Objective:To investigate the clinical and pathological characteristics, treatment and prognosis of endometriosis (EM)-associated ovarian mesonephric-like adenocarcinoma (MLA).Methods:Clinical and pathological data were collected from nine patients diagnosed with EM-associated ovarian MLA at the Obstetrics and Gynecology Hospital of Fudan University between January 2022 and December 2024. Histological slides were re-reviewed, immunohistochemical examination and molecular testing were performed, and patient follow-up was conducted.Results:(1) Clinical characteristics: the median age of the nine patients was 54 years (range: 38-69 years). All patients presented with a pelvic mass; five cases also reported abdominal pain. Tumor location included five cases in the right ovary, two in the left ovary, and two involving both ovaries. International Federation of Gynecology and Obstetrics (FIGO) staging showed 3 cases at stage Ⅰ, 4 at stage Ⅱ, and 2 at stage Ⅲ. (2) Pathological features: gross examination revealed mixed solid-cystic masses with solid areas appearing gray-white or yellow-brown; the median maximum tumor diameter was 9.0 cm (range: 2.6-13.0 cm). Microscopically, tumors exhibited various architectural patterns, including tubular, glandular, papillary, slit-like, sex cord-like, glomeruloid, and solid structures, with tubular and glandular patterns being most common. Tumor cells demonstrated mild to moderate nuclear atypia. Of the 11 tumor foci in the 9 cases, 8 showed coexistence of MLA with other tumor components, such as endometrioid carcinoma, borderline endometrioid or borderline seromucinous tumors. In 1 case of MLA mixed with a borderline endometrioid tumor, both components exhibited squamous metaplasia. Immunohistochemistry showed variable expression of GATA-binding protein 3, thyroid transcription factor-1, CD 10, and calretinin, with positive rates of 9/11, 8/11, 5/11, and 3/6, respectively. Two tumor foci (2/11) exhibited focal expression of estrogen receptor and progesterone receptor. All cases displayed wild-type p53 expression. Molecular testing via next-generation sequencing in five patients revealed pathogenic mutations in the KRAS gene (5/5), with 3 cases (3/5) harboring additional pathogenic mutations in other genes. (3) Treatment and prognosis: all patients underwent surgery, supplemented by chemotherapy and (or) targeted therapy. Five patients underwent comprehensive staging surgery, four received cytoreductive surgery, and one patient received targeted therapy. The median follow-up duration was 7 months (range: 2-27 months). Three patients (3/9) experienced recurrence, and no deaths were reported during the follow-up period. Conclusions:EM-associated ovarian MLA demonstrates diverse morphological patterns and frequently coexists with other tumor types. Accurate diagnosis relies on an integrated evaluation of histomorphology, immunohistochemistry, and molecular testing. The primary treatment for EM-associated ovarian MLA is surgery, followed by adjuvant chemotherapy. Patients harboring pathogenic KRAS p.G12C mutations may benefit from targeted therapies. Ovarian MLA is an aggressive tumor, prone to recurrence in the short term, and has a poor prognosis.

Objective To develop and validate a deep learning-based segmentation model for low-dose cervical spine X-ray,aiming to address the insufficient segmentation accuracy in low-dose protocols while balancing radiation protection and diagnostic accuracy.Methods A total of 1 363 patients cervical spine X-ray images data were collected.A dose-attenuation mathematical simulation sys-tem was constructed to generate 14 122 dynamic low-dose cervical spine images incorporating quantum noise,contrast degradation,and blur artifacts.A neural network model was developed for automated segmentation of low-dose cervical spine X-ray using this dataset.Results Within the dose range of 50%to 7.5%,the average reults of automatic segmentation by the neural network model and manual segmentation for each group were as follows:50%dose group,intersection over union(IoU)=0.98 vs 0.93(P=0.707)and Dice coefficient(Dice)=0.99 vs 0.96(P=0.749);10%dose group,IoU=0.97 vs 0.87(P=0.201)and Dice=0.99 vs 0.93(P=0.219);7.5%dose group,IoU=0.97 vs 0.67(P<0.01)and Dice=0.98 vs 0.80(P<0.01).Conclusion The developed deep learning model achieved robust cervical spine segmentation(IoU>0.96,Dice>0.98)below diagnostic dose thresholds[peak signal-to-noise ratio(PSNR)<38 dB,structural similarity index(SSIM)<0.65].Under ultra-low-dose conditions(PSNR=27.710 dB,SSIM=0.274),it demonstrated a 44.78%IoU improvement and 22.5%Dice improvement over manual segmentation.This model enables minimal radia-tion exposure while preserving diagnostic performance,confirming its theoretical feasibility for low-dose X-ray image analysis and clinical research potential.

PURPOSE: End-stage knee osteoarthritis (OA) patients are the primary candidates for total knee arthroplasty (TKA). However, most morphological refinements of TKA prosthesis are based on anatomical data from the knees of healthy individuals. This study aimed to determine whether differences exist in key bony morphological characteristics of the distal femur and proximal tibia between osteoarthritic knees and healthy knees. METHODS: This was a retrospective cross-sectional observational study with a case-control design. Patients who were aged ≥ 50 years, had no history of trauma, fracture, or surgery in the studied knee, and had no obvious knee flexion contracture were included in this study by CT scans. Patients who met the American College of Rheumatology clinical criteria for knee OA were included in the study group. Kellgren-Lawrence grade III or IV knees were studied (for bilateral cases, the more severely affected knee was chosen). Patients who presented with unilateral knee pain or trauma were included in the control group, with CT scans from the opposite (asymptomatic) knee used for analyzing. The studied knee had a Kellgren-Lawrence grade of 0 or I and showed no abnormalities upon physical examination. Archived knee CT scans from 160 patients were divided into 2 groups: the study group (80 moderate-to-severe OA knees) and the control group (80 healthy knees). After 3-dimensional reconstruction and virtual cutting using a CT workstation, 13 morphological parameters of the distal femur and proximal tibia were compared between the 2 groups using independent-samples t-tests. RESULTS: No significant group differences in the femoral anteroposterior dimension (p = 0.797), height of the lateral femoral condyle (p = 0.268), posterior condylar angle (p = 0.240), tibial anteroposterior dimension (p = 0.536), or tibial lateral anteroposterior dimension (p = 0.702) were observed. However, the femoral mediolateral dimension (p = 0.002), distal femoral aspect ratio (femoral mediolateral dimension/femoral anteroposterior dimension) (p < 0.001), height of the femoral trochlear groove (p < 0.001), height of the medial femoral condyle (p < 0.001), tibial mediolateral dimension (p = 0.001), proximal tibial aspect ratio (tibial mediolateral dimension/tibial anteroposterior dimension) (p = 0.004), tibial medial anteroposterior dimension (p = 0.005), and tibial asymmetry ratio (tibial medial anteroposterior dimension/tibial lateral anteroposterior dimension) (p = 0.006) were all significantly greater in the study group. CONCLUSION Knees with moderate-to-severe OA are significantly wider than healthy knees, and OA is a risk factor for increased tibial platform asymmetry. When refining the morphological parameters of TKA prostheses, the specific bony morphological characteristics of OA knees should be taken into account to reduce the potential risk of femoral or tibial component underhang and facilitate optimal balance between tibial component fit and rotational alignment.
Humans ; Osteoarthritis, Knee/pathology* ; Male ; Female ; Cross-Sectional Studies ; Retrospective Studies ; Arthroplasty, Replacement, Knee ; Middle Aged ; Aged ; Case-Control Studies ; Prosthesis Design ; Knee Prosthesis ; Femur/anatomy & histology* ; Tibia/anatomy & histology* ; Tomography, X-Ray Computed ; Knee Joint/diagnostic imaging*

Objective To develop and validate a deep learning-based segmentation model for low-dose cervical spine X-ray,aiming to address the insufficient segmentation accuracy in low-dose protocols while balancing radiation protection and diagnostic accuracy.Methods A total of 1 363 patients cervical spine X-ray images data were collected.A dose-attenuation mathematical simulation sys-tem was constructed to generate 14 122 dynamic low-dose cervical spine images incorporating quantum noise,contrast degradation,and blur artifacts.A neural network model was developed for automated segmentation of low-dose cervical spine X-ray using this dataset.Results Within the dose range of 50%to 7.5%,the average reults of automatic segmentation by the neural network model and manual segmentation for each group were as follows:50%dose group,intersection over union(IoU)=0.98 vs 0.93(P=0.707)and Dice coefficient(Dice)=0.99 vs 0.96(P=0.749);10%dose group,IoU=0.97 vs 0.87(P=0.201)and Dice=0.99 vs 0.93(P=0.219);7.5%dose group,IoU=0.97 vs 0.67(P<0.01)and Dice=0.98 vs 0.80(P<0.01).Conclusion The developed deep learning model achieved robust cervical spine segmentation(IoU>0.96,Dice>0.98)below diagnostic dose thresholds[peak signal-to-noise ratio(PSNR)<38 dB,structural similarity index(SSIM)<0.65].Under ultra-low-dose conditions(PSNR=27.710 dB,SSIM=0.274),it demonstrated a 44.78%IoU improvement and 22.5%Dice improvement over manual segmentation.This model enables minimal radia-tion exposure while preserving diagnostic performance,confirming its theoretical feasibility for low-dose X-ray image analysis and clinical research potential.

Objective:To investigate the clinical and pathological characteristics, treatment and prognosis of endometriosis (EM)-associated ovarian mesonephric-like adenocarcinoma (MLA).Methods:Clinical and pathological data were collected from nine patients diagnosed with EM-associated ovarian MLA at the Obstetrics and Gynecology Hospital of Fudan University between January 2022 and December 2024. Histological slides were re-reviewed, immunohistochemical examination and molecular testing were performed, and patient follow-up was conducted.Results:(1) Clinical characteristics: the median age of the nine patients was 54 years (range: 38-69 years). All patients presented with a pelvic mass; five cases also reported abdominal pain. Tumor location included five cases in the right ovary, two in the left ovary, and two involving both ovaries. International Federation of Gynecology and Obstetrics (FIGO) staging showed 3 cases at stage Ⅰ, 4 at stage Ⅱ, and 2 at stage Ⅲ. (2) Pathological features: gross examination revealed mixed solid-cystic masses with solid areas appearing gray-white or yellow-brown; the median maximum tumor diameter was 9.0 cm (range: 2.6-13.0 cm). Microscopically, tumors exhibited various architectural patterns, including tubular, glandular, papillary, slit-like, sex cord-like, glomeruloid, and solid structures, with tubular and glandular patterns being most common. Tumor cells demonstrated mild to moderate nuclear atypia. Of the 11 tumor foci in the 9 cases, 8 showed coexistence of MLA with other tumor components, such as endometrioid carcinoma, borderline endometrioid or borderline seromucinous tumors. In 1 case of MLA mixed with a borderline endometrioid tumor, both components exhibited squamous metaplasia. Immunohistochemistry showed variable expression of GATA-binding protein 3, thyroid transcription factor-1, CD 10, and calretinin, with positive rates of 9/11, 8/11, 5/11, and 3/6, respectively. Two tumor foci (2/11) exhibited focal expression of estrogen receptor and progesterone receptor. All cases displayed wild-type p53 expression. Molecular testing via next-generation sequencing in five patients revealed pathogenic mutations in the KRAS gene (5/5), with 3 cases (3/5) harboring additional pathogenic mutations in other genes. (3) Treatment and prognosis: all patients underwent surgery, supplemented by chemotherapy and (or) targeted therapy. Five patients underwent comprehensive staging surgery, four received cytoreductive surgery, and one patient received targeted therapy. The median follow-up duration was 7 months (range: 2-27 months). Three patients (3/9) experienced recurrence, and no deaths were reported during the follow-up period. Conclusions:EM-associated ovarian MLA demonstrates diverse morphological patterns and frequently coexists with other tumor types. Accurate diagnosis relies on an integrated evaluation of histomorphology, immunohistochemistry, and molecular testing. The primary treatment for EM-associated ovarian MLA is surgery, followed by adjuvant chemotherapy. Patients harboring pathogenic KRAS p.G12C mutations may benefit from targeted therapies. Ovarian MLA is an aggressive tumor, prone to recurrence in the short term, and has a poor prognosis.

ObjectiveTo explore the application value of CT pulmonary angiography (CTPA) in assessing the severity of acute pulmonary embolism (APE) and right heart function in rehabilitation patients. MethodsFrom January, 2013 to January, 2020, 133 inpatients (94 positive and 39 negative) who underwent CTPA examination in Beijing Bo'ai Hospital were involved. Positive patients were further divided into mild, moderate and severe groups based on the pulmonary artery obstruction index (PAOI). The clinical parameters and right heart function indicators were compared. Spearman correlation analysis was used to analyze the correlation between PAOI, and clinical parameters and right heart function indicators, and Logistic regression analysis was used to predict the risk factors of APE. ResultsThere was significant difference in lower extremity venous thrombosis, D-dimer, oxygen partial pressure, PAOI and left process of interventricular septum among four groups (H ≥ 12.350, P < 0.01). PAOI was moderately positively correlated with D-dimer (r = 0.443, P < 0.001) and left process of interventricular septum (r = 0.520, P < 0.001), and was weakly positively correlated with lower extremity venous thrombosis (r = 0.399, P < 0.001), left pulmonary artery diameter (r = 0.213, P = 0.014) and inferior vena cava regurgitation (r = 0.229, P = 0.008). Lower extremity venous thrombosis (OR = 7.708, P < 0.001) and left process of interventricular septum (OR = 3.641, P = 0.008) were independent risk factors for the onset of APE. The combination of the two indicators was effective for diagnosis of APE, and AUC was 0.795 (95% CI 0.715 to 0.874). ConclusionCTPA may be applied to evaluate the severity of APE and right heart function in rehabilitation patients.

ObjectiveTo apply resting-state functional magnetic resonance imaging in acute mild traumatic brain injury. MethodsFrom May, 2019 to May, 2021, 47 cases with acute mild traumatic brain injury in Beijing Bo'ai Hospital were selected as observation group, and 50 healthy people who visited for routine physical examination during the same period were as control group. They were scanned with resting-state functional magnetic resonance imaging to measure amplitude of low-frequency fluctuation (ALFF), and they were assessed with Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA) and Rivermead Behavioural Memory Test (RBMT). ResultsCompared with the control group, the scores of MMSE, MoCA and RBMT decreased in the observation group (t > 18.138, P < 0.001); while ALFF decreased in the brain areas of posterior cerebellar lobe, cerebellar tonsils, lower half moon lobules, right superior temporal gyrus, right middle temporal gyrus, right parietal lobe and right central posterior gyrus, etc; and ALFF increased in the brain areas of marginal lobe, cingulate gyrus, precuneus, left cerebellar, right superior temporal gyrus, right middle temporal gyrus, right superior frontal gyrus, right middle frontal gyrus, right inferior frontal gyrus, etc. ConclusionThere are disorders in multiple brain areas for patients after acute mild traumatic brain injury, which may associate to the cognitive impairment.

Objective To investigate the relationship between the expression of CENPF in NSCLC adenocarcinoma (LUAD) and the clinical prognosis of patients and its effect on the metastasis of lung adenocarcinoma cells. Methods The expression of CENPF in LUAD and its relationship with patient prognosis were analyzed by online bioinformatics. The expression of CENPF was verified by LUAD tissue microarray immunohistochemical staining. Kaplan-Meier analysis was performed to analyze the relationship between the expression of CENPF and the prognosis of patients with lung adenocarcinoma. Cox survival hazard ratio was used to analyze the factors affecting the survival of patients. Chi-square analysis was adopted to examine the relationship between CENPF expression and clinicopathological stage and grade of patients. The expression of CENPF in NCI-H2126 cells were knocked out by lentivirus, and then the proliferation, invasion, and migration abilities of the cells were detected. Changes in mRNA expression profiles after CENPF knockout were detected by RNA-seq. Bioinformatics analysis of downstream signaling pathways and the target genes of CENPF was also performed. Western blot was used to verify the target gene. Results CENPF was significantly upregulated in LUAD tumor tissue (P < 0.05) and significantly correlated with pathological stage (P=0.013). The higher expression of CENPF, the worse the prognosis of patients (P=0.01, P=0.027). After the expression was CENPF of knocked out, the cell proliferation, migration, and invasion abilities significantly reduced (P < 0.01). The expression of chemokine pathway genes in cells was enriched significantly (P < 0.001). ACKR3/CXCR7 and CDH2/N-cadherin were significantly downregulated, whereas CDH1/E-cadherin was significantly upregulated. After CENPF was knocked out, ACKR3/CXCR7 and N-cadherin were significantly downregulated, whereas E-cadherin significantly increased. Conclusion The expression of CENPF is negatively correlated with the clinical prognosis of patients with LUAD, and it promotes the occurrence of EMT by regulating the expression levels of N-cadherin and E-cadherin related to EMT through ACKR3/CXCR7.

Objective To diagnose crossed cerebellar diaschisis (CCD) after supratentorial cerebral infarction with intravoxel incoherent motion (IVIM) magnetic resonance imaging. Methods From May, 2016 to May, 2017, 120 patients with unilateral supratentorial cerebral infarction were divided as CCD positive and CCD negative, and investigated with IVIM. Results The cerebral infarction volume (CIV) was more in CCD positive patients than in CCD negative patients in acute and chronic phase (t>13.943, P<0.05). The apparent diffusion coefficients (ADC) was significantly different between affected and unaffected lateral cerebellar in CCD positive patients (t=11.413, P<0.05), and it was significantly different between CCD positive and negative patients in unaffected lateral cerebellar (t=10.026, P<0.05). Conclusion CIV and ADC can be the indexes of IVIM to evaluate the cerebral perfusion in patients with cerebral infarction.

Objective Annexin A3 (ANXA3) is a member of the annexin family,As the existing studies suggest,ANXA3 is closely related to tumor genesis,development,invasion,metastasis and prognosis.ANXA3 is down-regulated in prostate and kidney cancer,but it is up-regulated in breast cancer,liver cancer and other tumors.ANXA3 is related to tumor size,staging,lymphatic metastasis and prognosis.Silencing ANXA3 expression can not only inhibit the proliferation and invasion of colorectal cancer and hepatocellular carcinoma cells,but also restrain the migration of breast cancer cells.ANXA3 may also be involved in the regulation and maintenance of hepatocellular stem cells through HIF1a / Notch and JNK signaling pathways.The current studies have shown that ANXA3 can serve as a potential biological marker of tumor diagnosis,prediction of chemotherapy sensitivity,and provide a new target for oncotherapy.