Detection of cervical lymph node metastasis from oral cavity cancer using a non-radiating, noninvasive digital infrared thermal imaging system

Abstract

This study aimed to evaluate the diagnostic performance of a non-radiating, noninvasive infrared (IR) thermal imaging system in the detection of cervical lymph node metastasis from oral cavity cancer. In this prospective clinical trial, a total of 90 oral cavity cancer patients suspected of having cervical lymph node metastasis underwent IR imaging of the neck prior to neck dissection. Analysis of the IR images was performed by two methods: manual qualitative analysis and automatic analysis by an entropy-gradient support vector machine (EGSVM). The efficacies of the EGSVM-based infrared thermal imaging system and contrast-enhanced computed tomography (CT) were compared by using the Noninferiority Testing. Compared with manual qualitative analysis, the EGSVM-based automatic analysis had a higher sensitivity (84.8% vs. 71.7%), specificity (77.3% vs. 72.7%), accuracy (81.1% vs. 72.2%), positive predictive value (79.6% vs. 73.3%) and negative predictive value (82.9% vs. 71.1%). The EGSVM-based infrared thermal imaging system was noninferior to contrast-enhanced CT (P < 0.05). The EGSVM-based infrared thermal imaging system showed a trend of higher sensitivity, whereas contrast-enhanced CT showed a trend of higher specificity. The EGSVM-based infrared thermal imaging system is a promising non-radiating, noninvasive tool for the detection of lymph node metastasis from oral cavity cancer.

Figure 1

Timeline of the study process.

Figure 2

Manual qualitative analysis of IR imaging. (a) An asymmetric thermographic pattern that includes an elevated surface temperature and a vascular pattern; (b) Increased vascular density with a tortuous vascular morphologic pattern; (c) A unilateral dilated vascular image; (d) A surface temperature difference of over 1 °C.

Figure 3

Structure of the EGSVM system. (a) The image in the black box is the cropped image from raw images, which is converted into a grayscale image; (b) The expanding process of win and feature extraction. The win expands from win₁ to win_n. For win_k, the entropy of $Re\_imag{e}_{\mathrm{k}}$ and $Re\_imag{e}_k^{hist}$ are used in the feature extraction process; (c) The red circles and blue squares represent high-dimensional feature vectors; (d) The structure of automatic analysis system for new raw images.

Figure 4

Representative case correctly confirmed with both IR imaging and contrast-enhanced CT. (a) Elevated surface temperature and increased vascular density on the right side using IR imaging; (b) Metastatic lymph node in contrast-enhanced CT; (c) Microscopic appearance of metastatic lymph nodes (hematoxylin and eosin staining, original magnification ×200).