Gaetano Scebba

Machine learning models have become key in decision-making for many disciplines, including drug discovery and medicinal chemistry. Therefore, uncertainty quantification (UQ) in ML predictions has gained importance in recent years. Many investigations have focused on developing methodologies that provide accurate uncertainty estimates for ML-based predictions. Unfortunately, there is no UQ strategy that consistently provides robust estimates about model’s applicability on new samples. Depending on the dataset, prediction task, and algorithm, accurate uncertainty estimations might be unfeasible to obtain. The UNIQUE (UNcertaInty QUantification bEnchmarking) framework is introduced to facilitate a comparison of UQ strategies in ML-based predictions. This Python library unifies the benchmarking of multiple UQ metrics, including the calculation of nonstandard UQ metrics (combining information from the dataset and model), and provides a comprehensive evaluation.

Gait is a complex everyday activity that depends upon supraspinal activity and a host of cognitive functions such as attention and executive functions. As cognition declines in neurodegenerative diseases, the interaction and competition for neuronal resources during motor-cognitive dual-tasking (e.g., walking while talking) might be a sensitive measure of subtle functional impairments in early Alzheimer’s disease (AD). Here, we aim to identify gait deficits due to neuronal competition across the AD spectrum.

Chronic wounds significantly impact quality of life. They can rapidly deteriorate and require close monitoring of healing progress. Image-based wound analysis is a way of objectively assessing the wound status by quantifying important features that are related to healing. However, high heterogeneity of the wound types and imaging conditions challenge the robust segmentation of wound images. We present Detect-and-Segment (DS), a deep learning approach to produce wound segmentation maps with high generalization capabilities. In our approach, dedicated deep neural networks detected the wound position, isolated the wound from the perturbing background, and computed a wound segmentation map. We tested this approach on a diabetic foot ulcers data set and compared it to a segmentation method based on the full image. The Matthews correlation coefficient (MCC) improved from 0.29 (full image) to 0.85 (DS) on the diabetic foot ulcer data set. When the DS was tested on the independent data sets, the mean MCC increased from 0.17 to 0.85. Furthermore, the DS enabled the training of segmentation models with up to 90% less training data without impacting the segmentation performance. The proposed DS approach is a step towards automating wound analysis and reducing efforts to manage chronic wounds.

Continuous monitoring of respiratory activity is desirable in many clinical applications to detect respiratory events. Non-contact monitoring of respiration can be achieved with near- and far-infrared spectrum cameras. However, current technologies are not sufficiently robust to be used in clinical applications. For example, they fail to estimate an accurate respiratory rate (RR) during apnea. We present a novel algorithm based on multispectral data fusion that aims at estimating RR also during apnea. The algorithm independently addresses the RR estimation and apnea detection tasks. Respiratory information is extracted from multiple sources and fed into an RR estimator and an apnea detector whose results are fused into a final respiratory activity estimation. We evaluated the system retrospectively using data from 30 healthy adults who performed diverse controlled breathing tasks while lying supine in a dark room and reproduced central and obstructive apneic events. Combining multiple respiratory information from multispectral cameras improved the root mean square error (RMSE) accuracy of the RR estimation from up to 4.64 monospectral data down to 1.60 breaths/min. The median F1 scores for classifying obstructive (0.75 to 0.86) and central apnea (0.75 to 0.93) also improved. Furthermore, the independent consideration of apnea detection led to a more robust system (RMSE of 4.44 vs. 7.96 breaths/min). Our findings may represent a step towards the use of cameras for vital sign monitoring in medical applications.