EQUILIBRATION TIMES FOR DIGITIZED THERMOGRAPHIC EVALUATION
Ed Owens, M.S., D.C., Life College, Marietta, GA
A chief technical concern in conducting thermographic examinations is reliability of the results. Patient equilibration in a temperature controlled environment is essential. Standard thermographic protocol requires a 15 minute equilibration period before the first thermograph is recorded.(1) Three scans are taken of the same area at intervals of 15 to 20 minutes between scans. Each subsequent scan is compared to the previous to decide whether the patient is completely equilibrated. Using these guidelines, the minimum time required for triplicate studies would be 45 minutes.
The object of this study is to determine the length of time needed for a patient's thermographic scan to become stable, not necessarily with respect to overall temperature, but with respect to clinically significant patterns.
METHODS The Visi-Therm II computerized electronic thermography system (Stillwagon Seminars, Inc., 767 Dry Run Road, Monongahela, PA) was used to take a series of twelve thermograms, each three minutes apart on 25 subjects, as they equilibrated to ambient room temperature. The Visi-Therm II is a non-contact electronic infrared thermography system that uses twelve individual optical sensors mounted into a hand-held scanner to measure infra-red emissions in twelve bands up the spine.(2) The scanner is moved by hand up the back, from S2 to the Occiput, and has rollers that are kept in contact with the back to maintain uniform distance from the skin. The results of each scan were stored on floppy disks for later analysis.
Data analysis consisted of combining the serial data files into larger files which could be plotted and investigated using statistical methods. It was very difficult to compare whole scans over time using color changes to represent temperature, since the average temperature changed between scans.
Individual whole back graphs were normalized over time by calculating a color scale based on the warmest spot on the back. This method produced more similar scans, but did not provide a method for determining the amount of change between scans.
For more detailed analysis, programs were written in BASIC on an IBM Compatible computer that could plot bands of data, representing the thermal scan produced as a single sensor was moved up the back. It was then possible to compare adjacent scans on a column by column basis, looking for similarities in the locations of hot and cold deviations in each band.
A special algorithm was generated to shift adjacent scan plots by the average temperature of the scan, then calculate the square of the difference between adjacent scans. The result of this calculation was used as a measure of the amount of change that occurred over time in each thermal scan.
RESULTS The average temperature of the back was found to decrease smoothly from 35 to 32 degrees Celsius during the course of the 33 minute data collection period. A plot of the squared difference between adjacent thermograms with respect to time shows an approximately linear decrease from 6 units to 2 units over the first 12 minutes. After 12 minutes the plots level out, indicating that adjacent data is as uniform as it will get after 12 minutes.
CONCLUSIONS It was found that after 12 minutes, on the average, the scans were very consistent. Even though the patients continued to equilibrate for more than 36 minutes, as evidenced by an overall decrease in temperature, the significant features of the columnar scans were consistent after 12 minutes.
1. Rein, H. Thermographic evidence of soft tissue injuries. Shepard's/McGraw-Hill, Colorado Springs, CO, 1987.
2. Stillwagon, G. Thermography seminar notes. Stillwagon Seminars, Inc., 1984.