs (GPT-5, GPT-4.5, GPT-4.1, OpenAI-o3, -o1, and GPT-4 Turbo) and 4 blinded human evaluators (a nuclear medicine specialist, a gastrointestinal surgeon, and 2 radiology residents) assessed the presence of thoracic and abdominal LN metastases on a region-level basis and determined cN and cM staging on a patient-level basis. The model analyses were performed using the application programming interface in a zero-shot setting. Radiology reports served as the reference standard. Diagnostic agreement and accuracy were evaluated using Cohen κ and the Cochran Q test. Additionally, to account for the class imbalance in the dataset, the Matthews Correlation Coefficient was calculated as a robust metric for binary classification performance. Post hoc McNemar tests were performed with Bonferroni correction; statistical significance for pairwise comparisons was set at P<.0083 (adjusted from P<.05) using JMP Pro (version 18.0; SAS Institute Inc).

The average accuracy was 41/120 (34%) to 94/120 (78%) for LLMs and 72/120 (60%) to 102/120 (85%) for physicians, with significantly higher accuracy for physicians (P<.05) in the thoracic LN, abdominal LN, and cN stages. Interrater reliability was slight to fair for LLMs (κ: -0.07 to 0.25) and fair to substantial for physicians (κ: 0.27 to 0.74). Matthews Correlation Coefficient scores were consistently higher for physicians (0.28 to 0.75) than for LLMs (-0.07 to 0.32). Among the LLMs, GPT-5 demonstrated the highest overall accuracy, with newer LLMs showing improved diagnostic accuracy when compared with previous models in identifying abdominal LN metastases and cM staging, though they showed weaker consistency for cN staging. For example, in thoracic LN detection, GPT-5 achieved 76/120 (63%) accuracy, whereas other LLMs achieved 72/120 (60%) or lower accuracy.

Although current LLMs have not yet reached physician-level accuracy in comprehensive staging, recent models show promise in assisting with specific diagnostic tasks.

To examine the use of treatments with antibacterial properties as prophylaxis prior to radiotherapy (RT), either alone or in combination with chemotherapy (CT), to prevent and reduce radiation-induced oral mucositis (RIOM) in patients with head and neck cancer (HNC).

A systematic search following PRISMA guidelines was conducted across PubMed, Embase, Web of Science, and the Cochrane Library to identify relevant studies published in English through March 2025.

Eligible studies assessed prophylactic antibacterial interventions aimed at preventing RIOM.

From 86 retrieved citations, 9 articles met inclusion criteria. Antibacterial agents assessed included polymyxin, tobramycin, amphotericin (PTA), povidone iodine, SAMITAL, and Nigella sativa (NS). Evidence supporting povidone iodine, PTA, and SAMITAL was inconclusive or failed to demonstrate statistically significant reductions in RIOM severity. Several studies reported discordant findings, with statistically significant improvements in patient-reported symptoms or quality-of-life measures despite nonsignificant clinician-assessed scores. NS demonstrated potential benefits in reducing RIOM incidence and severity compared with standard of care and other antibacterial agents.

The systematic review highlights limited and inconsistent evidence supporting antibacterial prophylaxis for preventing and reducing RIOM severity in patients with HNC undergoing RT. Discrepancies between patient-reported outcomes and clinical-assessed grading suggest some treatments may provide symptomatic benefit not captured by traditional scoring systems. NS mouthwashes showed preliminary promise; however, evidence remains insufficient to establish superiority, and safety and regulatory concerns are persistent, particularly in immunosuppressed patients. Given the role of bacterial colonization and microbial dysbiosis in RIOM pathogenesis, larger, well-designed clinical trials with rigorous safety evaluations are warranted to investigate bacterial-directed preventive therapies.

Breast cancer is one of the most prevalent and biologically diverse malignancies in women worldwide, encompassing subtypes such as ductal carcinoma in situ (DCIS), lobular carcinoma, and triple-negative breast cancer (TNBC). These variants present complex therapeutic challenges. Chemotherapy remains a core treatment modality, particularly in aggressive or advanced stages, yet its systemic toxicity and lack of specificity limit its efficacy. In recent years, laser-based therapies have emerged as adjunctive strategies to enhance therapeutic precision. This review explores breast cancer classification, progression, and treatment, with an emphasis on chemotherapy, and critically examines the emerging role of laser technologies, including low-level laser therapy (LLLT), Photochemotherapy, and Photodynamic Therapy (PDT), as adjunctive or alternative therapeutic options. We highlight the potential of laser to modulate the tumor microenvironment, improve drug delivery, regulate mitochondrial function, and enhance apoptosis. PDT showed promise in activating localized cytotoxic effects while sparing surrounding tissues. However, heterogeneity in laser parameters and treatment protocols remains a significant barrier to clinical translation. This review underscores the translational potential of laser-assisted chemotherapy, identifies current gaps, and suggests future research directions for optimized treatment strategies in breast oncology.

DNA profiling is an established method for cancer treatment selection, while RNA profiling remains investigational. We explored associations between DNA and RNA alterations and between the number of genes with altered expression and overall survival (OS) using patient data from IMPACT2 (NCT02152254), a randomized study evaluating molecular profiling for guiding cancer therapy across tumor types. Molecular profiling, including DNA next-generation sequencing, was performed on all 829 patients in the IMPACT2 study. RNA profiling was performed by Tempus for 253 of 829 patients. We evaluated the concordance between DNA and RNA profiling, analyzed OS in 217 treated patients with RNA profiling, and assessed PD-L1 status and number of genes with altered expression. Fifty patients exhibited 58 concordant events, i.e., genomic and expression alteration(s) in the same gene, including 38 copy number events, and 41 patients had statistically significant concordance. We identified 123 gene pairs with significant associations between genomic and expression alterations (p < 0.05), including TP53 alterations with VEGFA overexpression. The median OS for patients with 0-2, 3-5, and ≥6 genes with altered expression was 9.8, 11.9, and 6.7 months, respectively (p = 0.03). These results underscore RNA profiling's potential actionability, and altered expression in ≥6 genes was associated with shorter OS. Significant concordance of TP53 alterations with VEGFA overexpression may partially explain tumor response to bevacizumab in TP53-mutant patients.

Endometrial stromal sarcoma (ESS) is a rare uterine malignancy originating from endometrial connective tissue. It is classified into benign and malignant subtypes. While it primarily affects perimenopausal women aged 45-55 years, younger patients may also be affected. ESS often presents with symptoms such as abnormal uterine bleeding, pelvic pain or pelvic mass, mimicking benign conditions like fibroids. Definitive treatment includes total hysterectomy with bilateral salpingo-oophorectomy, complemented by hormonal therapy for advanced cases.Despite its indolent nature, low-grade ESS (LGESS) requires long-term follow-up due to significant recurrence risk. We present a case of a woman in her late 20s, who underwent hysterectomy for severe bleeding and anaemia, with LGESS diagnosed later during a growing abdominal wall mass evaluation.Imaging and biopsy confirmed her diagnosis. She underwent extensive debulking surgery. Histological analysis revealed oval to spindle cells, low mitotic activity, no necrosis or atypical mitotic figures; Estrogen Receptor (ER), Progesterone Receptor (PR) and CD10 positivity, which were consistent with LGESS.

Kaposi's sarcoma (KS) is a malignant vascular neoplasm associated with human herpesvirus type 8 (HHV-8) infection. Conjunctival involvement is very rare and is considered part of classic KS. We present the case of a 57-year-old female patient with skin lesions diagnosed as classic KS not associated with HIV or immunosuppression, who later developed a conjunctival lesion. This lesion was treated with superficial radiotherapy in a single fraction using a low-energy X-ray device, resulting in good clinical outcomes and no signs of conjunctival recurrence after 5 years of follow-up. Single-fraction radiotherapy is an effective treatment for local control of the lesion and prevention of recurrence.

The identification of the central lymphatic system represents a significant advancement in contemporary neuroscientific research. As a vital conduit connecting the central nervous system to the peripheral lymphatic network, the primary components of the central lymphatic system, the glymphatic system and meningeal lymphatic vessels, maintain homeostasis within the central microenvironment through cerebrospinal fluid circulation, metabolic waste clearance, and immune cell trafficking. Recent studies affirm that structural and functional perturbations within the central lymphatic apparatus can lead to the accrual of neurotoxic proteins such as amyloid-β and tau, thereby contributing to the pathogenesis of neurodegenerative conditions like Alzheimer's and Parkinson's diseases. Additionally, in cerebrovascular pathology, impaired drainage mechanisms exacerbate blood-brain barrier disruption and cerebral edema, thereby impairing neurological recovery. Moreover, this system modulates immune cell infiltration and plays a pivotal role in the etiology and progression of multiple sclerosis, central nervous system infections, and intracranial neoplasms.

This review aimed to systematically synthesize the core components of the cerebral lymphatic system, namely the glymphatic system and meningeal lymphatic vessels, and clearly outline the key influencing factors, including aquaporin-4 activity, aging, hemodynamics, sleep, and circadian rhythms. The study aimed to emphasize the mechanisms by which the cerebral lymphatic system influences neurological pathology and summarize emerging therapeutic avenues targeting lymphatic pathways within the central nervous system.

Evidence suggests that the targeted modulation of the cerebral lymphatic system has the potential to treat various neurological diseases. Nevertheless, our understanding of the molecular mechanisms remains in its early stages, with significant research gaps. Consequently, future investigations should focus on molecular characterization to identify specific therapeutic targets, thereby facilitating the translation of preclinical insights into clinical practice.

Brain tumor is a common neurological surgical disease, where surgical resection is the primary treatment method. Neurosurgeons need to accurately determine the location of the tumor during tumor resection surgery, but existing clinical tumor identification technologies face numerous challenges, such as high equipment costs, long processing times, a certain degree of invasiveness, and insufficient image clarity. In this work, we propose a hyperspectral image detection algorithm based on the fusion of multiple features to maximize the determination of tumor boundaries. The algorithm establishes the machine learning models of Support Vector Machine (SVM) and Random Forest (RF) by integrating data features from optimal wavelengths, spectral indices, and textural features. Experimental results show that on different datasets, the classification accuracy of the three-feature fusion model is significantly higher than that of models using only two features or a single feature. Hyperspectral tumor image recognition can effectively help distinguish the tumors from the surrounding tissue, thereby enhancing the safety and thoroughness of tumor surgery.

To evaluate a multiparametric MRI protocol encompassing intravoxel incoherent motion diffusion-weighted imaging, diffusion kurtosis imaging, and dynamic contrast-enhanced MRI for discriminating aggressive (AEC) from non-aggressive type endometrial carcinoma (NAEC) according to the FIGO 2023 staging system.

This study involved retrospective analyses of a prospective dataset. 112 consecutive patients (77 NAEC and 35 AEC) underwent multiparametric MRI. Intratumoral and peritumoral quantitative MRI parameters were calculated. A multivariate logistic regression model comprising clinical data, conventional MRI features, and quantitative MRI metrics was constructed. Model performance was evaluated using receiver operating characteristic analysis, calibration curves, and bootstrap resampling (n = 1000).

AEC demonstrated significantly lower perfusion fraction (f) and mean diffusivity (MD), but higher pseudo diffusion coefficient (D*) and peritumoral D* (D*_peri) compared to NAEC (all p < 0.05). Multivariate analysis identified f, peritumoral mean kurtosis (MK_peri), and peritumoral maximum slope of increase (MaxSlope_peri) as independent predictors of AEC (AUC = 0.791, 95% CI: 0.692-0.891). Integration of menopause status, tumor location and extension beyond corpus, and quantitative MRI parameters yielded a combined model with a stratified bootstrap AUC of 0.830 (95% CI: 0.800-0.846), particularly for FIGO 2023 stage I-II patients (AUC = 0.832, 95%CI: 0.742-0.922). Significant differences in D*, D*_peri, f, and MD were observed among NAEC with and without squamous differentiation and AEC groups.

Multiparametric MRI, incorporating advanced quantitative sequences and conventional MRI features, could help effectively predict AEC before surgery.

The study bridges advanced imaging with the updated FIGO 2023 staging system, potentially offering a non-invasive assessment tool for endometrial carcinoma management.

Early diagnosis of lung cancer plays a pivotal role in ensuring improved treatment and survival of patients. This remains a major focus in clinical research. Artificial intelligence (AI) has transformed pathology by significantly improving diagnostic accuracy and efficiency. This study presents a robust deep learning model in the shape of the pretrained EfficientNetB7 model to classify colon and lung tissue histopathological images with an extremely high accuracy of 96%. The model's performance was optimized using advanced preprocessing methods, fine-tuning, and domain-specific data augmentation techniques. These strategies help reduce problems such as class imbalance and subtle histological variations. To address the issue of overfitting, multiple data augmentation techniques were combined, and an early stopping criterion was incorporated. This approach enabled efficient and cost-effective training. Robust validation of the model demonstrates high utility for clinical applications and enables pathologists to deliver timely and accurate diagnoses. Integrating advanced deep learning models into medical imaging workflows holds great promise for early and accurate cancer diagnosis, ultimately improving patient outcomes.