Detecting Aggressive Papillary Thyroid Carcinoma Using Hyperspectral Imaging and Radiomic Features.

Area of Science:

• Oncology
• Medical Imaging
• Computational Pathology

Background:

• Accurate assessment of tumor malignancy is crucial for effective cancer treatment.
• Papillary thyroid carcinoma (PTC) requires precise risk stratification for optimal patient management.
• Novel imaging techniques are needed to enhance diagnostic accuracy in surgical pathology.

Purpose of the Study:

• To investigate the utility of hyperspectral imaging (HSI) combined with radiomics for predicting tumor aggressiveness in ex-vivo papillary thyroid carcinoma (PTC) specimens.
• To identify specific radiomic features capable of differentiating between tumor risk levels.
• To evaluate the potential of HSI-based radiomics as a decision-making tool for oncologists.

Main Methods:

• Extraction of 107 unique radiomic features from fresh, ex-vivo PTC tissue specimens using hyperspectral imaging.
• Analysis of 72 tissue specimens from 44 patients with pathology-confirmed PTC.
• Utilizing dilated hyperspectral images to identify predictive shape features, specifically the least axis length.

Main Results:

• The shape feature 'least axis length' extracted from dilated hyperspectral images demonstrated high accuracy in predicting tumor aggressiveness.
• Radiomic features derived from HSI show promise in assessing tumor risk.
• The study successfully identified a potential imaging biomarker for PTC risk stratification.

Conclusions:

• Hyperspectral imaging-based radiomics offers a promising, non-invasive approach for evaluating PTC aggressiveness.
• The 'least axis length' feature is a potential predictor of tumor risk in PTC.
• This HSI-radiomic method can serve as a valuable tool to assist oncologists in clinical decision-making for intermediate to high-risk tumors.