Background: The homeostatic euthyroid set point of the hypothalamus-pituitary-thyroid axis of any given individual is unique and oscillates narrowly within substantially broader normal population ranges of circulating free thyroxine (FT4) and thyroid-stimulating hormone (TSH), otherwise termed 'thyroid function test (TFT)'. We developed a mathematical algorithm codenamed Thyroid-SPOT that effectively reconstructs the personalized set point in open-loop situations and evaluated its performance in a retrospective patient sample.
Methods: We computed the set points of 101 patients who underwent total thyroidectomy for non-functioning thyroid disease using Thyroid-SPOT on each patient's own serial post-thyroidectomy TFT. Every predicted set point was compared against its respective healthy pre-operative euthyroid TFT per individual and their separation (i.e. predicted-observed TFT) quantified.
Results: Bland-Altman analysis to measure the agreement between each pair of an individual's predicted and actual set points revealed a mean difference in FT4 and TSH of + 3.03 pmol/L (95% CI 2.64, 3.43) and - 0.03 mIU/L (95% CI - 0.25, 0.19), respectively. These differences are small compared to the width of the reference intervals. Thyroid-SPOT can predict the euthyroid set point remarkably well, especially for TSH with a 10-16-fold spread in magnitude between population normal limits.
Conclusion: Every individual's equilibrium euthyroid set point is unique. Thyroid-SPOT serves as an accurate, precise and reliable targeting system for optimal personalized restoration of euthyroidism. This algorithm can guide clinicians in L-thyroxine dose titrations to resolve persistent dysthyroid symptoms among challenging cases harbouring "normal TFT" within the laboratory ranges but differing significantly from their actual euthyroid set points.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1007/s40618-020-01390-7 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Foundation Models for Translational Cancer Biology.
Annu Rev Biomed Data Sci
January 2025
1Department of Computational Biomedicine, Cedars-Sinai Medical Center, Los Angeles, California, USA;
Cancer remains a leading cause of death globally. The complexity and diversity of cancer-related datasets across different specialties pose challenges in refining precision medicine for oncology. Foundation models offer a promising solution.
View Article and Find Full Text PDFWithin- and Between-Person Differences in Activity Factor Structure: Results from an Ecological Momentary Assessment Study.
J Gerontol B Psychol Sci Soc Sci
January 2025
Department of Human Development and Family Studies, Pennsylvania State University, State College, Pennsylvania, USA.
Objective: Studies using ecological momentary assessment (EMA) of activity participation rely on items tapping domains informed by factor analyses based on single time points. Analyses from a single time point focus on differences between participants and provide little insight into how activities cluster together within a person across moments or days. The present study compared the factor structure in activity participation between- and within-persons using an expanded set of momentary activity items in middle and older adulthood.
View Article and Find Full Text PDFAdaptive weighted progressive iterative approximation based on coordinate decomposition.
PLoS One
January 2025
School of Mathematics and Finance, Hunan University of Humanities, Science and Technology, Loudi, China.
During the iterative process of the progressive iterative approximation, it is necessary to calculate the difference between the current interpolation curve and the corresponding data points, known as the adjustment vector. To achieve more precise adjustments of control points, this paper decomposes the adjustment vector into its coordinate components and introduces a weight for each component. By dynamically adjusting these weights, we can accelerate the convergence of iterations and enhance approximation accuracy.
View Article and Find Full Text PDFGenerative artificial intelligence enables the generation of bone scintigraphy images and improves generalization of deep learning models in data-constrained environments.
Eur J Nucl Med Mol Imaging
January 2025
Department of Biomedical Imaging and Image-guided Therapy, Division of Nuclear Medicine, Medical University of Vienna, Spitalgasse 23, Vienna, 1090, Austria.
Purpose: Advancements of deep learning in medical imaging are often constrained by the limited availability of large, annotated datasets, resulting in underperforming models when deployed under real-world conditions. This study investigated a generative artificial intelligence (AI) approach to create synthetic medical images taking the example of bone scintigraphy scans, to increase the data diversity of small-scale datasets for more effective model training and improved generalization.
Methods: We trained a generative model on Tc-bone scintigraphy scans from 9,170 patients in one center to generate high-quality and fully anonymized annotated scans of patients representing two distinct disease patterns: abnormal uptake indicative of (i) bone metastases and (ii) cardiac uptake indicative of cardiac amyloidosis.
Spatial distribution of fine-grained floodplain deposits and anthropogenic materials based on official borehole data in the floodplain of Leipzig, Germany.
Data Brief
February 2025
Institute for Geography, Leipzig University, Johannisallee 19a, Leipzig, 04103, Germany.
This data set includes the spatial model of the thickness and distribution of fine-grained floodplain deposits in the Leipzig floodplain area. The data set originates from borehole records provided by the Saxon State Office for Environment, Agriculture, and Geology [1]. The data processing involved the categorization of the stratigraphic descriptions of the borehole logs.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!