The Evolution of a Malignancy Risk Prediction Model for Thyroid Nodules Using the Artificial Neural Network
Background: Clinically frank thyroid nodules are common and believed to be present in 4% to 10% of the adult population in the United States. In the current literature, fine needle aspiration biopsies are considered to be the milestone of a model which helps the physician decide whether a certain...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Shiraz University of Medical Sciences
2016-01-01
|
| Series: | Middle East Journal of Cancer |
| Subjects: | |
| Online Access: | http://mejc.sums.ac.ir/index.php/mejc/article/view/279/243 |
| Summary: | Background: Clinically frank thyroid nodules are common and believed to be present in 4%
to 10% of the adult population in the United States. In the current literature, fine needle aspiration
biopsies are considered to be the milestone of a model which helps the physician decide whether a
certain thyroid nodule needs a surgical approach or not. A considerable fact is that sensitivity and
specificity of the fine needle aspiration varies significantly as it remains highly dependent on the
operator as well as the cytologist’s skills. Practically, in the above group of patients, thyroid
lobectomy/isthmusectomy becomes mandatory for attaining a definitive diagnosis where the majority
(70%-80%) have a benign surgical pathology. The scattered nature of clinically gathered data and
analysis of their relevant variables need a compliant statistical method. The artificial neural network
is a branch of artificial intelligence. We have hypothesized that conduction of an artificial neural network
applied to certain clinical attributes could develop a malignancy risk assessment tool to help
physicians interpret the fine needle aspiration biopsy results of thyroid nodules in a context composed
of patient’s clinical variables, known as malignancy related risk factors.
Methods: We designed and trained an artificial neural network on a prospectively formed
cohort gathered over a four year period (2007-2011). The study population comprised 345 subjects
who underwent thyroid resection at Nemazee and Rajaee hospitals, tertiary care centers of Shiraz
University of Medical Sciences, and Rajaee Hospital as a level I trauma center in Shiraz, Iran after
having undergone thyroid fine needle aspiration. Histopathological results of the fine needle
aspirations and surgical specimens were analyzed and compared by experienced, board-certified
pathologists who lacked knowledge of the fine needle aspiration results for thyroid malignancy.
Results: We compared the preoperative fine needle aspiration and surgical histopathology
results. The results matched in 63.5% of subjects. On the other hand, fine needle aspiration biopsy
results falsely predicted malignant thyroid nodules in 16% of cases (false-negative). In 20.5% of
subjects, fine needle aspiration was falsely positive for thyroid malignancy. The Resilient back
Propagation (RP) training algorithm lead to acceptable accuracy in prediction for the designed artificial
neural network (64.66%) by the cross- validation method. Under the cross-validation method, a back
propagation algorithm that used the resilient back propagation protocol - the accuracy in prediction
for the trained artificial neural network was 64.66%.
Conclusion: An extensive bio-statistically validated artificial neural network of certain clinical,
paraclinical and individual given inputs (predictors) has the capability to stratify the malignancy risk
of a thyroid nodule in order to individualize patient care. This risk assessment model (tool) can virtually
minimize unnecessary diagnostic thyroid surgeries as well as FNA misleading. |
|---|---|
| ISSN: | 2008-6709 2008-6687 |