Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies.

The sudden outbreak and uncontrolled spread of COVID-19 disease is one of the most important global problems today. In a short period of time, it has led to the development of many deep neural network models for COVID-19 detection with modules for explainability. In this work, we carry out a systematic analysis of various aspects of proposed models. Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. We present the perspective of both: experts in the field - radiologists and deep learning engineers dealing with model explanations. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model.

Role of the rs2274907 allelic variant of the ITLN1 gene in patients with diabetic foot.

INTRODUCTION Diabetic foot (DF) is a serious complication of diabetes mellitus (DM) that occurs due to neuropathy or atherosclerosis of the lower limbs. Omentin (encoded by the ITLN1 gene) has been implicated as a protective factor in vascular complications of diabetes, likely due to its endothelial vasodilator activity and its anti­inflammatory actions. However, susceptibility to DF with respect to the allelic variants of the ITLN1 gene has not been studied so far. OBJECTIVES This study aimed to evaluate the association between the rs2274907 allelic variant of the ITLN1 gene and the occurrence of DF in patients with type 2 diabetes mellitus (T2DM).  PATIENTS AND METHODS The study included 670 individuals: 204 with T2DM and DF (DF group), 299 with T2DM without DF (T2DM group), and 167 healthy controls.  RESULTS Ischemic heart disease, retinopathy, nephropathy, neuropathy, obesity, hyperlipidemia, and active smoking were more frequent in the DF group than in the T2DM group. Allele A of the rs2274907 variant was observed more frequently in the DF group compared with healthy controls in an additive model (odds ratio [OR] = 0.7, P = 0.034). This effect was also sex­specific for males in both the additive and recessive models (OR = 0.6, P = 0.015 and OR = 0.52, P = 0.0017, respectively). However, no differences in the distribution of alleles was observed between the DF and T2DM groups. CONCLUSIONS The rs2274907 variant of the ITLN1 gene is associated with increased prevalence of DF.