Producing labels for unlabeled data is error-prone, making semi-supervised learning (SSL) troublesome. Often, little is known about when and why an algorithm fails to outperform a supervised baseline. Using benchmark datasets, we craft five common real-world SSL data scenarios: few-label, open-set, noisy-label, and class distribution imbalance/misalignment in the labeled and unlabeled sets. We propose a novel algorithm called Contrastive Credibility Propagation (CCP) for deep SSL via iterative transductive pseudo-label refinement. CCP unifies semi-supervised learning and noisy label learning for the goal of reliably outperforming a supervised baseline in any data scenario. Compared to prior methods which focus on a subset of scenarios, CCP uniquely outperforms the supervised baseline in all scenarios, supporting practitioners when the qualities of labeled or unlabeled data are unknown.
Background: Pancreatic perivascular epithelioid cell tumors (PEComas) are exceedingly uncommon mesenchymal tumors. Herein we have mainly focused on the radiologic features of pancreatic PEComas to achieve a better knowledge of the disease. Case report: We present the case of a 41-year-old woman who had a benign pancreatic PEComa that was misinterpreted as a malignant tumor based on a preoperative CT scan before undergoing surgery for a hysteromyoma. Laboratory tests revealed no significant abnormal findings except an elevated CA-125 (428 U/L). CT and MRI images showed a solid, well-defined, hypovascular nodule approximately 10 mm in diameter in the pancreatic uncinate process that was hyperintense on diffusion-weighted imaging (DWI) and hypointense on apparent diffusion coefficient (ADC) maps. Because of a suspected malignant pancreatic neoplasm, she underwent a pancreaticoduodenectomy. The pathologic examination confirmed the diagnosis of a pancreatic PEComa. Conclusions: Because of the rarity of pancreatic PEComas, we highlighted the radiologic characteristics to facilitate the diagnosis and therapy.