Evaluation method of infrared thermography on children with idiopathic thrombocytopenic purpura: Preliminary

Abstract

Listen

Translate article

PurposeIdiopathic thrombocytopenic purpura (ITP) is an immune disease with complex pathogenesis. However, the diagnosis of ITP lacks of specificity test index, the diagnosis is still at an exclusive diagnosis level. Clinical studies have found that ITP can cause a disorder of the skin temperature distribution due to the spontaneous immune response. Human skin temperature can be collected by infrared thermography (IRT), and we can speculate on the severity of the immune response by extracting the temperature distribution characteristics. In this paper we will explore the value of IRT for the diagnosis and evaluation on ITP, and try to add new non-invasive inspection tools for ITP laboratory inspection. MethodAccording to the human metabolic function status, a spatial heterogeneity algorithm for skin temperature distribution was proposed in this paper based on infrared thermography. The algorithm was used to quantify the degree of confusion in a target region. Thirty-three subjects were divided into a control group, an ITP1 group and an ITP2 group in this study. The temperature distribution spatial non-uniformity, average temperature and maximum temperature characteristics of 14 ROI regions were extracted from the infrared thermal imaging of every subject. Six target regions with smaller P value were selected and single-dimensional evaluation parameter was extracted by the method of area weighting. At the same time, the sensitivity and specificity of the parameter to ITP were examined. ResultThe 6 target regions selected by the non-uniformity feature had significant differences (P < 0.05) among the three groups with the statistical method of ANOVA, but there were no significantly statistical difference on the regions extracted by average temperature and maximum temperature. It can be found from the ROC curve that the evaluation method proposed in this paper has higher sensitivity and specificity, than the common methods. ConclusionThe skin temperature distribution characteristics of the ITP children can be displayed by an infrared thermography. All of the 6 target regions screened by the spatial non-uniformity algorithm proposed in this paper are torso parts, which match the conclusions of relevant literature research. The method presented in this paper has high sensitivity and specificity to ITP. Although there are some limitations, it can non-invasively and objectively describe the disease state and add a new tool for ITP laboratory examinations. This paper also provides a new method and idea for the medical evaluation of infrared thermography.