The segment-anything model (SAM), was introduced as a fundamental model for segmenting images. It was trained using over 1 billion masks from 11 million natural images. The model can perform zero-shot segmentation of images by using various prompts such as masks, boxes, and points. In this report, we explored (1) the accuracy of SAM on 12 public medical image segmentation datasets which cover various organs (brain, breast, chest, lung, skin, liver, bowel, pancreas, and prostate), image modalities (2D X-ray, histology, endoscropy, and 3D MRI and CT), and health conditions (normal, lesioned). (2) if the computer vision foundational segmentation model SAM can provide promising research directions for medical image segmentation. We found that SAM without re-training on medical images does not perform as accurately as U-Net or other deep learning models trained on medical images.
The combination of the U-Net based deep learning models and Transformer is a new trend for medical image segmentation. U-Net can extract the detailed local semantic and texture information and Transformer can learn the long-rang dependencies among pixels in the input image. However, directly adapting the Transformer for segmentation has ``token-flatten" problem (flattens the local patches into 1D tokens which losses the interaction among pixels within local patches) and ``scale-sensitivity" problem (uses a fixed scale to split the input image into local patches). Compared to directly combining U-Net and Transformer, we propose a new global-local fashion combination of U-Net and Transformer, named U-Netmer, to solve the two problems. The proposed U-Netmer splits an input image into local patches. The global-context information among local patches is learnt by the self-attention mechanism in Transformer and U-Net segments each local patch instead of flattening into tokens to solve the `token-flatten" problem. The U-Netmer can segment the input image with different patch sizes with the identical structure and the same parameter. Thus, the U-Netmer can be trained with different patch sizes to solve the ``scale-sensitivity" problem. We conduct extensive experiments in 7 public datasets on 7 organs (brain, heart, breast, lung, polyp, pancreas and prostate) and 4 imaging modalities (MRI, CT, ultrasound, and endoscopy) to show that the proposed U-Netmer can be generally applied to improve accuracy of medical image segmentation. These experimental results show that U-Netmer provides state-of-the-art performance compared to baselines and other models. In addition, the discrepancy among the outputs of U-Netmer with different scales is linearly correlated to the segmentation accuracy which can be considered as a confidence score to rank test images by difficulty without ground-truth.
The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.