Indeterminate thyroid cytology: detecting malignancy using analysis of nuclear images

Abstract

Background: Thyroid nodules diagnosed as “Atypia of Undetermined Significance/Follicular Lesion of Undetermined Significance” (AUS/FLUS) or “Follicular Neoplasm/Suspected Follicular Neoplasm” (FN/SFN)”, according to Bethesda's classification, represent a challenge in clinical practice. Computerized analysis of nuclear images (CANI) could be a useful tool for these cases. Our aim was to evaluate the ability of CANI to correctly classify AUS/FLUS and FN/SFN thyroid nodules for malignancy. Methods: We studied 101 nodules cytologically classified as AUS/FLUS (n=68) or FN/SFN (n=33) from 97 thyroidectomy patients. Slides with cytological material were submitted to manual selection and analysis of the follicular cell nuclei for morphometric and texture parameters using ImageJ software. The histologically benign and malignant lesions were compared for such parameters which were then evaluated for the capacity to predict malignancy using the Classification and Regression Trees Gini model. The Intraclass Coefficient of Correlation was used to evaluate method reproducibility. Results: In AUS/FLUS nodule analysis, the benign and malignant nodules differed for Entropy (p<0.05), while the FN/SFN nodules differed for Fractal analysis, coefficient of variation (CV) of roughness, and CV-Entropy (p<0.05). Considering the AUS/FLUS and FN/SFN nodules separately, it correctly classified 90.0% and 100.0% malignant nodules, with a correct global classification of 94.1% and 97%, respectively. We observed that reproducibility was substantially or nearly complete (0.61-0.93) in 10 of the 12 nuclear parameters evaluated. Conclusion: CANI demonstrated an high capacity for correctly classifying AUS/FLUS and FN/SFN thyroid nodules for malignancy. This could be a useful method to help increase diagnostic accuracy in the indeterminate thyroid cytology.

Digital analyses of nuclear features can help discriminate among non‐invasive follicular thyroid neoplasm with papillary‐like nuclear features, papillary thyroid carcinoma follicular subtype, and follicular carcinoma in cytological specimens

“…We demonstrated the importance of evaluating nuclear parameters using digital pathology, consistent with the findings of studies analysing follicular cells in cytological 9–11,14 and histological 15–17 samples. Studies with histological samples are more common in the literature 15–17 .…”

“…This is consistent with the concept of nuclear size as an important feature for discriminating benign from malignant cells. 13 We demonstrated the importance of evaluating nuclear parameters using digital pathology, consistent with the findings of studies analysing follicular cells in cytological [9][10][11]14 and histological [15][16][17] samples. Studies with histological samples are more common in the literature.…”

“…Studies with histological samples are more common in the literature 15–17 . In cytological samples, digital pathology has been shown to increase the diagnostic accuracy of malignancy in indeterminate nodules 10 and help differentiate malignant from benign nodules because, in the former, the minimum nuclear diameter, maximum nuclear diameter, perimeter, and nuclear area were more significant 9 . These findings, combined with the potential applications of artificial intelligence in digital pathology, offer a promising outlook for improving diagnostic accuracy in thyroid cytology 18 …”

“…Unlike molecular studies, which are expensive and not widely available, digital pathology has emerged as a promising tool; it is cost‐effective, objective, and easily applicable to the evaluation of the nuclear parameters (morphological and morphometric) of follicular cells 9 . Hayashi et al performed a computerised analysis of nuclear images of a cytological sample of indeterminate follicular cells (AUS/FLUS and FN/SFN nodules from the Bethesda classification); the authors found digital pathology to be valuable and effective in the diagnosis of malignancy 10 . Moreover, there are data favouring the basic measurement of nuclear morphological features as a potential tool to assist in the identification of malignant cases, helping in the clinical management of AUS cases 11 …”

Management of Atypia of Undetermined Significant Thyroid Nodule

“…Wide variation of AUS was seen since 42% physicians received benign histo diagnosis after surgery, and 74% received malignant histologic [14] Other modalities such as Computerized analysis of nuclear images (CANI) also reported beneficial in AUS case. Hayashi et al [17] reported sensitivity and specificity in detecting malignancy that higher than 80% of AUS cases with CANI [17].…”