Medical image analysis has witnessed remarkable advancements, even surpassing human-level performance in recent years, driven by the rapid development of advanced deep-learning algorithms. However, when the inference dataset slightly differs from what the model has seen during one-time training, the model performance is greatly compromised. The situation requires restarting the training process using both the old and the new data, which is computationally costly, does not align with the human learning process, and imposes storage constraints and privacy concerns. Alternatively, continual learning has emerged as a crucial approach for developing unified and sustainable deep models to deal with new classes, tasks, and the drifting nature of data in non-stationary environments for various application areas. Continual learning techniques enable models to adapt and accumulate knowledge over time, which is essential for maintaining performance on evolving datasets and novel tasks. Owing to its popularity and promising performance, it is an active and emerging research topic in the medical field and hence demands a survey and taxonomy to clarify the current research landscape of continual learning in medical image analysis. This systematic review paper provides a comprehensive overview of the state-of-the-art in continual learning techniques applied to medical image analysis. We present an extensive survey of existing research, covering topics including catastrophic forgetting, data drifts, stability, and plasticity requirements. Further, an in-depth discussion of key components of a continual learning framework, such as continual learning scenarios, techniques, evaluation schemes, and metrics, is provided. Continual learning techniques encompass various categories, including rehearsal, regularization, architectural, and hybrid strategies. We assess the popularity and applicability of continual learning categories in various medical sub-fields like radiology and histopathology. Our exploration considers unique challenges in the medical domain, including costly data annotation, temporal drift, and the crucial need for benchmarking datasets to ensure consistent model evaluation. The paper also addresses current challenges and looks ahead to potential future research directions.

How can we test AI performance? This question seems trivial, but it isn't. Standard benchmarks often have problems such as in-distribution and small-size test sets, oversimplified metrics, unfair comparisons, and short-term outcome pressure. As a consequence, good performance on standard benchmarks does not guarantee success in real-world scenarios. To address these problems, we present Touchstone, a large-scale collaborative segmentation benchmark of 9 types of abdominal organs. This benchmark is based on 5,195 training CT scans from 76 hospitals around the world and 5,903 testing CT scans from 11 additional hospitals. This diverse test set enhances the statistical significance of benchmark results and rigorously evaluates AI algorithms across various out-of-distribution scenarios. We invited 14 inventors of 19 AI algorithms to train their algorithms, while our team, as a third party, independently evaluated these algorithms on three test sets. In addition, we also evaluated pre-existing AI frameworks---which, differing from algorithms, are more flexible and can support different algorithms—including MONAI from NVIDIA, nnU-Net from DKFZ, and numerous other open-source frameworks. We are committed to expanding this benchmark to encourage more innovation of AI algorithms for the medical domain.

Anomaly detection has been a very challenging and active area of research for decades, particularly for video surveillance. However, most of the works detect predefined anomaly classes using static models. These frameworks have limited applicability for real-life surveillance where the data have concept drift. Under concept drift, the distribution of both normal and anomaly classes changes over time. An event may change its class from anomaly to normal or vice-versa. The non-adaptive frameworks do not handle this drift. Additionally, the focus has been on detecting local anomalies, such as a region of an image. In contrast, in CCTV-based monitoring, flagging unseen anomalous situations can be of greater interest. Utilizing multiple sensory information for anomaly detection has also received less attention. This extended abstract discusses these gaps and possible solutions.

PTZ cameras can be an effective replacement for multiple camera networks with their pan-tilt-zoom capability. However, the state of the art scheduling method for the PTZ cameras focuses mainly on tracking, not on coverage. In this paper, we aim to maximize coverage as well as information gain, thus, leading to effective surveillance. Towards this goal, we define an information map that represents the sensitivity of a region. We propose a scheduling algorithm in which the camera visits those states more often that are likely to be more important than others, thus, maximizing information gain. A probabilistic framework is used to maximize information gain and coverage simultaneously. Currently, there are no existing datasets and methods to evaluate PTZ camera scheduling methods. We build a real multi-camera dataset and develop a performance measure for this purpose. Experimental results show that the proposed stochastic scheduling algorithm based on adaptive information gain probability is better than traditional as well as other variants proposed in the paper in terms of information gain as well as coverage.

Drinking alcohol in excess leads to lower self-consciousness, damaging a persons judgment and thus enhances risk of aggressive behavior. It leads to various problems like social abuse, violence, crime, and road accidents. Hence, density of drunk people in a given area is one of the indicators of safety risk. In this work we propose a novel framework to determine density of drunk people in a smart city scenario. Smart cities provide multiple sources of information such as audio, video, and text (online social networks). We detect presence of drunk persons along with time and location by analyzing these information sources individually and then fuse this information to obtain a single drunk index for a given location. We put special focus text analysis and propose a more accurate method to detect drunk event (person) with an accuracy of 84.2%. Experimental results demonstrate the functionality and efficacy of the proposed framework.