This study investigated the usefulness of deep learning-based automatic detection of temporomandibular joint (TMJ) effusion using magnetic resonance imaging (MRI) in patients with temporomandibular disorder and whether the diagnostic accuracy of the model improved when patients\' clinical information was provided in addition to MRI images. The sagittal MR images of 2948 TMJs were collected from 1017 women and 457 men (mean age 37.19 ± 18.64 years). The TMJ effusion diagnostic performances of three convolutional neural networks (scratch, fine-tuning, and freeze schemes) were compared with those of human experts based on areas under the curve (AUCs) and diagnosis accuracies. The fine-tuning model with proton density (PD) images showed acceptable prediction performance (AUC = 0.7895), and the from-scratch (0.6193) and freeze (0.6149) models showed lower performances (p < 0.05). The fine-tuning model had excellent specificity compared to the human experts (87.25% vs. 58.17%). However, the human experts were superior in sensitivity (80.00% vs. 57.43%) (all p < 0.001). In gradient-weighted class activation mapping (Grad-CAM) visualizations, the fine-tuning scheme focused more on effusion than on other structures of the TMJ, and the sparsity was higher than that of the from-scratch scheme (82.40% vs. 49.83%, p < 0.05). The Grad-CAM visualizations agreed with the model learned through important features in the TMJ area, particularly around the articular disc. Two fine-tuning models on PD and T2-weighted images showed that the diagnostic performance did not improve compared with using PD alone (p < 0.05). Diverse AUCs were observed across each group when the patients were divided according to age (0.7083-0.8375) and sex (male:0.7576, female:0.7083). The prediction accuracy of the ensemble model was higher than that of the human experts when all the data were used (74.21% vs. 67.71%, p < 0.05). A deep neural network (DNN) was developed to process multimodal data, including MRI and patient clinical data. Analysis of four age groups with the DNN model showed that the 41-60 age group had the best performance (AUC = 0.8258). The fine-tuning model and DNN were optimal for judging TMJ effusion and may be used to prevent true negative cases and aid in human diagnostic performance. Assistive automated diagnostic methods have the potential to increase clinicians\' diagnostic accuracy.

Precision medicine translates through molecular assays and in minimally invasive diagnosis, evident in analyses of effusions that serve therapeutic and diagnostic purposes. This cost-effective and low-risk approach provides advantages, playing a pivotal role in late-stage oncology and frequently standing as the primary resource for cancer diagnosis and treatment pathways. This article outlines the workflow for managing serous fluid and explores how cytology effusion analysis extends beyond immunocytological diagnosis. Combined with current molecular tests it showcases the potential to be a skillful tool in precision cytopathology.

BACKGROUND: Sarcomas presenting as malignant effusions are rare, and diagnosing them on fluid cytology requires expertise and clinicoradiological correlation as cells undergo morphological changes, mimicking carcinoma or mesothelioma.
METHODS: We present a case of a 70-year-old man with abdominal distention and pain, initially suggestive of carcinoma on peritoneal fluid cytology. However, subsequent analysis with immunohistochemistry on the cell block revealed diffuse nuclear positivity for MDM2, leading to the diagnosis of dedifferentiated liposarcoma.
CONCLUSIONS: The cytological diagnosis of dedifferentiated liposarcoma is challenging and requires a high index of suspicion, with clinicoradiological correlation. Utilizing immunohistochemistry on cell block samples enhances diagnostic accuracy and guides appropriate patient management.

Current literature lacks data regarding the influence of serous fluid volume (SFV) on next-generation sequencing (NGS) performed on malignant cases. In this study, we highlight the impact of SFV and other parameters influencing the outcome of NGS analysis. We evaluated 827 samples of serous fluids from 607 patients. Of these, 72 samples underwent NGS analysis. Effusion volume, tumor cellularity, DNA, and RNA quality metrics, as well as clinicopathologic and molecular data were evaluated. Pleural fluid accounted for 56.3% of the fluid samples collected. The most common primary tumor site was gastrointestinal/pancreatobiliary, adenocarcinoma was the most common histologic type. Overall mean volume was 293 mL. The mean Qubit DNA of the 72 effusion samples that underwent NGS analysis was 14.3 ng/μL and mean Qubit RNA was 28.2 ng/μL. The mean Qubit DNA concentration increases in SFV up to 100 mL only. No correlation exists between SFV and mean tumor cellularity. In addition, 74.6% (50 of 67) of sequenced samples showed oncogenic drivers; KRAS was the most common driver followed by EGFR. Three cases displayed ALK fusions, and 1 case displayed NTRK1 fusion. The DNA yield is higher in SFV of 100 mL as a cutoff. Beyond 100 mL, there is no impact of SFV on DNA yield. SFV does not impact RNA yield and mean tumor cellularity. Effusion samples should be submitted for molecular testing despite low tumor cellularity. Our results as a pilot study are important in optimization of SFV for both diagnosis as well as NGS analysis for improving management.

OBJECTIVE: Neonatal systemic lupus erythematosus (NLE) is an acquired autoimmune disease. The presence of effusions, such as pleural effusion and pericardial effusion, is rare. The present study helped investigate the clinical characteristics and progression of children with NLE combined with effusions.
METHODS: Clinical data of patients diagnosed with NLE were retrospectively collected and analyzed from January 1, 2011, to December 31, 2023, at the Children\'s Hospital of Soochow University and Suzhou Municipal Hospital. Patients with NLE were divided into effusion and non-effusion groups based on the presence of effusion. Moreover, the clinical data of the newborns in both groups were compared and investigated.
RESULTS: Eleven (11/45, 24.44%) NLE patients had effusions, such as pleural effusion, testicular hydrocele, peritoneal effusion, pericardial effusion, and hydrocephalus. Other organs involved in effusion patients were cutaneous, gastrointestinal, hematologic, cardiac, and neurological. Among the patients with effusion, five cases of SLE in pregnant mothers, two cases of Sjogren\'s syndrome, one case of photoallergic symptoms, and three of pregnant mothers with no history of antenatal autoimmune disease. Pregnant mother\' autoimmune disease in remission prior to pregnancy, or stable low disease activity. Seven patients were positive for Anti-SSA, five of which were double positive for Anti-SSA and Anti-SSB. Compared with the non-effusion group, the effusion group patients had significantly higher lactate dehydrogenase, creatine kinase, and fibrinogen, significantly lower platelets, total protein, and albumin. These patients were likelier to have thrombocytopenia and coagulation abnormalities. Logistics regression analysis demonstrated that NLE patients with effusions are more likely to have decreased serum total protein levels. All NLE patients with effusion have self-resorption of the effusion.
CONCLUSIONS: 24.44% of patients had effusions in our study. NLE patients with effusion are more likely to have hematologic involvement and a more inflammatory response. The effusion in NLE patients is usually self-resorption, severe cases can be treated with nonsteroidal anti-inflammatory drugs/steroids. Key Points • NLE patients combined with effusions and were self-limiting, with pleural effusion being the most common. • NLE patients combined with effusions have a more inflammatory response, significant abnormalities in the blood routine and biochemical-related indexes.