AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT

Abstract Background The number of thyroid cancer diagnoses has been increasing for several decades, with a significant part of cases being detected incidentally (thyroid incidentaloma, TI) by imaging studies performed for reasons other than thyroid disease, including PET/CT with [18F]FDG. The chacte...

Full description

Saved in:
Bibliographic Details
Main Authors: Krystian Ślusarz, Mikolaj Buchwald, Adrian Szczeszek, Szymon Kupinski, Anna Gramek-Jedwabna, Wojciech Andrzejewski, Juliusz Pukacki, Robert Pękal, Marek Ruchała, Rafał Czepczyński, Cezary Mazurek
Format: Article
Language:English
Published: SpringerOpen 2025-04-01
Series:EJNMMI Research
Subjects:
Online Access:https://doi.org/10.1186/s13550-025-01228-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849726424313757696
author Krystian Ślusarz
Mikolaj Buchwald
Adrian Szczeszek
Szymon Kupinski
Anna Gramek-Jedwabna
Wojciech Andrzejewski
Juliusz Pukacki
Robert Pękal
Marek Ruchała
Rafał Czepczyński
Cezary Mazurek
author_facet Krystian Ślusarz
Mikolaj Buchwald
Adrian Szczeszek
Szymon Kupinski
Anna Gramek-Jedwabna
Wojciech Andrzejewski
Juliusz Pukacki
Robert Pękal
Marek Ruchała
Rafał Czepczyński
Cezary Mazurek
author_sort Krystian Ślusarz
collection DOAJ
description Abstract Background The number of thyroid cancer diagnoses has been increasing for several decades, with a significant part of cases being detected incidentally (thyroid incidentaloma, TI) by imaging studies performed for reasons other than thyroid disease, including PET/CT with [18F]FDG. The chacteristics of the detected TI cannot be determined solely on the basis of conventional parameters used in everyday clinical practice, such as SUVmax. In recent years, there has been a growing interest in radiomics, which is a quantitative method of analyzing radiological images based on the analysis of image texture. Textural analysis may be helpful, as it allows to characterize features invisible to the physician with the naked eye. Results Of the 54 patients who presented focal [18F]FDG-avid TI and had subsequent fine needle aspiration biopsy, 4 patients were excluded from the analysis due to the unavailability of the final diagnostic information. Hence, in the final analysis, data from 50 patients were used (39 females and 11 males) with a mean age of 58.5 ± 11.26. Of these 50 patients, 11 (22.0%) [18F]FDG-avid nodules were diagnosed as malignant. The performance of the XGBoost model in assessing [18F]FDG-avid TI was similar (0.846 [confidence interval, CI, 95% 0.737–0.956]) to SUVmax (0.797 [CI 95%: 0.622–0.973]; p = 0.60). Conclusions With an AI-based algorithm using radiomics features it is possible to detect the malignancy of thyroid nodule. However, no statistically significant differences were observed between the AI and radiomics approach, and when using a conventional measure, i.e., SUVmax.
format Article
id doaj-art-d5f1d01ac01f44f5b2acf86513f128e8
institution DOAJ
issn 2191-219X
language English
publishDate 2025-04-01
publisher SpringerOpen
record_format Article
series EJNMMI Research
spelling doaj-art-d5f1d01ac01f44f5b2acf86513f128e82025-08-20T03:10:10ZengSpringerOpenEJNMMI Research2191-219X2025-04-011511910.1186/s13550-025-01228-4AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CTKrystian Ślusarz0Mikolaj Buchwald1Adrian Szczeszek2Szymon Kupinski3Anna Gramek-Jedwabna4Wojciech Andrzejewski5Juliusz Pukacki6Robert Pękal7Marek Ruchała8Rafał Czepczyński9Cezary Mazurek10Department of Nuclear Medicine, AffideaPoznan Supercomputing and Networking Center, Polish Academy of SciencePoznan Supercomputing and Networking Center, Polish Academy of SciencePoznan Supercomputing and Networking Center, Polish Academy of ScienceDepartment of Nuclear Medicine, AffideaPoznan Supercomputing and Networking Center, Polish Academy of SciencePoznan Supercomputing and Networking Center, Polish Academy of SciencePoznan Supercomputing and Networking Center, Polish Academy of ScienceDepartment of Endocrinology, Metabolism and Internal Medicine, Poznan University of Medical SciencesDepartment of Nuclear Medicine, AffideaPoznan Supercomputing and Networking Center, Polish Academy of ScienceAbstract Background The number of thyroid cancer diagnoses has been increasing for several decades, with a significant part of cases being detected incidentally (thyroid incidentaloma, TI) by imaging studies performed for reasons other than thyroid disease, including PET/CT with [18F]FDG. The chacteristics of the detected TI cannot be determined solely on the basis of conventional parameters used in everyday clinical practice, such as SUVmax. In recent years, there has been a growing interest in radiomics, which is a quantitative method of analyzing radiological images based on the analysis of image texture. Textural analysis may be helpful, as it allows to characterize features invisible to the physician with the naked eye. Results Of the 54 patients who presented focal [18F]FDG-avid TI and had subsequent fine needle aspiration biopsy, 4 patients were excluded from the analysis due to the unavailability of the final diagnostic information. Hence, in the final analysis, data from 50 patients were used (39 females and 11 males) with a mean age of 58.5 ± 11.26. Of these 50 patients, 11 (22.0%) [18F]FDG-avid nodules were diagnosed as malignant. The performance of the XGBoost model in assessing [18F]FDG-avid TI was similar (0.846 [confidence interval, CI, 95% 0.737–0.956]) to SUVmax (0.797 [CI 95%: 0.622–0.973]; p = 0.60). Conclusions With an AI-based algorithm using radiomics features it is possible to detect the malignancy of thyroid nodule. However, no statistically significant differences were observed between the AI and radiomics approach, and when using a conventional measure, i.e., SUVmax.https://doi.org/10.1186/s13550-025-01228-4RadiomicsMachine learningAIThyroid incidentalomaThyroid cancerPET/CT
spellingShingle Krystian Ślusarz
Mikolaj Buchwald
Adrian Szczeszek
Szymon Kupinski
Anna Gramek-Jedwabna
Wojciech Andrzejewski
Juliusz Pukacki
Robert Pękal
Marek Ruchała
Rafał Czepczyński
Cezary Mazurek
AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
EJNMMI Research
Radiomics
Machine learning
AI
Thyroid incidentaloma
Thyroid cancer
PET/CT
title AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
title_full AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
title_fullStr AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
title_full_unstemmed AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
title_short AI may help to predict thyroid nodule malignancy based on radiomics features from [18F]FDG PET/CT
title_sort ai may help to predict thyroid nodule malignancy based on radiomics features from 18f fdg pet ct
topic Radiomics
Machine learning
AI
Thyroid incidentaloma
Thyroid cancer
PET/CT
url https://doi.org/10.1186/s13550-025-01228-4
work_keys_str_mv AT krystianslusarz aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT mikolajbuchwald aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT adrianszczeszek aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT szymonkupinski aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT annagramekjedwabna aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT wojciechandrzejewski aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT juliuszpukacki aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT robertpekal aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT marekruchała aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT rafałczepczynski aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct
AT cezarymazurek aimayhelptopredictthyroidnodulemalignancybasedonradiomicsfeaturesfrom18ffdgpetct