TY - JOUR T1 - Re-Modelling 99m-Technetium Pertechnetate Thyroid Uptake; Statistical, Machine Learning and Deep Learning Approaches JF - Journal of Nuclear Medicine Technology JO - J. Nucl. Med. Technol. DO - 10.2967/jnmt.121.263081 SP - jnmt.121.263081 AU - Geoffrey M. Currie AU - Basit M. Iqbal Y1 - 2021/12/01 UR - http://tech.snmjournals.org/content/early/2021/12/07/jnmt.121.263081.abstract N2 - Background: While normal ranges for 99mTc thyroid percentage uptake vary, the seemingly intuitive evaluation of thyroid function does not reflect the complexity of thyroid pathology and biochemical status. The emergence of artificial intelligence (AI) in nuclear medicine has driven problem solving associated with logic and reasoning that warrant re-examination of established benchmarks in thyroid functional assessment. Methods: There were 123 patients retrospectively analysed in the study sample comparing scintigraphic findings to grounded truth established through biochemistry status. Conventional statistical approaches were used in conjunction with an artificial neural network (ANN) to determine predictors of thyroid function from data features. A convolutional neural network (CNN) was also used to extract features from the input tensor (images). Results: Analysis was confounded by sub-clinical hyperthyroidism, primary hypothyroidism, sub-clinical hypothyroidism and T3 toxicosis. Binary accuracy for identifying hyperthyroidism was highest for thyroid uptake classification using a threshold of 4.5% (82.6%), followed by pooled physician 6interpretation with the aid of uptake values (82.3%). Visual evaluation without quantitative values reduced accuracy to 61.0% for pooled physician determinations and 61.4% classifying on the basis of thyroid gland intensity relative to salivary glands. The machine learning (ML) algorithm produced 84.6% accuracy, however, this included biochemistry features not available to the semantic analysis. The deep learning (DL) algorithm had an accuracy of 80.5% based on image inputs alone. Conclusion: Thyroid scintigraphy is useful in identifying hyperthyroid patients suitable for radioiodine therapy when using an appropriately validated cut-off for the patient population (4.5% in this population). ML ANN algorithms can be developed to improve accuracy as second readers systems when biochemistry results are available. DL CNN algorithms can be developed to improve accuracy in the absence of biochemistry results. ML and DL do not displace the role of the physician in thyroid scintigraphy but could be used as second reader systems to minimize errors and increase confidence. ER -