Uveal melanoma(UM)is the most common primary intraocular malignancy in adults, characterized by high invasiveness and unique metastatic biological features. Although local treatments(such as proton beam therapy and brachytherapy)can effectively control the primary lesion, approximately 50% of patients eventually develop distant metastasis, with the liver being the primary target organ(occurring in 90% of cases). This highlights a paradigm shift in treatment focus from mere local control to systemic prevention and management. For metastatic UM(mUM), current treatment strategies encompass biomarker-guided molecular targeted therapy, immunotherapy(including Tebentafusp, vaccines, and oncolytic virus therapy), and liver-directed therapy. Focusing on the synergy between local and systemic prevention and control, this article systematically elaborates on the precision local treatment for primary UM, the decision-making pathway for systemic treatment of metastatic UM based on molecular subtyping, the integration of local and systemic therapies for liver metastases, and the translational value of nanomedicine in addressing therapeutic bottlenecks. It provides insights for optimizing clinical management of mUM and developing novel therapeutic strategies.

Medical imaging technology plays a crucial role in clinical diagnosis and treatment. Image compression technology provides robust technical support for the storage and transmission of massive medical imaging data, serving as an effective safeguard for hospital data backup and telemedicine. The technology holds broad application prospects in the medical field, enabling the processing of various imaging modalities, multidimensional imaging, and medical video imaging. This study elaborates on general image and video compression algorithms, the application of compression algorithms in the medical field, and the performance metrics of medical image compression, thereby providing critical technical support for enhancing clinical diagnostic efficiency and data management security.

Objective:To investigate the longitudinal stability of the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and the Activity of Daily Living Scale (ADL).Methods:The longitudinal cognitive assessment results of 68 dementia patients admitted to the Dementia and Leukoencephalopathy Outpatient Clinic, Department of Neurology, Peking Union Medical College Hospital, from January 2021 to January 2024, were retrospectively analyzed, including the total and sub-items scores of the MMSE, MoCA, and ADL. Two different rules were applied to analyze the abnormality rates: rule 1, where the current test result being better than the previous one was considered an abnormality; rule 2, where the current test result being better than the previous average score was considered an abnormality (If a patient had only 2 cognitive assessments, rule 2 was considered the same as rule 1). Two rules were used to analyze the abnormality rates of the scales. The statistical analyses were repeated after excluding patients with possible anxiety and depression status.Results:In assessing the total score stability, MMSE showed the lowest abnormality rates [27.2% (31/114) under rule 1 and 29.8% (34/114) under rule 2], while MoCA had the highest abnormality rates [41.3% (26/63) and 46.0% (29/63), respectively]. The ADL abnormality rates were 27.7% (23/83) and 33.7% (28/83), respectively. Among MoCA sub-items, category cue, multiple choice cue, second memory trial, orientation, and clock showed higher abnormality rates [31.7%(20/63), 30.2%(19/63), 23.8%(15/63), 22.2%(14/63), 22.2%(14/63), respectively]. After excluding population with possible anxiety and depression status, the relative abnormality rates of MMSE and ADL sub-items did not significantly change, while the abnormality rate of orientation in MoCA sub-items decreased relatively.Conclusion:The MMSE and ADL exhibit good stability in long-term monitoring of dementia patients, serving as essential tools for assessing and following up cognitive changes.

Objective To achieve efficient compression of on-board radiotherapy projection images using dynamic video encoding algorithms.Methods The on-board radiotherapy imaging system primarily provides 2D X-ray projection images for patient positioning verification and 3D tomographic image reconstruction.Since multiple projection images acquired continuously exhibit strong spatiotemporal correlations,their similarities could be used to eliminate redundant information,thereby improving the image compression ratio.During image compression,the image sets obtained at different times were arranged into an image sequence which was input into a video encoder and output as a video file.During image decompression,the video file was input into a video decoder and output as an image sequence,and the images in the sequence were then assigned back to their original image sets.Three current dynamic video encoding algorithms(AVC,HEVC,and AV1)and the classic static image coding algorithm(JPEG 2000)were tested on a database of 2D projection images.The performance of various compression algorithms was evaluated using indicators such as compression ratio(CR),peak signal-to-noise ratio(PSNR),and structural similarity(SSIM).Moreover,visual comparison of projection images before and after compression was evaluated by clinical radiation oncologists.Results Dynamic video encoding algorithms achieved higher CR than the static image coding algorithm.The average CR of the 3 dynamic video encoding algorithms was as followed:CRAVC=11.50,CRHEVC=30.74,and CRAV1=27.10,while the average CR of the static image coding algorithm(JPEG 2000)was 5.28.For abdominal projection images,well-defined contours and textural details were preserved even when the CR reached 42.37.For head-neck projection images,although mild contour blurring emerged at a CR of 20.71,subsequent evaluation by clinical radiation oncologists confirmed that the reconstructed CBCT images still satisfied clinical requirements.Conclusion These dynamic video encoding algorithms effectively utilize the strong correlation information between multiple projection images,reduce the storage of redundant information,and greatly improve the image CR.

Objective To achieve efficient compression of on-board radiotherapy projection images using dynamic video encoding algorithms.Methods The on-board radiotherapy imaging system primarily provides 2D X-ray projection images for patient positioning verification and 3D tomographic image reconstruction.Since multiple projection images acquired continuously exhibit strong spatiotemporal correlations,their similarities could be used to eliminate redundant information,thereby improving the image compression ratio.During image compression,the image sets obtained at different times were arranged into an image sequence which was input into a video encoder and output as a video file.During image decompression,the video file was input into a video decoder and output as an image sequence,and the images in the sequence were then assigned back to their original image sets.Three current dynamic video encoding algorithms(AVC,HEVC,and AV1)and the classic static image coding algorithm(JPEG 2000)were tested on a database of 2D projection images.The performance of various compression algorithms was evaluated using indicators such as compression ratio(CR),peak signal-to-noise ratio(PSNR),and structural similarity(SSIM).Moreover,visual comparison of projection images before and after compression was evaluated by clinical radiation oncologists.Results Dynamic video encoding algorithms achieved higher CR than the static image coding algorithm.The average CR of the 3 dynamic video encoding algorithms was as followed:CRAVC=11.50,CRHEVC=30.74,and CRAV1=27.10,while the average CR of the static image coding algorithm(JPEG 2000)was 5.28.For abdominal projection images,well-defined contours and textural details were preserved even when the CR reached 42.37.For head-neck projection images,although mild contour blurring emerged at a CR of 20.71,subsequent evaluation by clinical radiation oncologists confirmed that the reconstructed CBCT images still satisfied clinical requirements.Conclusion These dynamic video encoding algorithms effectively utilize the strong correlation information between multiple projection images,reduce the storage of redundant information,and greatly improve the image CR.

Objective:To investigate the longitudinal stability of the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and the Activity of Daily Living Scale (ADL).Methods:The longitudinal cognitive assessment results of 68 dementia patients admitted to the Dementia and Leukoencephalopathy Outpatient Clinic, Department of Neurology, Peking Union Medical College Hospital, from January 2021 to January 2024, were retrospectively analyzed, including the total and sub-items scores of the MMSE, MoCA, and ADL. Two different rules were applied to analyze the abnormality rates: rule 1, where the current test result being better than the previous one was considered an abnormality; rule 2, where the current test result being better than the previous average score was considered an abnormality (If a patient had only 2 cognitive assessments, rule 2 was considered the same as rule 1). Two rules were used to analyze the abnormality rates of the scales. The statistical analyses were repeated after excluding patients with possible anxiety and depression status.Results:In assessing the total score stability, MMSE showed the lowest abnormality rates [27.2% (31/114) under rule 1 and 29.8% (34/114) under rule 2], while MoCA had the highest abnormality rates [41.3% (26/63) and 46.0% (29/63), respectively]. The ADL abnormality rates were 27.7% (23/83) and 33.7% (28/83), respectively. Among MoCA sub-items, category cue, multiple choice cue, second memory trial, orientation, and clock showed higher abnormality rates [31.7%(20/63), 30.2%(19/63), 23.8%(15/63), 22.2%(14/63), 22.2%(14/63), respectively]. After excluding population with possible anxiety and depression status, the relative abnormality rates of MMSE and ADL sub-items did not significantly change, while the abnormality rate of orientation in MoCA sub-items decreased relatively.Conclusion:The MMSE and ADL exhibit good stability in long-term monitoring of dementia patients, serving as essential tools for assessing and following up cognitive changes.

Objective:To analyze the clinical and imaging features of patients with COVID-19-related osmotic demyelination syndrome (ODS).Methods:COVID-19-related ODS cases diagnosed in the Department of Neurology, Peking Union Medical College Hospital from January 2020 to September 2023 were retrospectively reviewed. And their past medical history, possible triggers, clinical manifestations, imaging manifestations, treatment and prognosis were summarized.Results:A total of 5 patients with COVID-19-related ODS were included. Electrolyte disturbances acted as an inducement of ODS in all patients (5/5),4 of whom with hyponatremia. Four of 5 patients first presented with disturbance of consciousness, followed by predominant dystonia. Imaging of all patients (5/5) showed isolated extrapontine myelinolysis (EPM). With the prolongation of the course of disease, such signal intensity could return to normal, and lesions showed atrophic changes in some patients. The patients′ clinical symptoms were partly relieved within a few days to a few months after treatment.Conclusions:COVID-19-related ODS is mostly associated with hyponatremia, and EPM is more common. COVID-19 should be considered as a risk factor for ODS.

Objective To noninvasively predict isocitrate dehydrogenase(IDH)status of glioma via combining imaging and clini-cal features before surgery,so as to provide basis for individualized clinical treatment decision.Methods A total of 47 patients with glioma confirmed by pathological and molecular genetic tests were included,including 20 with IDH mutant type and 27 with IDH wild type.After diffusion tensor imaging(DTI)scanning,fractional anisotropy(FA)and mean diffusivity(MD)values of tumor paren-chyma were calculated.Combining DTI parameters with MRI morphological features of tumor,blood neutrophil/lymphocyte ratio(NLR)and patient's age,binary logistic regression model was established to effectively predict IDH status of glioma patients before surgery.Results There were significant differences in FAmean/FANAWM,MDmin,NLR,tumor location and age between IDH mutant type and IDH wild type groups(P<0.05).The binary logistic regression model concluding,FAmean/FANAWM,MDmin,cystic degeneration,NLR and age,predicted IDH status of glioma with area under the curve(AUC)of 0.961 and 95%confidence interval(CI)of 0.914-1.00.Conclusion The regression model established via combining DTI,MRI morphological features and blood NLR has great performance in classifying IDH status of glioma,and can help predict IDH status noninvasively before surgery,so as to assist clinical individualized treatment.

Objective:To clarify the effect and the mechanism of G protein-coupled receptor class C group 5 member A (GPRC5A) on lipopolysaccharide (LPS)-induced inflammatory response in human gingival fibroblasts (GFs), thus to provide a foundation for delving into the role of G protein coupled receptor (GPCR) in periodontitis.Methods:Gingival tissue samples were collected from 3 individuals periodontally healthy (health group) and 3 patients with periodontitis (periodontitis group) in Shandong Stomatological Hospital from December 2022 to February 2023. The expressions of GPRC5A of the two groups were detected by immunohistochemistry staining. GFs used in this study were isolated from a portion of gingiva for the extraction of impacted teeth in School and Hospital of Stomatology, Cheeloo College of Medicine, Shandong University from December 2022 to February 2023. GFs were isolated with enzymic digestion and transfected with 30, 50 and 80 μmol/L small interfering RNA-GPRC5A (siGPRC5A) or small interfering RNA-negative control (siNC), regarded as the experimental group and the negative control one, respectively. The silencing efficiency of siGPRC5A was evaluated by real-time fluorescence quantitative PCR (RT-qPCR). Experiments were then conducted using these cells which were divided into four groups of negative control (NC), LPS, siGPRC5A+LPS and siGPRC5A. The mRNA and protein levels of GPRC5A in GFs under 1 mg/L LPS-induced GFs inflammatory state were evaluated by RT-qPCR and Western blotting analysis after GPRC5A knockdown. RT-qPCR was used to detect the gene expression levels of the inflammatory cytokines in GFs induced by LPS, namely, interleukin (IL)-1β, IL-6, IL-8, tumor necrosis factor (TNF)-α, prostaglandin endoperoxide synthase 2 (PTGS2) after GPRC5A knockdown. Western blotting analysis and immunofluorescence staining were used to further investigate the activation of nuclear factor-kappa B (NF-κB) signaling pathway.Results:Immunohistochemistry staining showed that the expression of GPRC5A in gingival tissues of periodontitis group (0.132±0.006) increased compared with that in periodontally healthy group (0.036±0.019) ( t=8.24, P=0.001). Meanwhile, RT-qPCR results showed that the gene expression levels of GPRC5A at different time point (2, 6, 12, 24 h) in LPS-induced GFs (0.026±0.002, 0.042±0.005, 0.004±0.000, 0.016±0.000) were upregulated compared with those in the NC group (0.004±0.000, 0.004±0.000, 0.002±0.000, 0.007±0.000) (all P<0.001), respectively, and peaked at 6 h. The 50 μmol/L group displayed the most significant decrease in siGPRC5A expression (31.16±3.29) compared with that of the siNC group (100.00±4.88) ( F=297.98, P<0.001). The results of RT-qPCR and Western blotting analysis showed that siGPRC5A (0.27±0.03, 0.71±0.00) suppressed the expressions of GPRC5A at both gene and protein levels, while LPS (1.30±0.10, 1.43±0.03) was able to promote the expressions of GPRC5A compared with those of the NC group (1.00±0.01, 1.00±0.00)(all P<0.001). The siGPRC5A+LPS group (0.39±0.03, 1.06±0.16) also inhibited the increase of GPRC5A at both gene and protein levels induced by LPS (1.30±0.10, 1.43±0.03) ( F=208.38, P<0.001; F=42.04, P<0.001). RT-qPCR results showed that the expressions of IL-8, IL-1β, IL-6, TNF-α, and PTGS2 at the gene level in LPS group were highly increased compared with those in the NC group (all P<0.001). siGPRC5A significantly suppressed LPS-induced expressions of these inflammatory cytokines in GFs (all P<0.001). Western blotting analysis showed that the levels of p65 and IκBα protein phosphorylation in the LPS group were highly increased compared with those in the NC group, and siGPRC5A could effectively suppressed LPS-induced protein phosphorylation (all P<0.01). Furthermore, immunofluorescence staining showed that NF-κB p65 in the control group was mainly concentrated in the cytoplasm, and partially translocated to the nucleus under the stimulation of LPS. siGPRC5A was able to inhibit LPS-induced intranuclear translocation of p65 to a certain extent. Conclusions:GPRC5A expression was upregulated in periodontitis, and GPRC5A knockdown inhibited LPS-induced inflammation. Moreover, GPRC5A played a role in inflammation regulation by interacting with NF-κB signaling pathway.

Objective:To investigate the clinical significance of mycoplasm pneumoniae-DNA(MP-DNA) load detected in bronchoalveolar lavage fluid of children with refractory mycoplasma pneumoniae pneumonia(RMPP) in the assessment of the condition.Methods:A total number of 98 hospitalized children with RMPP diagnosed in the Department of Pediatric Respiratory of Shengjing Hospital of China Medical University from May to November in 2018 were selected as the subjects.Bronchoscopy was performed on each subject and the alveolar lavage fluid was collected to detect the MP-DNA load quantitatively.According to the test results, all children were divided into three groups: high-load group(>10 6/ml, 28 cases), medium-load group(10 3 to 10 6/ml, 54 cases), and low-load group(≤10 3/ml, 16 cases). The clinical manifestations, relevant laboratory data, imaging data, and bronchoscopy findings of the children in different load groups were compared. Results:Compared with the low-load group and medium-load groups, the level of C-reactive protein was the highest in the high-load group with statistical significance( P<0.05). The difference of lactate dehydrogenase levels among the three groups was statistically significant( P=0.001). The positive incidences of nodular protuberances, mucosal erosion and necrosis, bronchial inflammatory stenosis and mucus plugs occlusion in the high-load group were 35.7%, 32.1% and 64.3%, respectively, which were significantly higher than those in the medium-load group(13.0%, 5.6%, 38.95) and low-load group(0, 0, 12.5%), and the differences were statistically significant( P<0.05). The imaging manifestation of large areas consolidation/lung atelectasis and pleural effusion in the high-load group were 100% and 60.7%, respectively, which were significantly higher than those in the medium-load group(61.1%, 27.8%)and low-load group(25.0%, 12.5%), and the differences were statistically significant( P<0.05). Compared with the low-load group and medium-load group, the hospitalization days was the longest in the high-load group, and the fever clearance time after the application of macroclides was also the longest( P<0.05). Conclusion:The MP-DNA load of children with RMPP is closely related with the clinical features.The clinical manifestations of children with higher DNA-load are more serious.The level of local MP-DNA load can guide clinicians to use drugs rationally, better predicting the outcome of the disease.