Challenges for Driver Action Recognition with Face Masks

Advanced Driver Assistance Systems (ADAS) are enabling technologies in Intelligent Transportation Systems. Modern ADAS include algorithms to classify drivers' actions and distractions, aiming at identifying situations in which the driver is inattentive. Such systems typically include components for Driver Action Recognition (DAR) and Visual Distraction Classification (VDC), which prevent risky situations during semi-autonomous driving. DAR and VDC often rely on cameras that track the driver and classify actions based on image recognition algorithms. The COVID-19 pandemic has changed several common social behaviours, including the widespread use of face mask even during driving. In some cases (taxi, bus) face covering policies are compulsory in many legislations. We here show that these behavioural changes challenge state-of-theart DAR and VDC systems, with the average F1-score in some scenarios dropping by around 30% when exposed to images of drivers wearing masks. Noting a lack of public datasets to update the ML classifiers performing such tasks, we contribute MASKDAR, a dataset for Action Recognition of Drivers wearing face Masks. Finally, using MASKDAR we show the importance of including subjects with face masks in datasets for DAR.

Challenges for Driver Action Recognition with Face Masks

Advanced Driver Assistance Systems (ADAS) are enabling technologies in Intelligent Transportation Systems. Modern ADAS include algorithms to classify drivers' actions and distractions, aiming at identifying situations in which the driver is inattentive. Such systems typically include components for Driver Action Recognition (DAR) and Visual Distraction Classification (VDC), which prevent risky situations during semi-autonomous driving. DAR and VDC often rely on cameras that track the driver and classify actions based on image recognition algorithms. The COVID-19 pandemic has changed several common social behaviours, including the widespread use of face mask even during driving. In some cases (taxi, bus) face covering policies are compulsory in many legislations. We here show that these behavioural changes challenge state-of-the-art DAR and VDC systems, with the average F1-score in some scenarios dropping by around 30% when exposed to images of drivers wearing masks. Noting a lack of public datasets to update the ML classifiers performing such tasks, we contribute Maskdar, a dataset for Action Recognition of Drivers wearing face Masks. Finally, using Maskdarwe show the importance of including subjects with face masks in datasets for DAR. © 2022 IEEE.

How breaks in nature can affect the users’ wellbeing: an experience based survey during the lockdown : Marco Gola

Background In the occasion of COVID-19 pandemic in Italy, the life of citizens was greatly disrupted - from healthcare professionals to the smart workers - and consequently also the state of mood. On the basis of the scientific evidences in relation to the relationship between the built environment and health, a research group has promoted an investigation on the benefits that greenery can have on the psychophysical state of the users, especially healthcare staff and users at home. Objectives The methodology adopted is the Profile of Mood States, which provides experiential activity in nature - without any technological distraction- to evaluate the benefits on mood. The methodology adopted refer to the shorter version (34 items) designed by prof. Grove at the University of Western Australia. In relation to the COVID-19 pandemic, the experience based questionnaire is differentiated for healthcare staff and general users. The questionnaire is composed of a few questions, to be completed before and after an experience in nature of 20/30 minutes. The investigation requires to be carried out in private gardens, balcony and/or terrace with greenery, public green areas, etc. Results 300 participants (subdivided into 225 general users and 75 healthcare professionals) took part in the investigations. Data analysis highlighted the higher performances in anxiety, depression, anger, force, fatigue and confusion, in particular for users who had the experience in garden (-50/70%), and among the healthcare staff the best outcomes are related to who did the investigation during or after the workshift (-60/-90%). Conclusions Although it is well-known the benefits that nature affects positively on well-being and stress level of users, the investigation underlines that brief breaks in the nature - especially in period of great stress such as pandemic - can influence the well-being and mental health of users.

Hyperoxemia in invasively ventilated COVID-19 patients-Insights from the PRoVENT-COVID study.

OBJECTIVE: We determined the prevalences of hyperoxemia and excessive oxygen use, and the epidemiology, ventilation characteristics and outcomes associated with hyperoxemia in invasively ventilated patients with coronavirus disease 2019 (COVID-19). METHODS: Post hoc analysis of a national, multicentre, observational study in 22 ICUs. Patients were classified in the first two days of invasive ventilation as 'hyperoxemic' or 'normoxemic'. The co-primary endpoints were prevalence of hyperoxemia (PaO2 > 90 mmHg) and prevalence of excessive oxygen use (FiO2 ≥ 60% while PaO2 > 90 mmHg or SpO2 > 92%). Secondary endpoints included ventilator settings and ventilation parameters, duration of ventilation, length of stay (LOS) in ICU and hospital, and mortality in ICU, hospital, and at day 28 and 90. We used propensity matching to control for observed confounding factors that may influence endpoints. RESULTS: Of 851 COVID-19 patients, 225 (26.4%) were classified as hyperoxemic. Excessive oxygen use occurred in 385 (45.2%) patients. Acute respiratory distress syndrome (ARDS) severity was lowest in hyperoxemic patients. Hyperoxemic patients were ventilated with higher positive end-expiratory pressure (PEEP), while rescue therapies for hypoxemia were applied more often in normoxemic patients. Neither in the unmatched nor in the matched analysis were there differences between hyperoxemic and normoxemic patients with regard to any of the clinical outcomes. CONCLUSION: In this cohort of invasively ventilated COVID-19 patients, hyperoxemia occurred often and so did excessive oxygen use. The main differences between hyperoxemic and normoxemic patients were ARDS severity and use of PEEP. Clinical outcomes were not different between hyperoxemic and normoxemic patients.

The effect of age on ventilation management and clinical outcomes in critically ill COVID-19 patients-insights from the PRoVENT-COVID study

Introduction: We analyzed the association of age with ventilation practice and outcomes in critically ill COVID-19 patients requiring invasive ventilation. Methods: Posthoc analysis of the PRoVENT-COVID study, an observational study performed in 22 ICUs in the first 3 months of the national outbreak in the Netherlands. The coprimary endpoint was a set of ventilator parameters, including tidal volume normalized for predicted bodyweight, positive end-expiratory pressure, driving pressure, and respiratory system compliance in the first 4 days of invasive ventilation. Secondary endpoints were other ventilation parameters, the use of rescue therapies, pulmonary and extrapulmonary complications in the first 28 days in the ICU, hospital- and ICU stay, and mortality. Results: 1122 patients were divided into four groups based on age quartiles. No meaningful differences were found in ventilation parameters and in the use of rescue therapies for refractory hypoxemia in the first 4 days of invasive ventilation. Older patients received more often a tracheostomy, developed more frequently acute kidney injury and myocardial infarction, stayed longer in hospital and ICU, and had a higher mortality. Conclusions: In this cohort of invasively ventilated critically ill COVID-19 patients, age had no effect on ventilator management. Higher age was associated with more complications, longer length of stay in ICU and hospital and a higher mortality.

Practice of tracheostomy in patients with acute respiratory failure related to COVID-19 - Insights from the PRoVENT-COVID study.

OBJECTIVE: Invasively ventilated patients with acute respiratory failure related to coronavirus disease 2019 (COVID-19) potentially benefit from tracheostomy. The aim of this study was to determine the practice of tracheostomy during the first wave of the pandemic in 2020 in the Netherlands, to ascertain whether timing of tracheostomy had an association with outcome, and to identify factors that had an association with timing. METHODS: Secondary analysis of the 'PRactice of VENTilation in COVID-19' (PRoVENT-COVID) study, a multicenter observational study, conducted from March 1, 2020 through June 1, 2020 in 22 Dutch intensive care units (ICU) in the Netherlands. The primary endpoint was the proportion of patients receiving tracheostomy; secondary endpoints were timing of tracheostomy, duration of ventilation, length of stay in ICU and hospital, mortality, and factors associated with timing. RESULTS: Of 1023 patients, 189 patients (18.5%) received a tracheostomy at median 21 [17 to 28] days from start of ventilation. Timing was similar before and after online publication of an amendment to the Dutch national guidelines on tracheostomy focusing on COVID-19 patients (21 [17-28] vs. 21 [17-26] days). Tracheostomy performed ≤ 21 days was independently associated with shorter duration of ventilation (median 26 [21 to 32] vs. 40 [34 to 47] days) and higher mortality in ICU (22.1% vs. 10.2%), hospital (26.1% vs. 11.9%) and at day 90 (27.6% vs. 14.6%). There were no patient demographics or ventilation characteristics that had an association with timing of tracheostomy. CONCLUSIONS: Tracheostomy was performed late in COVID-19 patients during the first wave of the pandemic in the Netherlands and timing of tracheostomy possibly had an association with outcome. However, prospective studies are needed to further explore these associations. It remains unknown which factors influenced timing of tracheostomy in COVID-19 patients.

Vaccine-induced severe thrombotic thrombocytopenia following COVID-19 vaccination: a report of an autoptic case and review of the literature.

OBJECTIVE: Vaccine-induced immune thrombocytopenia (VITT) is a new syndrome occurring primarily in healthy young adults, with a female predominance, after receiving the first dose of ChAdOx1 nCoV-19 vaccine. We describe VITT syndrome characterized by severe thrombosis and thrombocytopenia found in our patient, with fatal outcome. CASE REPORT: A 58-year-old man, after 13 days from the first administration of ChAdOx1 nCoV-19 vaccine (AstraZeneca), presented with abdominal pain, diarrhea and vomitus. Laboratory tests revealed a severe thrombocytopenia, low fibrinogen serum levels and marked increase of D-dimer serum levels. The patient quickly developed a multiple organ failure, till death, three days after the hospital admission. RESULTS: At histology, in the lungs, interalveolar septa appeared thickened with microthrombi in the capillaries and veins. Interalveolar septa appeared thickened and showed vascular proliferation. Thrombi were detected in the capillaries of glomerular tufts. In the hearth, thrombi were observed in veins and capillaries. In the liver, voluminous fibrin thrombi were diffusely observed in the branches of the portal vein. Microthrombi were also found in the vasa vasorum of the wall of abdominal aorta. In the brain, microthrombi were observed in the capillaries of the choroid plexuses. Diffuse hemorrhagic necrosis was observed in the intestinal wall with marked congestion of the venous vessels. CONCLUSIONS: In our patient, the majority of data necessary for a VITT final diagnosis were present: thrombocytopenia and thrombosis in pulmonary, portal, hepatic, renal and mesenteric veins, associated with a marked increase of D-dimer serum levels. The finding of cerebral thrombosis in choroid plexuses, is a new finding in VITT. These features are suggestive for a very aggressive form of VITT.

Impaired ventilation is not independently associated with 28-day mortality in coronavirus disease 2019 (COVID19) acute respiratory distress syndrome (ARDS)

INTRODUCTION Surrogates for impaired ventilation such as estimated dead-space fractions and the ventilatory ratio are independently associated with an increased risk of mortality in the acute respiratory distress syndrome (ARDS) and small case series of COVID-19 related ARDS. METHODS This study aimed to quantify the dynamics and determine the prognostic value of surrogate markers of impaired ventilation in patients with COVID-19 related ARDS. The present study is a secondary analysis of the PRactice Of VENTilation in COVID-19 patients (PROVENT-COVID) in 22 intensive care unit hospitals in the Netherlands. Surrogates of impaired ventilation such as the estimated dead space fraction (by Harris-Benedict-VD/VT HB and direct method-VD/VT DIR), ventilatory ratio (VR), and end-tidal-to-arterial PCO2 ratio (PETCO2/PaCO2) were used. RESULTS 927 consecutive patients admitted with COVID-19 related ARDS were included in this study. Surrogates of impaired ventilation were significantly higher in non-survivors than survivors at baseline and during the following days of mechanical ventilation (p <0.001). As ARDS severity increased, mortality increased with successive tertiles for VD/VT HB and VD/VT DIR, and VR, and decreased with successive tertiles for PETCO2/PaCO2. Mortality over the first 28 days was higher in patients in the high group of dead space fraction by VD/VT HB (16.4% vs. 12.3%;p = 0.003), but similar in the groups considering the dead space fraction by VD/VT DIR (15.4% vs. 13.3%;p = 0.100), and VR (15.5% vs. 13.2%;p = 0.080) (Figure 2). After adjustment for a base risk model that included chronic comorbidities, ventilation and oxygenation parameters, none of the surrogates of impaired ventilation measured at the start of ventilation or the following days were significantly associated with 28-day mortality. CONCLUSION Surrogate markers for impaired ventilation are abnormal at the start of invasive ventilation in patients with COVID-19 related ARDS and worsen during consequent days. Ventilation impairment seems to be more extensive in non-survivors than in survivors, but they do not yield prognostic information when added to a baseline risk model. In the absence of bedside capnography, surrogates of impaired ventilation may serve as an important tool to assess the severity of COVID-19 related ARDS along with other variables such as oxygenation abnormalities and respiratory mechanics.