Assisting the diagnosis of overt hypothyroidism with pattern recognition methods, making use of a set of routine tests, and their multiple correlation with total T4

Thyroid dysfunction diagnosis from routine laboratory tests based on machine learning

“…Similar to the present study, Aoki's et al 17 study used pattern recognition methods such as neural networks to predict the likelihood of thyroid dysfunction from a set of routine test parameters such as ALP, S-Cr, and TC. Their results suggested that most patients with overt thyroid dysfunction could be screened by using a set of routine clinical data without measuring thyroid hormone levels.…”

“…For the hypothyroidism classification in this study, the hypothyroidism group was used as a positive label, whereas both the control and hyperthyroidism groups were negatively labeled (referred to as "crosstalk on"). On the other hand, related studies 16,17 performed classification by setting thyroid dysfunction patients (with hyperthyroidism or hypothyroidism) as a positive label and only the control group as a negative label (referred to as "crosstalk off"). Therefore, we evaluated the performance of the models with similar settings as these studies.…”

“…Several previous studies revealed promising results from the use of machine learning approaches for predicting thyroid dysfunction (16,17).…”

“…In another study, the incidence of myocardial infarction or cerebral infarction was predicted using the results of a health checkup 10 . Numerous studies have also attempted to assess the efficacy of detecting misdiagnosed diseases, including thyroid dysfunction [11][12][13][14][15][16][17] . Aoki et al 16,17 found that there were strong, multiple correlations between a set of routine clinical parameters and FT4 in patients with both overt hyperthyroidism and overt hypothyroidism.…”

“…Numerous studies have also attempted to assess the efficacy of detecting misdiagnosed diseases, including thyroid dysfunction [11][12][13][14][15][16][17] . Aoki et al 16,17 found that there were strong, multiple correlations between a set of routine clinical parameters and FT4 in patients with both overt hyperthyroidism and overt hypothyroidism. These studies used pattern recognition methods such as neural networks and predicted the likelihood of thyroid dysfunction from a set of routine clinical tests.…”