ABSTRACT
The density of mitotic figures (MF) within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of MF by pathologists is subject to a strong inter-rater bias, limiting its prognostic value. State-of-the-art deep learning methods can support experts but have been observed to strongly deteriorate when applied in a different clinical environment. The variability caused by using different whole slide scanners has been identified as one decisive component in the underlying domain shift. The goal of the MICCAI MIDOG 2021 challenge was the creation of scanner-agnostic MF detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were provided. In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance. The winning algorithm yielded an F1 score of 0.748 (CI95: 0.704-0.781), exceeding the performance of six experts on the same task.
Subject(s)
Algorithms , Mitosis , Humans , Neoplasm Grading , PrognosisABSTRACT
Amino acids (AAs) are important nitrogen-containing organics in water, and a large number of reports have proven that they were the precursors of many nitrogen-containing disinfection by-products, some of which have cytotoxicity and carcinogenicity. However, little has been done on their occurrence in drinking source water. Therefore, a trace determination method via solid-phase extraction coupled with ultra-high pressure liquid chromatography tandem mass spectrometry (UPLC-MS/MS) for 15 free AAs (FAAs) was developed, which was successfully applied for drinking source water samples. For sample preparation, strong cation-exchange stationary solid-phase extraction (SPE) cartridge showed better extraction performance to that of reverse phase stationary oasis HLB SPE cartridge. The optimal water pH was determined to be 2.8 before extraction. Strong matrix effects for most FAAs were observed in this work; thus, sample extraction with SPE was recommended to eliminate the matrix effects. The developed method showed excellent linearity (R2 > 0.991), low limits of detection (LODs, 0.01-0.27 nmol/L), and good recoveries of 69.8-117.9% in drinking source water with low relative standard deviations (RSDs, 0.3-13.2%). The developed method was finally applied to eight drinking source water samples, and the top five FAAs were found to be serine, glycine, leucine, alanine, and isoleucine.
