Search for: All records

Incorporating learning based components in the current state-of-the-art cyber-physical systems (CPS) has been a challenge due to the brittleness of the underlying deep neural networks. On the bright side, if executed correctly with safety guarantees, this has the ability to revolutionize domains like autonomous systems, medicine, and other safety-critical domains. This is because it would allow system designers to use high-dimensional outputs from sensors like camera and LiDAR. The trepidation in deploying systems with vision and LiDAR components comes from incidents of catastrophic failures in the real world. Recent reports of self-driving cars running into difficult to handle scenarios is ingrained in the software components which handle such sensor inputs. The ability to handle such high-dimensional signals is due to the explosion of algorithms which use deep neural networks. Sadly, the reason behind the safety issues is also due to deep neural networks themselves. The pitfalls occur due to possible over-fitting and lack of awareness about the blind spots induced by the training distribution. Ideally, system designers would wish to cover as many scenarios during training as possible. However, achieving a meaningful coverage is impossible. This naturally leads to the following question: is it feasible to flag out-of-distribution (OOD) samples without causing too many false alarms? Such an OOD detector should be executable in a fashion that is computationally efficient. This is because OOD detectors often are executed as frequently as the sensors are sampled. Our aim in this article is to build an effective anomaly detector. To this end, we propose the idea of a memory bank to cache data samples which are representative enough to cover most of the in-distribution data. The similarity with respect to such samples can be a measure of familiarity of the test input. This is made possible by an appropriate choice of distance function tailored to the type of sensor we are interested in. Additionally, we adapt conformal anomaly detection framework to capture the distribution shifts with a guarantee of false alarm rate. We report the performance of our technique on two challenging scenarios: a self-driving car setting implemented inside the simulator CARLA with image inputs and autonomous racing car navigation setting with LiDAR inputs. From the experiments, it is clear that a deviation from the in-distribution setting can potentially lead to unsafe behavior. It should be noted that not all OOD inputs lead to precarious situations in practice, but staying in-distribution is akin to staying within a safety bubble and predictable behavior. An added benefit of our memory-based approach is that the OOD detector produces interpretable feedback for a human designer. This is of utmost importance since it recommends a potential fix for the situation as well. In other competing approaches, such feedback is difficult to obtain due to reliance on techniques which use variational autoencoders. 

Background Early diagnosis is essential for effective stroke therapy. Strokes in hospitalized patients are associated with worse outcomes compared with strokes in the community. We derived and validated an algorithm to identify strokes by monitoring upper limb movements in hospitalized patients. Methods and Results A prospective case–control study in hospitalized patients evaluated bilateral arm accelerometry from patients with acute stroke with lateralized weakness and controls without stroke. We derived a stroke classifier algorithm from 123 controls and 77 acute stroke cases and then validated the performance in a separate cohort of 167 controls and 33 acute strokes, measuring false alarm rates in nonstroke controls and time to detection in stroke cases. Faster detection time was associated with more false alarms. With a median false alarm rate among nonstroke controls of 3.6 (interquartile range [IQR], 2.1–5.0) alarms per patient per day, the median time to detection was 15.0 (IQR, 8.0–73.5) minutes. A median false alarm rate of 1.1 (IQR. 0–2.2) per patient per day was associated with a median time to stroke detection of 29.0 (IQR, 11.0–58.0) minutes. There were no differences in algorithm performance for subgroups dichotomized by age, sex, race, handedness, nondominant hemisphere involvement, intensive care unit versus ward, or daytime versus nighttime. Conclusions Arm movement data can be used to detect asymmetry indicative of stroke in hospitalized patients with a low false alarm rate. Additional studies are needed to demonstrate clinical usefulness. 

Classification of clinical alarms is at the heart of prioritization, suppression, integration, postponement, and other methods of mitigating alarm fatigue. Since these methods directly affect clinical care, alarm classifiers, such as intelligent suppression systems, need to be evaluated in terms of their sensitivity and specificity, which is typically calculated on a labeled dataset of alarms. Unfortunately, the collection and particularly labeling of such datasets requires substantial effort and time, thus deterring hospitals from investigating mitigations of alarm fatigue. This article develops a lightweight method for evaluating alarm classifiers without perfect alarm labels. The method relies on probabilistic labels obtained from data programming—a labeling paradigm based on combining noisy and cheap-to-obtain labeling heuristics. Based on these labels, the method produces confidence bounds for the sensitivity/specificity values from a hypothetical evaluation with manual labeling. Our experiments on five alarm datasets collected at Children’s Hospital of Philadelphia show that the proposed method provides accurate bounds on the classifier’s sensitivity/specificity, appropriately reflecting the uncertainty from noisy labeling and limited sample sizes. 

Providing reliable model uncertainty estimates is imperative to enabling robust decision making by autonomous agents and humans alike. While recently there have been significant advances in confidence calibration for trained models, examples with poor calibration persist in most calibrated models. Consequently, multiple techniques have been proposed that leverage label-invariant transformations of the input (i.e., an input manifold) to improve worst-case confidence calibration. However, manifold-based confidence calibration techniques generally do not scale and/or require expensive retraining when applied to models with large input spaces (e.g., ImageNet). In this paper, we present the recursive lossy label-invariant calibration (ReCal) technique that leverages label-invariant transformations of the input that induce a loss of discriminatory information to recursively group (and calibrate) inputs – without requiring model retraining. We show that ReCal outperforms other calibration methods on multiple datasets, especially, on large-scale datasets such as ImageNet.