Association of body mass index with long-term outcomes in older adults hospitalized for COVID-19: an observational study.

Both underweight and obesity have been associated with poor prognosis in COVID-19. In an older populations of patients hospitalized for SARS-CoV-2 infection, we aimed to evaluate the association between body mass index (BMI) and short and long-term prognosis. Among 434 consecutive patients aged ≥ 70 years and hospitalized for suspected COVID-19 at a university hospital, 219 patients (median age of 83 years, 53% male) testing positive for COVID-19 and for whom BMI was recorded at admission, agreed to participate. Among them, 39 had a BMI < 20 kg/m2, 73 had a BMI between 20 and 24.9 kg/m2 and 107 had a BMI ≥ 25 kg/m2. After adjustment for confounders, BMI < 20 kg/m2 was associated with a higher risk of one-year mortality (hazard ratio (HR) [95% confidence interval]: 1.75 [1.00-3.05], p = 0.048), while BMI ≥ 25 kg/m2 was not (HR: 1.04 [0.64-1.69], p = 0.9). However, BMI was linearly correlated with both in-hospital acute respiratory failure (p = 0.02) and cardiovascular events (p = 0.07). In this cohort of older patients hospitalized for COVID-19, low BMI, rather than high BMI, appears as an independent risk factor for death after COVID-19. The pathophysiological patterns underlying this excess mortality remain to be elucidated.

A Cross-Sectional Reproducibility Study of a Standard Camera Sensor Using Artificial Intelligence to Assess Food Items: The FoodIntech Project.

Having a system to measure food consumption is important to establish whether individual nutritional needs are being met in order to act quickly and to minimize the risk of undernutrition. Here, we tested a smartphone-based food consumption assessment system named FoodIntech. FoodIntech, which is based on AI using deep neural networks (DNN), automatically recognizes food items and dishes and calculates food leftovers using an image-based approach, i.e., it does not require human intervention to assess food consumption. This method uses one-input and one-output images by means of the detection and synchronization of a QRcode located on the meal tray. The DNN are then used to process the images and implement food detection, segmentation and recognition. Overall, 22,544 situations analyzed from 149 dishes were used to test the reliability of this method. The reliability of the AI results, based on the central intra-class correlation coefficient values, appeared to be excellent for 39% of the dishes (n = 58 dishes) and good for 19% (n = 28). The implementation of this method is an effective way to improve the recognition of dishes and it is possible, with a sufficient number of photos, to extend the capabilities of the tool to new dishes and foods.