1. A novel deep-learning algorithm called dual-view selective instance segmentation network (DVSISN) has been developed for segmenting unstained adherent cells in differential interference contrast (DIC) images.
2. The algorithm uses a dual-view segmentation (DVS) method with pairs of original and rotated images to predict the bounding box and its corresponding mask for each cell instance.
3. Experimental results demonstrate that the algorithm achieves an AP_segm of 0.555, which is 23.6% higher than a benchmark.
The article is generally reliable and trustworthy, as it provides detailed information about the development of a novel deep-learning algorithm for segmenting unstained adherent cells in DIC images, as well as experimental results demonstrating its effectiveness compared to a benchmark. However, there are some potential biases that should be noted. For example, the authors do not provide any information about how they selected their dataset or what criteria were used to evaluate the performance of their algorithm; this could lead to bias in favor of their own results if they chose datasets that would be more likely to produce favorable outcomes for their algorithm. Additionally, the authors do not discuss any potential risks associated with using their algorithm or any possible limitations; this could lead to readers overestimating the capabilities of the algorithm without considering potential drawbacks or limitations. Finally, while the authors provide evidence for their claims regarding the effectiveness of their algorithm compared to a benchmark, they do not explore any counterarguments or present both sides equally; this could lead readers to form an overly positive opinion about the efficacy of their approach without considering alternative perspectives or approaches.