dental radiology Editor: Allan G. Farman, BDS, PhD (Odont), MBA Department of Primary Patient Care University of Louisville School of Dentistry Louisville, Ky. 40292 Electronic thermography of normal facial structures: A pilot study Barton M. Gratt, DDS,a Andrew Pullinger, DDS, 174S,~ Edward A. Sickles, MD,c and J. Jack Lee, DDS, MS,d Los Angeles and San Francisco, Cali$ UCLA SCHOOL OF DENTISTRY AND UCSF SCHOOL OF MEDICINE New electronic thermographic instruments capable of routine clinical examination need to be evaluated for their potential as a diagnostic aid in dentistry. This study assessed thermal symmetry of the face and neck in 20 normal subjects with the use of frontal and lateral views, at 1 .O” C and 0.5” C sensitivity, under controlled conditions. Electronic thermographic images were analyzed for thermal symmetry, by means of a grid matching technique, in 12 anatomic regions and the overall face. Results indicated that thermal symmetry for the entire face was high (70.2%). The 12 specific facial areas demonstrated varying levels of thermal symmetry. Regions of high symmetry on frontal projections included the anterior portion of the neck (82.0%), the TMJ (80.0%), the lower lip (78.8%) and the upper lip (77.39). The temporal region (48.7%) was found to be of relatively low thermal symmetry. Regions of high symmetry on lateral projections included the nasal region (89.5%) and the inframandibular region (67.0%). The posterior neck region (44.2OI0) was found to be of relatively low thermal symmetry. In general, normal subjects demonstrated high levels of thermal symmetry over most regions of the face. This pilot investigation is an early step in the evaluation of electronic thermography for future use in dentistry. (ORAL SURC ORAL MED ORAL PATHOL 1989;68:346-51) R eports from early medical history have indicated the use of body temperature as a diagnostic sign, In 400 B.C., Hippocrates used the palm of his hand to judge the skin temperature of sick patients.’ In 1872, Wunderlich, in Germany, introduced fever measure- ments as a routine procedure. Later, when reliable thermometric instruments were developed, measure- ment of the core temperature of the body became the This study was funded in part from Grant #87-007 from the UCLA Dental Research Institute, The Center for the Health Sciences, Los Angeles, Calif. aProfessor, Section of Oral Radiology, UCLA Dental Research Institute, UCLA School of Dentistry. bAssociate Professor, Section of Gnathology and Occlusion, UCLA Dental Research Institute, UCLA School of Dentistry. cProfessor and Chief, Breast Imaging Section, Department of Ra- diology, UCSF School of Medicine. dSenior Statistician, UCLA School of Dentistry. 7116112654 346 primary procedure for the diagnosis of fever. Since 1951, it has been customary to take daily body tem- perature measurements on all hospital patients. l Long-range medical research has been expanding in the fields of magnetic resonance imaging, scintig- raphy, positron emission tomography scanning, elec- tromyography, and thermography. All of these tech- niques are based at least in part on physiologic pro- cesses,thereby providing complementary information to the anatomic imaging provided by examinations such as conventional film x-ray and computerized to- mography scanning. Thermography (a generic title given to various methods of heat pattern identification and analysis) was first used as a descriptive procedure in diagnosis in the 1960s. There have been many im- provements in thermographic equipment and tech- nique since that time.’ Thermal symmetry is present in many parts of the
Transcript
dental radiology Editor: Allan G. Farman, BDS, PhD (Odont), MBA Department of Primary Patient Care University of Louisville School of Dentistry Louisville, Ky. 40292
Electronic thermography of normal facial structures: A pilot study Barton M. Gratt, DDS,a Andrew Pullinger, DDS, 174S,~ Edward A. Sickles, MD,c and J. Jack Lee, DDS, MS,d Los Angeles and San Francisco, Cali$
UCLA SCHOOL OF DENTISTRY AND UCSF SCHOOL OF MEDICINE
New electronic thermographic instruments capable of routine clinical examination need to be evaluated for
their potential as a diagnostic aid in dentistry. This study assessed thermal symmetry of the face and neck
in 20 normal subjects with the use of frontal and lateral views, at 1 .O” C and 0.5” C sensitivity, under controlled conditions. Electronic thermographic images were analyzed for thermal symmetry, by means of a
grid matching technique, in 12 anatomic regions and the overall face. Results indicated that thermal
symmetry for the entire face was high (70.2%). The 12 specific facial areas demonstrated varying levels of
thermal symmetry. Regions of high symmetry on frontal projections included the anterior portion of the neck (82.0%), the TMJ (80.0%), the lower lip (78.8%) and the upper lip (77.39). The temporal region (48.7%)
was found to be of relatively low thermal symmetry. Regions of high symmetry on lateral projections
included the nasal region (89.5%) and the inframandibular region (67.0%). The posterior neck region
(44.2OI0) was found to be of relatively low thermal symmetry. In general, normal subjects demonstrated high levels of thermal symmetry over most regions of the face. This pilot investigation is an early step in the
evaluation of electronic thermography for future use in dentistry.
(ORAL SURC ORAL MED ORAL PATHOL 1989;68:346-51)
R eports from early medical history have indicated the use of body temperature as a diagnostic sign, In 400 B.C., Hippocrates used the palm of his hand to judge the skin temperature of sick patients.’ In 1872, Wunderlich, in Germany, introduced fever measure- ments as a routine procedure. Later, when reliable thermometric instruments were developed, measure- ment of the core temperature of the body became the
This study was funded in part from Grant #87-007 from the UCLA Dental Research Institute, The Center for the Health Sciences, Los Angeles, Calif. aProfessor, Section of Oral Radiology, UCLA Dental Research Institute, UCLA School of Dentistry. bAssociate Professor, Section of Gnathology and Occlusion, UCLA Dental Research Institute, UCLA School of Dentistry. cProfessor and Chief, Breast Imaging Section, Department of Ra- diology, UCSF School of Medicine. dSenior Statistician, UCLA School of Dentistry. 7116112654
346
primary procedure for the diagnosis of fever. Since 1951, it has been customary to take daily body tem- perature measurements on all hospital patients. l
Long-range medical research has been expanding in the fields of magnetic resonance imaging, scintig- raphy, positron emission tomography scanning, elec- tromyography, and thermography. All of these tech- niques are based at least in part on physiologic pro- cesses, thereby providing complementary information to the anatomic imaging provided by examinations such as conventional film x-ray and computerized to- mography scanning. Thermography (a generic title given to various methods of heat pattern identification and analysis) was first used as a descriptive procedure in diagnosis in the 1960s. There have been many im- provements in thermographic equipment and tech- nique since that time.’
Thermal symmetry is present in many parts of the
Volume 68 Number 3
Electronic thermography of normal facial structures 347
Fig. 1. Facial thermograms of normal subject taken with 1 .O” C thermal sensitivity setting. Each band of color represents a 1 .O” C (temperature) difference. Adjustable temperature scale is set to show black as cold, blue as cool, pink as warm, and red as hot.
body in a normal population. Bassett and coworkers2 reported thermal symmetry in the hands of 19 of 20 normal subjects. In computerized thermographic studies of normal persons, Uematsu3, 4 reported the average temperature differences between correspond- ing sides of the body to be less than 0.3” C. Feldman and Nickolo@ studied the neck and upper extremities of 100 normal factory workers and demonstrated that 94% had symmetric temperature distribution (< lo C difference). Lovisatti and associate@ reported ther- mal symmetry in the lower extremities in 88% of 17 subjects without symptoms. Examinations of 52 per- sons without symptoms by Hubbard’ demonstrated 90% to have symmetric thermal patterns in the arms and legs. Raskin and colleagues8 reported thermal symmetry as the normal pattern in 85 persons in their study of the low back area. Pulst and Haller,9 in their study of peripheral nerve injuries, demonstrated ther- mal symmetry in 15 subjects without symptoms used as controls. Gillstroml” found symmetric tempera- tures in the lower extremities in all of the 17 normal subjects examined in his study. Thus, the literature clearly documents that in the normal situation, blood flow through the skin of most parts of the body pro- duces a symmetric thermal pattern. However, to date
no one has investigated the thermal symmetry of the complex structures of the face. In this study, we an- alyzed in detail the facial thermal patterns of 20 nor- mal volunteers without symptoms.
MATERIALS AND METHODS
Subject selection criteria. Twenty subjects who granted informed consent for study were examined; they consisted of 12 men and 8 women, having a mean age of 24.8 years. Criteria for selection as normal subjects included a negative history for facial trauma, facial pain, headaches, sinus disease/symptoms, and temporomandibular joint (TMJ) symptoms, in addi- tion to the absence of prominent facial scars or blem- ishes. Furthermore, to be considered normal, all sub- jects had to pass a clinical (dental) examination, demonstrating normal dental occlusion (Class I), no TMJ sounds, and no facial or TMJ tenderness on palpation. The target population was normal subjects without facial pain.
All subjects complied with the preexamination procedures recommended by the Academy of Neuro- Muscular Thermography.5
Thermography equipment. Thermography was conducted with the use of one of two electronic units
348 Gratt et al. ORAL SURC ORAL MED ORAL PATHOL September 1989
Fig. 2. Acetate grid overlay used for matching analysis. The sides of each square are 5 mm in size.
Assessment by Grid Matching
Right I Left
Right Left % Matching
VS. = 100
VS. 0
= 50
VS. = 30
vs. I!?!! = 0 .:.:.:. $$v. vs. $g = 30 •B :q:.
Fig. 3. Diagram demonstrating matching techniques used in obtaining percentages of thermal facial sym- metry. The various shades of gray indicate different temperature readings. A square located on the right side of the face is compared to a corresponding square of the opposite side. Perfectly (color) matched squares are scored at lOO%, partially matched squares are scored at from 90% to lo%, and nonmatching squares are scored at 0%.
available at the UCLA School of Dentistry. Subjects were imaged with an Inframetrics 535 unit (Infra- metrics Corp., Bedford, Massachusetts), coupled to a JVC 12-inch television monitor, model TM-90 U
(JVC Corp., Yokohama, Japan), and an AGA 750 thermovision system, including scanner, computer, and monitor (Pharos Corp., Secaucus, New Jersey). Hard-copy images were recorded on 35 mm color
Volume 68 Number 3
Electronic thermography of normal facial structures 349
Fig. 4. Diagram of 12 anatomic zones used for analysis: (1) nasal region, (2) superior labial region, (3) in- ferior labial and mental region, (4) frontal region (forehead), (5) periorbital region, (6) superior buccal region (upper cheek), (7) inferior buccal region (lower cheek), (4) frontal region (forehead), (8) temporal region (temple), (91 TMJ (preauricular region), (10) inframandibular region (upper neck region), (11) anterior neck region, and (12) posterior neck region.
Table I. Percentage of thermal facial symmetry by Table II. Percentage of thermal facial symmetry by anatomic region (frontal views) anatomic region (lateral views)
Region 1.0” C sensitivity 0.5” C sensitivity % matching (SD) % matching (SD) Region
1 .O’ C sensitivity 0.5’ C sensitivity % matching (SD) ) % matching (SD)
slide film. Both thermography units were calibrated to 0.1’ C accuracy before use.
The Inframetrics and AGA units both required liquid nitrogen cooling (Cryogenic, Inc., Division of V. B. Anderson Co., Santa Ana, California). Room conditions for conducting the thermographic exami- nations included a draft-free environment (no win- dows, closed doors), temperature control (68” to 73” F), variable lighting, a swivel chair for patient posi- tioning, and a small hand-held electric fan (General Electric Co., Milwaukee, Wisconsin).
Facial imaging. Color thermograms of the face were taken with right and left lateral projections and a frontal projection, at both 1.0” C and 0.5” C sen- sitivity. Before the examination, each subject’s face was cleared of hair (tied back), and the face was wiped with a damp cloth and then air dried with the small electric fan. Then, after a wait of 20 minutes for fa- cial temperature equilibration, three series of color thermograms were made, 15 minutes apart.
Image analysis. Color slides were enlarged to 5~ 7 inch color prints (Fig. 1) to facilitate subsequent
350 Gratt et al. ORAL SURG ORAL MED ORAL PATHOL September 1989
both 1.0” C and 0.5” C sensitivity settings for eval- uation of the face, until additional research confirms this conclusion.
Ten of the twelve anatomic regions of the face and neck assessed in this study demonstrated primarily symmetric patterns (anatomic areas containing greater than 50% thermal symmetry). The two ana- tomic regions that showed the lowest levels of thermal symmetry, the region of the posterior neck and the temporal region of the forehead (temple), also pro- duced the highest levels of heat emission of the twelve regions studied. We postulate that since both of these regions contain large superficial blood vessels that common anatomic variations in location of these ves- sels may cause the observed low levels of thermal symmetry.
The full scope of the clinical utility of electronic thermography (ET) in dentistry is not clear at this time. ET has already been shown to aid in the diag- nosis of headache,’ l-l3 atypical odontolgia,14 and oral postsurgical evaluation. 1 5 On the other hand, ET does not appear to be of use in the evaluation of dental granuloma. l6 Recent reports 17, l* indicate that tem- perature measurements may be useful in the assess- ment of patients with rheumatoid arthritis of the TMJ. Additional research in these and other areas will help to determine the ultimate role, if any, of ET in dentistry.
In summary, overall thermal symmetry usually is present in the face and neck regions of normal subjects. Areas of highest thermal symmetry include nasal, inframandibular, inferior labial, inferior buc- cal, superior buccal, superior labial, and TMJ regions of the face and neck. The posterior neck and tempo- ral region of the face demonstrated the lowest degree of thermal symmetry. Electronic thermography has potential for use as an alternative diagnostic tech- nique in dentistry.
We thank the Illustration Department in the UCLA School of Dentistry for art work; the Inframetrics Corp., Bedford, Massachusetts, and the Agema Corp., Secaucus, New Jersey, for equipment and technical support; Profes- sor Andrew Dixon, UCLA School of Dentistry, for ana- tomic terminology; and especially Dr. Charles Wexler, M.D., Encino, California, for direction in thermographic techniques.
analysis, conducted with the use of an acetate overlay matrix with 5 mm squares (Fig. 2), which was super- imposed over the facial thermograms. Assessments were made of thermal symmetry by comparing the temperature distribution within each square with the corresponding square of the opposite side of the face (Fig. 3). When two corresponding squares were judged to have exactly the same color patterns (tem- peratures), they were scored as matching. Matching scores (%) were given for each pair of squares rang- ing from a high of 100% (perfect match) to a low of 0% (no match). Matching assessments were made for one half of the face and for 12 anatomic locations on the face (Fig. 4), with the use of both frontal and lat- eral projections at 1.0” C and 0.5” C sensitivity.
RESULTS
The results of the half-face analysis of normal sub- jects’ thermograms indicated a high degree of thermal symmetry. For frontal projections imaged at 1.0” C sensitivity, there was 70.2% symmetry (SD = 26.2%), and for images taken at 0.5” C sensitivity, there was 62.2% symmetry (SD = 30.1%). For lateral projec- tions imaged at 1 .O” C sensitivity, the symmetry score was 61.0% (SD = 30.9%), and for images taken at 0.5” C sensitivity, the score was 59.9% (SD = 24.9%).
Symmetry scores for the 12 anatomic regions of the face are shown in Tables I and II. Levels of thermal symmetry ranged from 82.0% (anterior portion of the neck, frontal views) and 69.5% (nasal region, lateral views) down to 46.7% (temporal region, frontal views) and 44.2% (posterior portion of the neck, lateral views). These results indicate that there are wide variations in facial symmetry and wide ranges in their variability (standard deviation). As expected, ther- mograms imaged with 1.0” C sensitivity demon- strated higher levels of thermal symmetry in both frontal and lateral views than those with 0.5% sensi- tivity.
DISCUSSION
A 100% level of thermal facial symmetry can be obtained with the use of electronic thermography at imaging sensitivities of 4” C and above. However, at these sensitivities, few if any abnormal conditions will be detected because abnormalities usually demon- strate less than a 2.0” C difference from normal values.4 On the other hand, if low sensitivity settings are used (e.g., 0.1’ C), abnormal states may be iden- tified readily but the majority of normal patients would also show significant areas of thermal asym- metry about the face. At this time, it appears that a sensitivity of 1 .O” C is most likely to produce clinically relevant data. However, we recommend the use of
REFERENCES
1. Engel J-M, Flesch U. Thermological methods. Weinheim, Federal Republic of Germany: VCH Verlagsgesellschaft, 1985:i-iv, 1-8.
2. Bassett LW, Gold RH, Clements PJ, Furst D. Hand thermog- raphy in normal subjects and scleroderma. Acta Thermo- graphica 1980;5: l-5.
3. Uematsu S. Thermographic imaging of sensory dermatome. Sot Neuroscience Abstracts 1983;9:324-9.
Volume 68 Number 3
Electronic thermography of normal facial structures 351
4.
5.
10.
11.
12.
Uematsu S. Thermographic imaging of cutaneous sensory segment in patients with peripheral nerve injury. Skin-tem- perature stability between sides of the body. J Neurosurg 1985;62:716-29. Feldman F, Nikoloff EL. Normal thermographic standards for cervical spine and upper extremities. Skeletal Radio1 1984;12:242-9. Lovisatti L, Mora L, Pistolesi GF. Thermographic patterns of lower limb arterial disease. Bib1 Radio1 1975;6:107-14. Hubbard JE. Statistical review of thermography in a neurol- ogy practice. Postgrad Med special edition, March, 1986. Raskin MM, Martinex-Lopez M, Sheldon JJ. Lumbar ther- mography in discogenic disease. Radiology 1976; 119: 149-52. Pulst SM. Haller P. Thermoaraohic assessment of imuaired sympathetic function in peripheral nerve injuries. J Neural 1981;226:35-9. Gillstrom P. Thermography in low back pain and sciatica. Arch Orthop Trauma Surg 1985;104:31-6. Swerdlow B, Dieter JN. The validity of the vascular “cold patch” in the diagnosis of chronic headache. Headache 1986;26:22-5. Kudrow L. A distinctive facial thermographic pattern in clus- ter headache-the “chai” sign. Headache 1985;25:33-6.
13.
14.
15.
16.
17.
18.
Reprint requests to: Dr. Barton M. Gratt UCLA School of Dentistry Los Angeles, CA 90024- 1668
Drummond PD, Lance JW. Thermographic changes in cluster headache. Neurology 1984;34:1292-8. Gratt BM, Sickles EA, Graff-Radford SB, Solberg WK. Elec- tronic thermography in the diagnosis of atypical odontalgia: a pilot study. ORAL SURG ORAL MED ORAL PATHOL (in press, 1989). Amutpueba C, Anekceeba AH. Applications of thermography in evaluation of the postoperative patient. Stomatologiia 1986;12:29-30 (Russian). Crandell CE, Hill RP. Thermography in dentistry: a pilot study. ORAL SURG ORAL MED ORAL PATHOL 1966;21:316-20. Akerman S. Morphologic, radiologic, and thermometric as- sessment of degenerative and inflammatory temporomandibu- lar ioint disease. Swed Dent J Suoul 1987:52:1-110. Kopp S, Haraldson T. Skin surface iemperature over the TMJ and masseter muscle in patients with craniomandibular disor- der. Swed Dent J 1988;12:63-7.
