Prediction of Malignant Thyroid Nodules Using 18F-FDG PET/CT–Based Radiomics Features in Thyroid Incidentalomas

Ko, Woo Seog; Kim, Seong-Jang

doi:10.1097/rlu.0000000000004637

Cited by 3 publications

(3 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ko et al [36] performed a meta-analysis to investigate the potential of radiomics for the prediction of malignant thyroid nodules in patients with thyroid incidentaloma across 5 studies with 518 patients and found a pooled sensitivity of FDG PET/CT of 0.77 and a pooled specificity of 0.67. The authors concluded that FDG PET/CT-based radiomics features demonstrated a good diagnostic performance for the prediction of malignant thyroid nodules.…”

Section: Classification and Prediction Of Thyroid Nodulesmentioning

confidence: 99%

Clinical Applications of Radiomics in Nuclear Medicine

Lohmann,

Bundschuh,

Miederer

et al. 2023

Nuklearmedizin

View full text Add to dashboard Cite

Radiomics is an emerging field of artificial intelligence that focuses on the extraction and analysis of quantitative features such as intensity, shape, texture and spatial relationships from medical images. These features, often imperceptible to the human eye, can reveal complex patterns and biological insights. They can also be combined with clinical data to create predictive models using machine learning to improve disease characterization in nuclear medicine. This review article examines the current state of radiomics in nuclear medicine and shows its potential to improve patient care. Selected clinical applications for diseases such as cancer, neurodegenerative diseases, cardiovascular problems and thyroid diseases are examined. The article concludes with a brief classification in terms of future perspectives and strategies for linking research findings to clinical practice.

show abstract

Section: Classification and Prediction Of Thyroid Nodulesmentioning

confidence: 99%

Clinical Applications of Radiomics in Nuclear Medicine

Lohmann,

Bundschuh,

Miederer

et al. 2023

Nuklearmedizin

View full text Add to dashboard Cite

show abstract

“…In the recent meta-analysis by Ko and colleagues 78 , the use of 18 F-FDG PET/CT-based RFs applied to thyroid incidentaloma was investigated. The authors included 5 studies, with differences in terms of RFs extraction and selection methods, types and number of features, and modelling methods.…”

Section: Other Biomarkers For Itn Risk Assessment: Pet/ct Imaging And...mentioning

confidence: 99%

Molecular testing in indeterminate thyroid nodules: an additional tool for clinical decision-making

Fumagalli,

Serio

2023

Pathologica

View full text Add to dashboard Cite

Summary Thyroid nodules are commonly encountered in clinical practice, affecting up to 50% of the population. The large majority of thyroid lumps are benign incidental findings detected by imaging, while approximately 5-15% harbor malignancy. For a target patient’s care, it is of paramount importance to identify and treat thyroid malignancy, while preventing unnecessary invasive surgery in patients with benign lesions. Although fine needle aspiration (FNA) associated with cytological examination provides malignant risk information, 20-30% of diagnoses fall into the “indeterminate thyroid nodule” (ITN) category. ITN clinical management remains a challenging issue for physicians since the ITN risk of malignancy varies from 5% to 40% and most thyroid nodules undergo overtreatment with surgery procedures. ITN molecular testing may better define malignant risk in the single nodule and is able to discriminate with accuracy benign from malignant nodules. Nowadays there are different technologies and different molecular panels, each with its own specificity, sensitivity and predictive values. In view of widespread introduction of molecular testing, some outstanding questions remain and are addressed in the present review such as the presence of molecular panels acting as “rule in” or “rule out” tools, the effective impact of testing results in the clinical decision-making process, and the prohibitive cost of commercial assays associated with the lack of test reimbursement in national health systems.

show abstract

“…CT imaging features from peritumor tissue including shape and texture features were used to predict distance metastasis in locally advanced NSCLC (Dou et al 2018). Recent studies extracted and evaluated radiomic features to predict the clinical prognosis of patients (van Timmeren et al 2017, Ko and Kim 2023).…”

Section: Introductionmentioning

confidence: 99%

GWO+RuleFit: rule-based explainable machine-learning combined with heuristics to predict mid-treatment FDG PET response to chemoradiation for locally advanced non-small cell lung cancer

Duan,

Liu,

Wang

et al. 2024

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective: Vital rules learned from FDG-PET radiomics of tumor subregional response can provide clinical decision support for precise treatment adaptation. We combined a rule-based machine learning (ML) model (RuleFit) with a heuristic algorithm (Gray Wolf Optimizer, GWO) for mid-chemoradiation FDG-PET response prediction in patients with locally advanced non-small cell lung cancer. Approach: Tumors subregions were identified using K-means clustering. GWO+RuleFit consists of three main parts: (i) a random forest is constructed based on conventional features or radiomic features extracted from tumor regions or subregions in FDG-PET images, from which the initial rules are generated; (ii) GWO is used for iterative rule selection; (iii) the selected rules are fit to a linear model to make predictions about the target variable. Two target variables were considered: a binary response measure (∆SUVmean⩾20% decline) for classification and a continuous response measure (∆SUVmean) for regression. GWO+RuleFit was benchmarked against common ML algorithms and RuleFit, with leave-one-out cross-validated performance evaluated by the area under the receiver operating characteristic curve (AUC) in classification and root-mean-square error (RMSE) in regression. Main results: GWO+RuleFit selected 15 rules from the radiomic feature dataset of 23 patients. For treatment response classification, GWO+RuleFit attained numerically better cross-validated performance than RuleFit across tumor regions and sets of features (AUC:0.58-0.86 vs. 0.52-0.78, p=0.170-0.925). GWO+Rulefit also had the best or second-best performance numerically compared to all other algorithms for all conditions. For treatment response regression prediction, GWO+RuleFit (RMSE:0.162-0.192) performed better numerically for low-dimensional models (p=0.097-0.614) and significantly better for high-dimensional models across all tumor regions except one (RMSE:0.189-0.219, p<0.004). Significance: The GWO+RuleFit selected rules were interpretable, highlighting distinct radiomic phenotypes that modulated treatment response. GWO+Rulefit achieved parsimonious models while maintaining utility for treatment response prediction, which can aid clinical decisions for patient risk stratification, treatment selection, and biologically driven adaptation.Clinical trial: NCT02773238

show abstract

Prediction of Malignant Thyroid Nodules Using 18F-FDG PET/CT–Based Radiomics Features in Thyroid Incidentalomas

Cited by 3 publications

References 33 publications

Clinical Applications of Radiomics in Nuclear Medicine

Clinical Applications of Radiomics in Nuclear Medicine

Molecular testing in indeterminate thyroid nodules: an additional tool for clinical decision-making

GWO+RuleFit: rule-based explainable machine-learning combined with heuristics to predict mid-treatment FDG PET response to chemoradiation for locally advanced non-small cell lung cancer

Contact Info

Product

Resources

About