Convolutional neural network-based computer-aided diagnosis in Hiesho (cold sensation)

A new biomarker for screening of Hiesho (cold sensation) in young females through feet temperature recovery rate after cooling test

“…Lastly, the temperature measurement method used in this study was semi-automatic. Considering the potential benefits of advanced image processing technology [2], and machine learning in identifying optimal biomarkers [22], future research focusing on automated and intelligent approaches to Hiesho assessment could yield more significant insights.…”

“…Cold sensation, commonly referred to as Hiesho, presents a prevalent health concern predominantly observed in females [1]. Hiesho manifests in three primary subtypes: peripheral artery disease type, hypo-metabolism type, and autonomic nerve type [2]. The primary symptom of Hiesho is a feeling of coldness, especially in the hands and feet, even in environments where a healthy individual would not experience such a sensation [2].…”

“…Hiesho manifests in three primary subtypes: peripheral artery disease type, hypo-metabolism type, and autonomic nerve type [2]. The primary symptom of Hiesho is a feeling of coldness, especially in the hands and feet, even in environments where a healthy individual would not experience such a sensation [2]. Although not considered a remarkable symptom in Western medicine, approximately 50% of young Japanese women suffer from Hiesho [3,4].…”

“…Interestingly, excessive sympathetic nerve activity has been reported in individuals with Hiesho, resulting in higher CVC activity [7]. Consequently, women with Hiesho experience decreased peripheral blood flow and skin temperature [2,9]. When exposed to cold environments, such as immersing their feet in cold water, the state of sensitivity to cold persists, leading to chronic stimulation of the sympathetic nerves, which tends to cause changes in blood circulation [10].…”

Detection and Classification of Teacher-Rated Children's Activity Levels Using Millimeter-wave Radar and Machine Learning: A Pilot Study in a Real Primary School Environment

“…Furthermore, we employed machine learning (ML), which has found extensive application in computer-aided diagnosis for analyzing both imaging and non-imaging data [16]. Once adequately trained with relevant features, ML has the potential to serve as a supplementary opinion or provide supporting information in the school-based evaluation process, thereby mitigating the workload for teachers, school-based healthcare professionals and children [17]. The identification of specific features from millimeter-wave radar data, successfully validated to be informative for classifying restlessness, holds the promise of training ML models to develop a computer-aided screening system for children's activity levels.…”