Robust in-vehicle respiratory rate detection using multimodal signal fusion.

Continuous health monitoring in private spaces such as the car is not yet fully exploited to detect diseases in an early stage. Therefore, we develop a redundant health monitoring sensor system and signal fusion approaches to determine the respiratory rate during driving. To recognise the breathing movements, we use a piezoelectric sensor, two accelerometers attached to the seat and the seat belt, and a camera behind the windscreen. We record data from 15 subjects during three driving scenarios (15 min each) city, highway, and countryside. An additional chest belt provides the ground truth. We compare the four convolutional neural network (CNN)-based fusion approaches: early, sensor-based late, signal-based late, and hybrid fusion. We evaluate the performance of fusing for all four signals to determine the portion of driving time and the signal combination. The hybrid algorithm fusing all four signals is most effective in detecting respiratory rates in the city ([Formula: see text]), highway ([Formula: see text]), and countryside ([Formula: see text]). In summary, 60% of the total driving time can be used to measure the respiratory rate. The number of signals used in the multi-signal fusion improves reliability and enables continuous health monitoring in a driving vehicle.

Robust in-vehicle heartbeat detection using multimodal signal fusion.

A medical check-up during driving enables the early detection of diseases. Heartbeat irregularities indicate possible cardiovascular diseases, which can be determined with continuous health monitoring. Therefore, we develop a redundant sensor system based on electrocardiography (ECG) and photoplethysmography (PPG) sensors attached to the steering wheel, a red, green, and blue (RGB) camera behind the steering wheel. For the video, we integrate the face recognition engine SeetaFace to detect landmarks of face segments continuously. Based on the green channel, we derive colour changes and, subsequently, the heartbeat. We record the ECG, PPG, video, and reference ECG with body electrodes of 19 volunteers during different driving scenarios, each lasting 15 min: city, highway, and countryside. We combine early, signal-based late, and sensor-based late fusion with a hybrid convolutional neural network (CNN) and integrated majority voting to deliver the final heartbeats that we compare to the reference ECG. Based on the measured and the reference heartbeat positions, the usable time was 51.75%, 58.62%, and 55.96% for the driving scenarios city, highway, and countryside, respectively, with the hybrid algorithm and combination of ECG and PPG. In conclusion, the findings suggest that approximately half the driving time can be utilised for in-vehicle heartbeat monitoring.

Focalizing regions of biomarker relevance facilitates biomarker prediction on histopathological images.

Image-based AI has thrived as a potentially revolutionary tool for predicting molecular biomarker statuses, which aids in categorizing patients for appropriate medical treatments. However, many methods using hematoxylin and eosin-stained (H&E) whole-slide images (WSIs) have been found to be inefficient because of the presence of numerous uninformative or irrelevant image patches. In this study, we introduced the region of biomarker relevance (ROB) concept to identify the morphological areas most closely associated with biomarkers for accurate status prediction. We actualized this concept within a framework called saliency ROB search (SRS) to enable efficient and effective predictions. By evaluating various lung adenocarcinoma (LUAD) biomarkers, we showcased the superior performance of SRS compared to current state-of-the-art AI approaches. These findings suggest that AI tools, built on the ROB concept, can achieve enhanced molecular biomarker prediction accuracy from pathological images.

Scientific discovery in the age of artificial intelligence.

Artificial intelligence (AI) is being increasingly integrated into scientific discovery to augment and accelerate research, helping scientists to generate hypotheses, design experiments, collect and interpret large datasets, and gain insights that might not have been possible using traditional scientific methods alone. Here we examine breakthroughs over the past decade that include self-supervised learning, which allows models to be trained on vast amounts of unlabelled data, and geometric deep learning, which leverages knowledge about the structure of scientific data to enhance model accuracy and efficiency. Generative AI methods can create designs, such as small-molecule drugs and proteins, by analysing diverse data modalities, including images and sequences. We discuss how these methods can help scientists throughout the scientific process and the central issues that remain despite such advances. Both developers and users of AI toolsneed a better understanding of when such approaches need improvement, and challenges posed by poor data quality and stewardship remain. These issues cut across scientific disciplines and require developing foundational algorithmic approaches that can contribute to scientific understanding or acquire it autonomously, making them critical areas of focus for AI innovation.

Investigating impacts of COVID-19 on urban mobility and emissions.

The COVID-19 pandemic has severely impacted human activities in a way never documented in modern history. The prevention policies and measures have abruptly changed well-established urban mobility patterns. In this context, we exploit different sources of urban mobility data to gain insights into the effects of restrictive policies on the daily mobility and exhaust emissions in pandemic and post-pandemic periods. Manhattan, the most densely populated borough in New York City, is chosen as the study area. We collect data generated by taxis, sharing bikes, and road detectors between 2019 and 2021, and estimate exhaust emissions using the COPERT (Computer Programme to calculate Emissions from Road Transport) model. A comparative analysis is conducted to identify important changes in urban mobility and emission patterns, with a particular focus on the lockdown period in 2020 and its counterparts in 2019 and 2021. The results of the paper fuel the discussion on urban resilience and policy-making in a post pandemic world.

Erratum: Author Correction: Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence.

[This corrects the article DOI: 10.1038/s42256-021-00421-z.].

Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence.

Artificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses. However, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalised model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the AI model can be distributedly trained and independently executed at each host institution under a federated learning framework (FL) without data sharing. Here we show that our FL model outperformed all the local models by a large yield (test sensitivity /specificity in China: 0.973/0.951, in the UK: 0.730/0.942), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals leaving out the FL) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans (CTs) from 3,336 patients collected from 23 hospitals located in China and the UK. Collectively, our work advanced the prospects of utilising federated learning for privacy-preserving AI in digital health.

Evaluating the Epley Canalolith Repositioning Procedure With and Without a Visual Assistive Device.

HYPOTHESIS: The primary goal of this study was to examine how accuracy is affected when we employ a guidance device to assist with the execution of the Epley canalolith repositioning procedure. BACKGROUND: Benign paroxysmal positional vertigo is a common cause of vestibular vertigo. Treatment is noninvasive and generally effective when performed correctly. Deficiencies in clinical application result in unnecessary failures in response for those affected. METHODS: Ten participants were each taken through six iterations of the Epley canalolith repositioning procedure. Iterations were divided evenly between those conducted with and without the use of a guidance device. One clinician performed all 60 procedures. Head movements were recorded using motion capture cameras and strategically placed motion tracking markers. RESULTS: Results showed that the guidance device significantly improved the latter phase maneuver accuracy. Rotation error was significantly reduced for hold3 with-device (M = 20.23°, SD = 12.08°) versus without-device (M = 40.13°, SD = 14.62°, p  =  0.001). Maximal rotation error during rotation4 of the maneuver demonstrated a similar reduction of error with-device (M = 24.44°, SD = 10.43°) versus without-device (M = 41.36°, SD = 12.89°, p  =  0.002). CONCLUSION: A simple visual guidance device can increase the execution accuracy of canalith repositioning procedures. Further research is required to show how such improvements influence treatment efficacy.

Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence.

Artificial intelligence provides a promising solution for streamlining COVID-19 diagnoses; however, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalized model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the artificial intelligence (AI) model can be distributedly trained and independently executed at each host institution under a federated learning framework without data sharing. Here we show that our federated learning framework model considerably outperformed all of the local models (with a test sensitivity/specificity of 0.973/0.951 in China and 0.730/0.942 in the United Kingdom), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals without the federated learning framework) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans from 3,336 patients collected from 23 hospitals located in China and the United Kingdom. Collectively, our work advanced the prospects of utilizing federated learning for privacy-preserving AI in digital health.

'There were more wires than him': the potential for wireless patient monitoring in neonatal intensive care.

BACKGROUND: The neonatal intensive care unit (NICU) can be one of the most stressful hospital environments. Alongside providing intensive clinical care, it is important that parents have the opportunity for regular physical contact with their babies because the neonatal period is critical for parent-child bonding. At present, monitoring technology in the NICU requires multiple wired sensors to track each baby's vital signs. This study describes the experiences that parents and nurses have with the current monitoring methods, and reports on their responses to the concept of a wireless monitoring system. METHODS: Semistructured interviews were conducted with six parents, each of whom had babies on the unit, and seven nurses who cared for those babies. The interviews initially focused on the participants' experiences of the current wired system and then on their responses to the concept of a wireless system. The transcripts were analysed using a general inductive approach to identify relevant themes. RESULTS: Participants reported on physical and psychological barriers to parental care, the ways in which the current system obstructed the efficient delivery of clinical care and the perceived benefits and risks of a wireless system. The parents and nurses identified that the wires impeded baby-parent bonding; physically and psychologically. While a wireless system was viewed as potentially enabling greater interaction, staff and parents highlighted potential concerns, including the size, weight and battery life of any new device. CONCLUSIONS: The many wires required to safely monitor babies within the NICU creates a negative environment for parents at a critical developmental period, in terms of physical and psychological interactions. Nurses also experience challenges with the existing system, which could negatively impact the clinical care delivery. Developing a wireless system could overcome these barriers, but there remain challenges in designing a device suitable for this unique environment.

New least squares solutions for estimating the average centre of rotation and the axis of rotation.

A new method is proposed for estimating the parameters of ball joints, also known as spherical or revolute joints and hinge joints with a fixed axis of rotation. The method does not require manual adjustment of any optimisation parameters and produces closed form solutions. It is a least squares solution using the whole 3D motion data set. We do not assume strict rigidity but only that the markers maintain a constant distance from the centre or axis of rotation. This method is compared with other methods that use similar assumptions in the cases of random measurement errors, systematic skin movements and skin movements with random measurement noise. Simulation results indicate that the new method is superior in terms of the algorithm used, the closure of the solution, consistency and minimal manual parameter adjustment. The method can also be adapted to joints with translational movements.