Majid Hassan-Ghomi, Bahareh Nikooyeh, Soudabeh Motamed, Tirang R. Neyestani,
Volume 31, Issue 1 (1-2017)
Abstract
Background: In several disease conditions, patients must inevitably be nourished by enteral feeding (EF). Though in many countries, commercial formulas are routinely used for EF, in Iran still home-made formulas are commonly employed as commercial formulas are not covered by insurance. This may pose patients to malnutrition and bring about further costs. The aim of this study was to evaluate the efficacy of EF commercial formulas in comparison with home-made formulas and thus to make further evidence for insurance policy-making
Methods: Medline, Cochrane, Embass and Center for Review & Dissemination (CRD) as well as IranDoc and SID databases were searched. Keywords included formula, ICU, and enteral nutrition or tube feeding. No clinical trial study on the efficacy of EF formulas was found. Therefore, the compositions of available formulas and their cost-effectiveness were evaluated based on the clinical guidelines of scientific bodies such as American Society for Parenteral and Enteral Nutrition (ASPEN), European Society for Parenteral and Enteral Nutrition (ESPEN) and relative articles available in PubMed. In addition, the expert opinions were also taken into consideration.
Results: Domestic commercial formulas seemed to less merit dietary recommended intakes, i.e. the amount of some nutrients were much higher, and some others were much lower than the recommended values. The amount of several micronutrients including vitamins B1, B6, C, D and K, as well as iron, calcium and magnesium were not sufficient to meet the body needs in most commercial formulas upon receiving 2000 kilocalories and less.
Conclusion: Clinical studies on the efficacy of commercial formulas in comparison with home-made formulas are needed. Meanwhile, making suitable conditions for increasing the diversity of artificial nutrition products in the market would help clinical nutritionists to make better choices according to their patients conditions and to reduce the costs, as well.
Zahra Ghomi, Reza Mirshahi, Arash Khameneh Bagheri, Ali Fattahpour, Saeed Mohammadiun, Abdorreza Alavi Gharahbagh, Abtin Djavadifar, Hossein Arabalibeik, Rehan Sadiq, Kasun Hewage,
Volume 34, Issue 1 (2-2020)
Abstract
Background: Lung CT scan has a pivotal role in diagnosis and monitoring of COVID-19 patients, and with growing number of affected individuals, the need for artificial intelligence (AI)-based systems for interpretation of CT images is emerging. In current investigation we introduce a new deep learning-based automatic segmentation model for localization of COVID-19 pulmonary lesions.
Methods: A total of 2469 CT scan slices, containing 1402 manually segmented abnormal and 1067 normal slices form 55 COVID-19 patients and 41 healthy individuals, were used to train a deep convolutional neural network (CNN) model based on Detectron2, an open-source modular object detection library. A dataset, including 1224 CT slices of 18 COVID-19 patients and 9 healthy individuals, was used to test the model.
Results: The accuracy, sensitivity, and specificity of the trained model in marking a single image slice with COVID-19 lesion were 0.954, 0.928, and 0.961, respectively. Considering a threshold of 0.4% for percentage of lung involvement, the model was capable of diagnosing the patients with COVID-19 pneumonia, with a sensitivity of 0.982% and a specificity of 88.5%. Furthermore, the mean Intersection over Union (IoU) index for the test dataset was 0.865.
Conclusion: The deep learning-based automatic segmentation method provides an acceptable accuracy in delineation and localization of COVID-19 lesions, assisting the clinicians and researchers for quantification of abnormal findings in chest CT scans. Moreover, instance segmentation is capable of monitoring longitudinal changes of the lesions, which could be beneficial to patients’ follow-up.
