ABSTRACT

Multiple sclerosis (MS) is a progressive and degenerative disease that causes nerve conduction blocks due to demyelination in the central nervous system. Most MS patients experience a worsening of clinical signs and neurological symptoms when they are exposed to heat due to a thermoregulatory dysfunction. This paper proposes a novel methodology to understand temperature changes in MS patients by obtaining and evaluating texture features from infrared thermography (IRT) images. For that purpose, images of the legs of a MS patient and a healthy control subject with similar physical characteristics (while at rest and in a standing position) were recorded using a FLIR A655SC infrared camera. In the quantitative analysis of the resulting IRT images, three texture features (average, entropy, and uniformity) were computed, and the results were compared using statistical techniques. The statistical analysis showed that temperatures in the MS patient were not normally distributed, while those in the healthy control subject were normally distributed. In addition, significant differences in average, entropy, and uniformity were found between subjects. This methodology enables a quantitative evaluation of thermal distributions over different regions of the body and can be used in further studies into temperature changes in MS patients.

摘要

多发性硬化症 (MS) 是一种进行性和退行性疾病，由于中枢神经系统脱髓鞘而导致神经传导阻滞。大多数 MS 患者在因体温调节功能障碍而暴露于高温时会出现临床症状和神经系统症状的恶化。本文提出了一种新的方法，通过从红外热成像 (IRT) 图像中获取和评估纹理特征来了解 MS 患者的温度变化。为此，使用 FLIR A655SC 红外热像仪记录了一名 MS 患者和一名身体特征相似的健康对照受试者的腿部图像（在休息和站立时）。在对生成的 IRT 图像进行定量分析时，计算了三个纹理特征（平均、熵和均匀性），并使用统计技术对结果进行比较。统计分析表明，MS 患者的体温不是正态分布的，而健康对照组的体温是正态分布的。此外，受试者之间的平均、熵和均匀性存在显着差异。这种方法能够对身体不同区域的热分布进行定量评估，并可用于进一步研究 MS 患者的温度变化。