Using an artificial intelligence model to detect and localize visible clinically significant prostate cancer in prostate magnetic resonance imaging: a multicenter external validation study.

Background: An increasing number of patients with suspected clinically significant prostate cancer (csPCa) are undergoing prostate multiparametric magnetic resonance imaging (mpMRI). The role of artificial intelligence (AI) algorithms in interpreting prostate mpMRI needs to be tested with multicenter external data. This study aimed to investigate the diagnostic efficacy of an AI model in detecting and localizing visible csPCa on mpMRI a multicenter external data set. Methods: The data of 2,105 patients suspected of having prostate cancer from four hospitals were retrospectively collected to develop an AI model to detect and localize suspicious csPCa. The lesions were annotated based on pathology records by two radiologists. Diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) values were used as the input for the three-dimensional U-Net framework. Subsequently, the model was validated using an external data set comprising the data of 557 patients from three hospitals. Sensitivity, specificity, and accuracy were employed to evaluate the diagnostic efficacy of the model. Results: At the lesion level, the model had a sensitivity of 0.654. At the overall sextant level, the model had a sensitivity, specificity, and accuracy of 0.846, 0.884, and 0.874, respectively. At the patient level, the model had a sensitivity, specificity, and accuracy of 0.943, 0.776, and 0.849, respectively. The AI-predicted accuracy for the csPCa patients (231/245, 0.943) was significantly higher than that for the non-csPCa patients (242/312, 0.776) (P<0.001). The lesion number and tumor volume were greater in the correctly diagnosed patients than the incorrectly diagnosed patients (both P<0.001). Among the positive patients, those with lower average ADC values had a higher rate of correct diagnosis than those with higher average ADC values (P=0.01). Conclusions: The AI model exhibited acceptable accuracy in detecting and localizing visible csPCa at the patient and sextant levels. However, further improvements need to be made to enhance the sensitivity of the model at the lesion level.

Clinical application of super-low-positioned intestinal decompression tube in the treatment of intestinal obstruction: a pilot study.

OBJECTIVE: This study aimed to evaluate the clinical application and efficacy of a super-low-positioned intestinal decompression tube in the treatment of intestinal obstruction. METHODS: A total of 130 patients with postoperative small bowel obstruction were included in this study. The patients were divided into a super-low-positioned intestinal decompression group and a conventional intestinal decompression group. The clinical data, treatment outcomes, and complications were compared between the two groups. RESULTS: The technical success rate of placing the super-low-positioned intestinal decompression tube was 100%, with no intraoperative complications. The patients in the super-low-positioned intestinal decompression group had a significantly shorter hospital stay (8.3 ± 5.2 vs 17.7 ± 13.3, P < 0.001) and a higher non-operative treatment success rate (83.6% vs 57.9%, P = 0.001) compared to the conventional intestinal decompression group. Multivariate logistic regression analysis showed that the placement of a super-low-positioned intestinal decompression tube was an independent protective factor for treatment outcomes (P = 0.001). The hospital stay was significantly shorter in the super-low-positioned intestinal decompression group compared to the conventional group in both successful non-operative treatment patients (6.9 ± 3.0 vs 11.2 ± 7.5, P < 0.001) and failed non-operative treatment patients (16.2 ± 7.4 vs 26.6 ± 14.4, P < 0.001). The super-low-positioned intestinal decompression tube effectively relieved the "Self-strangulation" phenomenon in patients with intestinal obstruction. CONCLUSION: The super-low-positioned intestinal decompression tube is a safe and effective method for the treatment of intestinal obstruction, with better treatment outcomes and shorter hospital stays compared to conventional intestinal decompression. Further prospective studies are needed to validate these findings.

Molybdenum target mammography-based prediction model for metastasis of axillary sentinel lymph node in early-stage breast cancer.

Sentinel lymph node (SLN) status is closely related to axillary lymph node metastasis in breast cancer. However, SLN biopsy has certain limitations due to invasiveness and diagnostic efficiency. This study aimed to develop a model to predict the risk of axillary SLN metastasis in early-stage breast cancer based on mammography, a noninvasive, cost-effective, and potential complementary way. Herein, 649 patients with early-stage breast cancer (cT1-T2) who received SLN biopsy were assigned to the training cohort (n = 487) and the validation cohort (n = 162). A prediction model based on specific characteristics of tumor mass in mammography was developed and validated with R software. The performance of model was evaluated by receiver operating characteristic curve, calibration plot, and decision curve analysis. Tumor margins, spicular structures, calcification, and tumor size were independent predictors of SLN metastasis (all P < .05). A nomogram showed a satisfactory performance with an AUC of 0.829 (95% CI = 0.792-0.865) in the training cohort and an AUC of 0.825 (95% CI = 0.763-0.888) in validation cohort. The consistency between model-predicted results and actual observations showed great Hosmer-Lemeshow goodness-of-fit (P = .104). Patients could benefit from clinical decisions guided by the present model within the threshold probabilities of 6% to 84%. The prediction model for axillary SLN metastasis showed satisfactory discrimination, calibration abilities, and wide clinical practicability. These findings suggest that our prediction model based on mammography characteristics is a reliable tool for predicting SLN metastasis in patients with early-stage breast cancer.

Short-term outcomes of a new gastrointestinal decompression tube combined with conservative treatment in patients with esophagojejunal anastomotic leakage after total gastrectomy.

To compare the short-term outcomes of a new gastrointestinal decompression tube combined with conservative treatment in patients with esophagojejunal anastomotic leakage (EJAL) after total gastrectomy. We retrospectively analyzed the data of 81 patients with EJAL who had undergone total gastrectomy and Roux-en-Y reconstruction at Fujian Medical University Union Hospital between January 2014 and December 2021. The patients were divided into experimental (12 patients with new gastrointestinal decompression tube plus conservative treatment) and control (69 patients with conservative treatment) groups, according to the different treatment methods they received. Anatomic defect size linearly correlated with time to clinical success, hospital stay, and hospital cost in the control group. The two groups showed no significant differences in anastomotic defect size, time of defect after surgery, hospitalization cost, and time of antibiotic use. However, the time to clinical success was significantly shorter in the experimental group than in the control group (16.0 ± 8.3 vs. 23.6 ± 17.8, P = 0.04), as was the length of hospital stay (30.1 ± 6.3 vs. 36.8 ± 16.7, P = 0.017). Furthermore, when the defect size was ≥ 4 mm, the time to clinical success, hospital stay, and hospital cost in the experimental group were lower than those in the control group (P < 0.05). Placement of a new gastrointestinal decompression tube is a safe treatment. When the defect size is ≥ 4 mm, the time to clinical success, length of hospital stay, and hospital cost can be reduced.

Altered Topological Properties of Gray Matter Structural Covariance Networks in Minimal Hepatic Encephalopathy.

Background and Aims: Liver cirrhosis commonly induces brain structural impairments that are associated with neurological complications (e.g., minimal hepatic encephalopathy (MHE)), but the topological characteristics of the brain structural network are still less well understood in cirrhotic patients with MHE. This study aimed to conduct the first investigation on the topological alterations of brain structural covariance networks in MHE. Methods: This study included 22 healthy controls (HCs) and 22 cirrhotic patients with MHE. We calculated the gray matter volume of 90 brain regions using an automated anatomical labeling (AAL) template, followed by construction of gray matter structural covariance networks by thresholding interregional structural correlation matrices as well as graph theoretical analysis. Results: MHE patients showed abnormal small-world properties of the brain structural covariance network, i.e., decreased clustering coefficient and characteristic path length and lower small-worldness parameters, which indicated a tendency toward more random architecture. In addition, MHE patients lost hubs in the prefrontal and parietal regions, although they had new hubs in the temporal and occipital regions. Compared to HC, MHE patients had decreased regional degree/betweenness involving several regions, primarily the prefrontal and parietal lobes, motor region, insula and thalamus. In addition, the MHE group also showed increased degree/betweenness in the occipital lobe and hippocampus. Conclusion: These results suggest that MHE leads to altered coordination patterns of gray matter morphology and provide structural evidence supporting the idea that MHE is a neurological complication related to disrupted neural networks.