Objective differential classification of thyroid lesions by nuclear quantitative assessment

Area of Science:

• Pathology
• Oncology
• Medical Imaging

Background:

• Accurate diagnosis of thyroid lesions is crucial for effective treatment.
• Distinguishing between follicular adenomas, follicular carcinomas, and papillary carcinomas can be challenging.
• Quantitative nuclear parameters offer potential for improved diagnostic accuracy.

Purpose of the Study:

• To evaluate the diagnostic efficiency of quantitative nuclear parameters in differentiating thyroid lesions.
• To assess the utility of morphometric and stereological methods combined with discriminant analysis for thyroid pathology.
• To investigate nuclear size and form factor differences among follicular adenomas, follicular carcinomas, and papillary carcinomas.

Main Methods:

• Image analysis of thyroid tissue samples from 55 patients.
• Estimation of quantitative nuclear parameters (size and form) using morphometric and stereological techniques.
• Application of stepwise discriminant analysis for lesion classification.

Main Results:

• Nuclear size increased progressively from follicular adenomas to follicular carcinomas, and further to papillary carcinomas.
• Three-dimensional nuclear volume estimates showed more significant size increases.
• Nuclear form factors did not differ between adenomas and carcinomas, but elongation and irregularity increased in papillary tumors (p < 0.01).
• Overall diagnostic accuracy was 75%, with 85% efficiency for papillary carcinoma detection.

Conclusions:

• Quantitative nuclear parameters, particularly nuclear volume, are valuable for discriminating thyroid lesions.
• Morphometric and stereological analysis combined with discriminant analysis provides a robust method for thyroid lesion diagnosis.
• While discrimination between follicular adenomas and carcinomas was less efficient (69%), papillary carcinomas were identified with high accuracy.