Precision and Robust Models on Healthcare Institution Federated Learning for Predicting HCC on Portal Venous CT Images.

Hepatocellular carcinoma (HCC), the most common type of liver cancer, poses significant challenges in detection and diagnosis. Medical imaging, especially computed tomography (CT), is pivotal in non-invasively identifying this disease, requiring substantial expertise for interpretation. This research introduces an innovative strategy that integrates two-dimensional (2D) and three-dimensional (3D) deep learning models within a federated learning (FL) framework for precise segmentation of liver and tumor regions in medical images. The study utilized 131 CT scans from the Liver Tumor Segmentation (LiTS) challenge and demonstrated the superior efficiency and accuracy of the proposed Hybrid-ResUNet model with a Dice score of 0.9433 and an AUC of 0.9965 compared to ResNet and EfficientNet models. This FL approach is beneficial for conducting large-scale clinical trials while safeguarding patient privacy across healthcare settings. It facilitates active engagement in problem-solving, data collection, model development, and refinement. The study also addresses data imbalances in the FL context, showing resilience and highlighting local models' robust performance. Future research will concentrate on refining federated learning algorithms and their incorporation into the continuous implementation and deployment (CI/CD) processes in AI system operations, emphasizing the dynamic involvement of clients. We recommend a collaborative human-AI endeavor to enhance feature extraction and knowledge transfer. These improvements are intended to boost equitable and efficient data collaboration across various sectors in practical scenarios, offering a crucial guide for forthcoming research in medical AI.

Identification of new defective interfering RNA species associated with porcine reproductive and respiratory syndrome virus infection.

In the present study, seven new defective interfering (DI) RNA species of porcine reproductive and respiratory syndrome virus (PRRSV) were identified. RT-PCR, Northern blot and sequence analyses indicated that these DI RNA specie have deletions of 8513-9176 nucleotides located between Nsp1/Nsp2 and Nsp10. Compared with the previous DI RNAs of PRRSV reported, they have three distinct characteristics: much smaller deletion sizes; different nucleotide repeats (2-12nt) used in the junction sites and in-frame deletions. The results further suggested that the similarity-assisted RNA recombination may be the main cause of generation of DI RNAs in PRRSV and probably in other arteriviruses.

Successful medical treatment of spinal epidural abscess.

Spinal epidural abscess is a rare but potentially fatal disease. A 67-year-old female suffered fever and painful swelling of the right knee and lower leg for one week. Both synovial fluid and blood cultures yielded methicillin-sensitive Staphylococcus aureus. Low back pain developed and fever was sustained despite the administration of intravenous oxacillin. Magnetic resonance imaging (MRI) of the thoracolumbar spine revealed spinal epidural abscess from T12 to S1. Because of severe hypoalbuminemia and general anasarca and followed by exploratory laparotomy for massive duodenal bleeding, she did not receive surgical intervention for the spinal epidural abscess. After intravenous administration of oxacillin 2 g 4-hourly for 12 weeks, she recovered and follow-up MRI confirmed the efficacy of the medical treatment. She remained well at 1-year follow-up. In a patient with minimal neurological deficit or surgical contraindication, spinal epidural abscess can be successfully treated with a medical regimen.

Characterization and diagnostic accuracy of serous cystadenomas and mucinous neoplasms of the pancreas with multi-slice helical computed tomography / 中国医学科学院学报

<p><b>OBJECTIVE</b>To evaluate the feasibility of using multi-slice helical computed tomography (MS-CT) to accurately distinguish serous cystadenomas from mucinous cystadenomas or cystadenocarcinomas of the pancreas and to determine their radiographic appearances that can be applied for differentiative diagnosis.</p><p><b>METHODS</b>We performed a single-blind retrospective analysis of CT images of 30 patients with pathologically proven primary cystic pancreatic neoplasms (12 cases of serous cystadenomas, 14 cases of mucinous cystadenomas, and 4 cases of mucinous cystadenocarcinomas) to reach a diagnosis of either serous cystadenoma or mucinous cystic tumor. CT features such as tumor location, septations, presence of calcification, features of cystic wall, papillary excrescences, and size of the largest cyst were recorded. Statistical analysis was performed to evaluate the efficacy of certain CT findings in the differentiation of serous cystadenomas and mucinous neoplasms.</p><p><b>RESULTS</b>Totally 9 (75.0%) serous cystadenomas and 16 (88.9%) mucinous tumors were correctly diagnosed. Three serous cystadenomas were misdiagnosed as mucinous cystadenomas, while 2 mucinous neoplasms were misdiagnosed as serous cystadenomas. And 9 (75.0%) serous cystadenomas were located at the pancreatic head and neck areas, while 12 (66.7%) mucinous neoplasms were located at the pancreatic body and tail areas (P < 0.05). The presence of calcification, especially central calcification, had statistical significance in differentiating serous cystadenoma from mucinous neoplasms (P < 0.05). The size of the largest cyst over 2 cm was positive associated with mucinous neoplasms (P < 0.05).</p><p><b>CONCLUSION</b>CT characteristics between serous cystadenomas and mucinous neoplasms of the pancreas have distinct difference, which validates the values of CT in differentiating these tumors. However, atypical CT appearances may compromise its diagnostic accuracy.</p>