Novel thermal imaging analysis technique for detecting inflammation in thyroid eye disease

J Clin Endocrinol Metab. 2014 Dec;99(12):4600-6. doi: 10.1210/jc.2014-1957.

Abstract

Context: The disease phase in thyroid eye disease (TED) is commonly assessed by clinical investigation of cardinal signs of inflammation and using the clinical activity score (CAS). Although CAS is the current gold standard, the clinical assessment would benefit if a more objective tool were available.

Objective: The aim of this work was to explore the clinical value of a novel thermal imaging analysis technique to objectively quantify the thermal characteristics of the eye and peri-orbital region and determine the disease phase in TED.

Design: This was a cross-sectional study comparing consecutive patients with active TED (CAS ≥ 3/7) attending a tertiary center, with a group of consecutive patients with inactive TED (CAS <3).

Patients: Thermal images were acquired from 30 TED patients, 17 with active disease and 13 with inactive disease.

Interventions: Patients underwent standard ophthalmological clinical assessments and thermal imaging.

Main outcome measures: Five novel thermal eye parameters (TEP) were developed to quantify the thermal characteristics of the eyes in terms of the highest level of inflammation (TEP1), overall level of inflammation (TEP2), right-left asymmetry in the level of inflammation (TEP3), maximum temperature variability across the eyes (TEP4), and right-left asymmetry in the temperature variability (TEP5).

Results: All five TEP were increased in active TED. TEP1 gave the largest accuracy (77%) at separating the two groups, with 65% sensitivity and 92% specificity. A statistical model combining all five parameters increased the overall accuracy, compared to using only one parameter, to 93% (94% sensitivity and 92% specificity). All five of the parameters were also found to be increased in patients with chemosis compared to those without.

Conclusions: The potential diagnostic value of this novel thermal imaging analysis technique has been demonstrated. Further investigation on a larger group of patients is necessary to confirm these results.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Algorithms
  • Cross-Sectional Studies
  • Diagnostic Imaging / methods*
  • Eye / physiopathology
  • Female
  • Graves Ophthalmopathy / complications
  • Graves Ophthalmopathy / diagnosis*
  • Humans
  • Inflammation / diagnosis*
  • Inflammation / etiology
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Temperature
  • Thermography
  • Young Adult