EXTERNAL VALIDATION OF AIBX V2 ALGORITHM FOR THYROID NODULE RISK STRATIFICATION
External validation of AIBx version 2 was presented at the Amercian Thyroid Association's annual meeting (2022) at Montreal by Dr. Singh et.al.
AIBx V2 had a negative predictive value of 98.5%, which was similar to fine needle aspiration biopsies. Sensitivity and specificity for the model were 92% and 86% respectively. Full abstract can be viewed at Late Breaking Abstracts | Thyroid (liebertpub.com)
Combining Image Similarity and Predictive AI Models to Decrease Subjectivity in Thyroid Nodule Diagnosis and Improve Malignancy Prediction
Research by Nair et.al. presented at SIIM CMIMI 2022.
AIBx V2 was able to predict the histopathologically proven presence of malignancy accurately. None of the thyroid cancers were missed by the algorithm. The AI algorithm was able to reduce unnecessary biopsy in 50.5% of the nodules.
To read the full abstract and see the video presentation, please visit: - Call for Papers (ymaws.com) & Combining Image Similarity and Predictive AI Models to Decrease Subjectivity in Thyroid Nodule Diagn (siim.org)
Abstract
Background: Current classification systems for thyroid nodules are very subjective. Artificial intelligence (AI) algorithms have been used to decrease subjectivity in medical image interpretation. One out of 2 women over the age of 50 years may have a thyroid nodule and at present the only way to exclude malignancy is through invasive procedures for those that are suspicious on ultrasonography. Hence, there exists a need for noninvasive objective classification of thyroid nodules. Some cancers have benign appearance on ultrasonogram. Hence, we decided to create an image similarity algorithm rather than image classification algorithm.
External validation study in Denmark for AIBx version 1.
Initial internal validation study in the United States. PubMed link
National and international presentations
2019 Annual meeting of American Thyroid Association. Invited oral podium presentation.
2020 Thyroid, Head and Neck Cancer (THANC) Foundation, virtual journal club.
2020 University of Arizona College of Medicine Phoenix/ Banner University Medical Center-Phoenix Division of Endocrinology Grand Rounds.
2021 European Society of Endocrinology international meeting eECE 2021. Selected for oral presentation.
Pitch competition
Selected for pitch competition at Harvard Catalyst TRANSCEND program. 2021 cohort.
References
Thomas, J., & Haertling, T. (2020). AIBx, artificial intelligence model to risk stratify thyroid nodules. Thyroid.
Swan,K, Thomas,J, Nielsen,V, Jespersen,M & Bonnema, S. External validation of AIBx, an artificial intelligence model for risk stratification, in surgically resected thyroid nodules
Other articles and book chapters referencing AIBx:
Orloff, L. A. (2020). Artificial Intelligence plus Human Interpretation for Thyroid Nodule Risk Stratification: An Image Similarity Model Keeps the Physician in the Loop. Clinical Thyroidology, 32(6), 276-278.
Unnikrishnan, A. G., & Kalra, S. (2020). Could artificial intelligence help in the risk stratification of thyroid nodules?. Thyroid Research and Practice, 17(2), 51.
Thomas, J. (2020). Application of Artificial Intelligence in Thyroidology. Artificial Intelligence: Applications in Healthcare Delivery, 273.
Thomas, J., Ledger, G. A., & Mamillapalli, C. K. (2020). Use of artificial intelligence and machine learning for estimating malignancy risk of thyroid nodules. Current Opinion in Endocrinology, Diabetes and Obesity, 27(5), 345-350.
Wang, S., Xu, J., Tahmasebi, A., Daniels, K., Liu, J. B., Curry, J., ... & Eisenbrey, J. R. (2020). Incorporation of a Machine Learning Algorithm With Object Detection Within the Thyroid Imaging Reporting and Data System Improves the Diagnosis of Genetic Risk. Frontiers in Oncology, 10, 2481.
李铃睿, 杜博, & 陈创. (2020). 人工智能在甲状腺癌精准化诊疗中的研究进展.
For up to date citing articles click here