MedicineLakex

Thermografie-centrum.nl

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier's archiving and manuscript policies are
encouraged to visit:

Author's personal copy
Infrared Physics & Technology 52 (2009) 42–47 Contents lists available at ScienceDirect Infrared Physics & Technology Normalized methodology for medical infrared imaging J.V.C. Vargas a,*, M.L. Brioschi a, F.G. Dias b, M.B. Parolin c, F.A. Mulinari-Brenner d, J.C. Ordonez e, D. Colman c a Programa de Pós-Graduação em Engenharia Mecânica, Departamento de Engenharia Mecânica, Universidade Federal do Paraná, Curitiba, PR, 81531-980, Brazilb Programa de Pós-Graduação em Engenharia, Universidade Federal do Paraná, Curitiba, PR, 81531-980, Brazilc Transplante Hepático, Hospital de Clínicas da Universidade Federal do Paraná, Curitiba, Paraná, 80060-160, Brazild Departamento de Medicina Interna, Universidade Federal do Paraná, Curitiba, Paraná, 80060-160, Brazile Center for Advanced Power Systems and Department of Mechanical Engineering, Florida State University, Tallahassee, FL 32310, USA A normalized procedure for medical infrared imaging is suggested, and illustrated by a leprosy and hep- Received 7 August 2008 atitis C treatment follow-up, in order to investigate the effect of concurrent treatment which has not been Available online 7 December 2008 reported before. A 50-year-old man with indeterminate leprosy and a 20-year history of hepatitis C wasmonitored for 587 days, starting from the day the patient received treatment for leprosy. Standard ther- apy for hepatitis C started 30 days later. Both visual observations and normalized infrared imaging were conducted periodically to assess the response to leprosy treatment. The primary end points were effec- Skin thermography tiveness of the method under different boundary conditions over the period, and rapid assessment of the Indeterminate leprosy response to leprosy treatment. The patient achieved sustained hepatitis C virological response 6 monthsafter the end of the treatment. The normalized infrared results demonstrate the leprosy treatment suc-cess in spite of the concurrent hepatitis C treatment, since day 87, whereas repigmentation was visuallyassessed only after day 182, and corroborated with a skin biopsy on day 390. The method detected theeffectiveness of the leprosy treatment in 87 days, whereas repigmentation started only in 182 days. Hep-atitis C and leprosy treatment did not affect each other.
Ó 2008 Elsevier B.V. All rights reserved.
consists of the average skin temperature at the affected re-gion ± the standard deviation of the measurements [13], (ii) a com- According to Wallace [1], medical thermography was conceived pound thermographic index (CTI) in correlation with low density and first applied to patients by Lawson [2] in Montreal, and lymphocytes (LDL) [14,15], (iii) a normalized thermographic index reported on two patients with breast cancers, showing increased (DTn), which consists of the evaluation of the TI at the affected re- temperature over the tumor site. Williams et al. [3,4] reported on gion minus the TI of a normal selected region [16], and (iv) a ds in- an infrared survey of 200 cases of breast cancer, also detecting dex (difference between temperatures of tissues of a breast tumor increased temperature on the affected skin regions. These studies and normal tissues) [17]. Inoue et al. [18] evaluated patients with might be considered the starting points of medical infrared rheumatoid arthritis after 20 min of thermal stabilization in a room thermal imaging.
at 20 °C showing that the HDI results correlated better with clinical In an attempt to provide objective means to analyze skin sur- observations than the TI.
face temperature readings, Collins et al. [5–7] developed the ther- An alternative to medical infrared imaging is the so called skin- mographic index to quantify infrared thermal imaging. The method contact thermography. Recently, a wearable device for skin-con- was then utilized in several studies with different disorders [8–12].
tact thermography using silicon sensors was studied by Giansanti All studies showed the effect of medications on the thermographic [19], who designed a cubic–spline interpolation procedure to im- index, which returned to basal levels with the symptoms remis- prove the spatial resolution of the device. The methodology does sion. In general, the results established the concept of a stable envi- not compare to infrared imaging in terms of spatial resolution, ronment as one of a number of essentials to reliable technique.
but it is required when subjects need to be monitored for 24 h, Other methods for medical infrared imaging quantification have since the infrared camera cannot be affixed to a body segment or been proposed, such as: (i) a heat distribution index (HDI), which to a breast for 24 h.
The previously proposed methods for medical thermography (TI, HDI, CTI, DTn, ds, and skin-contact thermography) measure lo-cal temperatures that are significantly affected by ambient temper- * Corresponding author. Tel.: +55 41 3361 3307; fax: +55 41 3361 3129.
ature and patient metabolism. A way to normalize temperature E-mail address: [email protected] (J.V.C. Vargas).
1350-4495/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.infrared.2008.11.003

Author's personal copy
J.V.C. Vargas et al. / Infrared Physics & Technology 52 (2009) 42–47 area of the affected region, m2 thermographic index, °C compound thermographic index, °C central body temperature, °C difference between temperatures of tissues of a breast ambient temperature, °C tumor and normal tissues, °C cartesian coordinates, m heat distribution index, °C low density lymphocytes multidrug therapy normalized thermographic index, °C rifampin, ofloxacin, and minocycline single dose inde- dimensionless temperature, Eq. (2) terminate leprosy treatment average dimensionless temperature, Eq. (4) affected region domain, Fig. 1 skin surface temperature at a point, °C readings for any specific location would therefore be of much use in medical practice. Normalized infrared imaging could provide means for early and accurate detection of the response to treat- The dimensionless temperature defined by Eq. (1) is expected to ment, mainly when other treatments are conducted simulta- deliver normalized temperature readings, independently of mea- neously due to co-infections. The literature reports a relatively suring units, for any particular skin location, whatever body and high prevalence of hepatitis C combined with leprosy in several ambient temperatures are registered. The dimensionless tempera- countries [20,21], but, within the knowledge of the authors, does ture, as defined by Eq. (1), was first introduced in engineering by not document the resulting effect of combined peginterferon Pohlhausen [23], who used it to present the normalized tempera- alfa-2a and ribavirin on leprosy patients undertaking concurrent ture profile solution to the thermal boundary layer problem of lam- multidrug therapy (MDT) and vice versa.
inar forced convection on a flat plate.
In addition to the definition of a local measuring quantity, it is 1.1. Objectives of the paper necessary to specify a representative quantity for the skin regionof interest, namely, the affected region. Fig. 1 shows a portion of Two objectives were sought in this work: (i) to suggest a nor- the affected skin, in which a polygonal line defines a domain X malized methodology for general thermography interpretation, with respect to two Cartesian axes x and y. The polygonal line and (ii) to test the methodology in the treatment follow-up of an should be appropriately specified to encompass the entire region indeterminate leprosy and hepatitis C case, for early and accurate of interest. Each region provides a dimensionless temperature field, detection of leprosy skin lesion evolution.
which depends on x and y, i.e., h(x, y). Using the mean value theo-rem for integrals, the average dimensionless temperature for the 2. Materials and methods entire region of interest is therefore evaluated by: The methodological flow was divided in three steps:  The development of the theoretical methodology for normalized Eq. (2) defines the quantity to be obtained through the infrared infrared imaging.
camera temperature readings in the entire selected region of inter-  The experimental procedure for temperature data acquisition est, and through the measured central body and ambient and calculation of uncertainties, and  The performance evaluation of the suggested normalized infra- red imaging methodology with a case study of concurrent hep- 2.2. Temperature measurements atitis C and indeterminate leprosy.
The infrared images were obtained using a SAT-S160 Infrared camera, manufactured by SAT (Guangxhou SAT Infrared Technol- The energy conservation principle [22] states that, for any time interval, the variation of the energy of a system results from the ex-change of matter and energy with the exterior, and the system en-ergy variation is equal and opposite to the external world energyvariation. Considering the system as a portion of an individual skin(the region of interest), its energy variation is a result of the ex-change of matter and energy with the ambient and the rest ofthe individual's body. Therefore, since energy is directly relatedto temperature, a general methodology to interpret the tempera-ture readings obtained from an individual skin should considerthe local environmental conditions and the individual metabolism.
An appropriate dimensionless variable was identified to inter- pret the infrared camera temperature readings. The variable is awell known dimensionless temperature in engineering heat trans-fer, and combines the locally measured temperature with the cen-tral body and ambient temperatures, as follows: Fig. 1. The skin region of interest, namely, the affected region.
Author's personal copy
J.V.C. Vargas et al. / Infrared Physics & Technology 52 (2009) 42–47 ogy CO., LTD), with a spatial resolution of 2.2 mrad, 160  120 pix- simultaneously and directly calculate the region of interest average els, a bias limit of ±2% (temperature accuracy), and a temperature resolution of 0.1 °C. The camera was placed at a horizontal distance Before the beginning of the experiments, several tests were per- of 1 m away from the region of interest on the patient's skin in or- formed with three different ambient temperatures in an environ- der to ensure the same view factor and dimensions for the selected mentally controlled laboratory, i.e., T1 = 15.5, 20 and 26 °C. The region of interest in all thermographs. Although the literature tests consisted of, for the same individual, to perform the data shows discrepancies for the values of human skin emissivity, it acquisition of the temperature field on the selected region of inter- was possible to attribute an emissivity of 0.975 to the human skin est and body temperature. Then, the value of  h was computed for surface based on relatively recent work [24–26].
these three different conditions, resulting in approximately equal The average of the measurements taken with four high preci- h for all cases. Uncertainties were computed based on sion thermistors of type YSI44004 (YSI Inc., USA), standard type these three values, taken as the precision limit of the temperature Bead I, with a maximum diameter of 2.4 mm, was utilized to esti- measurements (two times the standard deviation [28]), since the mate the ambient temperature close to the patient in the testing temperature bias limit (±2% of the actually measured temperature room, T1. A fifth and a sixth thermistor were placed under the with the infrared camera and ±0.001 °C with the thermistors) was two armpits of the individual being tested, and the average of considered negligible in presence of the temperature precision lim- the measurements was used to estimate the patient's central body it (at least one order of magnitude smaller). In this way it was dem- temperature, Tb ± 0.001 °C. All tests started at least ten minutes onstrated that the dimensionless group really normalizes the after the body temperature thermistors were placed, in order to results, making them independent of the ambient and body tem- read the actual body temperature of the subject.
perature conditions. For the sake of accuracy, in the present study, The thermistors were immersed in a constant temperature bath, the tests were all performed within the ambient temperature and sixty-four temperature measurements were made at 20 °C, range mentioned above, i.e., 15.5 °C 6 T1 6 26 °C.
30 °C, 40 °C, . . , 80 °C. The largest standard deviation of these mea-surements was 0.0006 °C, therefore the bias limit was considered ±0.001 °C for all the thermistors.
The experimental work involved the acquisition of temperature A 50 year-old man was referred to an outpatient clinic for treat- data in real time. This task was performed through the utilization ment of a single lesion with the approximate shape of an ellipse of a computational data acquisition system which consisted of a (major axis: 6 cm, minor axis: 4 cm) in the cervical region, as digital multimeter board, NI PCI-4060, a NI PCI-6703 analog output shown in Fig. 1. Microscopy was conducted on a rectified lamellar board, a SCXI-1127 32-channel high voltage multiplexer, and and epidermis corneal layer, with preserved thickness, measuring accessories, all manufactured by National Instruments, USA, which 0.3  0.3  0.3 cm, collected from the affected region. The superfi- allows for the sequential data acquisition from 32 channels at cial and deep dermis presented moderate lymphocyte inflamma- interval times of 0.1 s. All the data were processed by a home made tory infiltrate, with plasmocytes, predominantly perivascular and Labview application (National Instruments, USA) to convert the perineural, sometimes interstitial, with the presence of one acid- sensors signals in readable temperatures. In this way, the temper- fast bacillus. Therefore the patient was diagnosed with indetermi- ature measurements originated from the six thermistors utilized in nate leprosy. Because of his past history of an untreated chronic the experiments were read almost simultaneously, i.e., in less than hepatitis C and evidence of elevated liver enzymes and positive a second. The program also generated files with the numerical data qualitative HCV–RNA (Cobas Amplicor Hepatitis C Virus Test, ver- measured for the patient, and allowed the calculation of the aver- sion 2.0, Roche Molecular Systems, Branchburg, NJ, USA), a liver age dimensionless temperature according to Eq. (2), with the re- biopsy was performed before starting leprosy treatment. The liver gion of interest temperature field data obtained with the infrared specimens showed moderate necroinflammatory activity and ad- camera and a bi-dimensional numerical integration procedure vanced fibrosis (METAVIR stage F3). HCV genotyping was deter- (Simpson's rule [27]). With the equipment described in this sec- mined as 3a (Inno LipA HCV II, Innogenetics, Ghent, Belgium).
tion, it was possible to measure all required temperatures almost The patient was then monitored for 587 days after receiving a single dose of 600 mg of rifampin, 400 mg of ofloxacin, and100 mg of minocycline, i.e., the so called ROM treatment for inde- terminate leprosy. Thirty days later, he started the current stan-dard therapy for hepatitis C [29,30], with 180 lg peginterferon alfa-2a by subcutaneous injection once a week, plus 1000 mg oralribavirin daily for 24 weeks, achieving sustained virological re- sponse 6 months after the end of the treatment, and thereafter (negative quantitative HCV–RNA). Blood tests were utilized to fol-low the response to the hepatitis C treatment, and both visual and normalized infrared imaging were conducted periodically (0, 15,28, 46, 59, 73, 87, 144, 182, 204, 257, 367 and 587 days) to assess the response to leprosy treatment. On day 390, a biopsy was con- ducted on a punch of pink–red skin tissue extracted from thescarred repigmented area, which was soft and elastic, measuring 0.3  0.3  0.3 cm, showing that the epidermis had no particulari-ties, and that in the mid and deep dermis there was disorganiza- tion of thickened collagen groups, with the Ziehl–Neelsen staining method showing no evidence of acid-fast bacilli. The pri-mary end points were the effectiveness of the normalized temper- Fig. 2. The patient's response to hepatitis C treatment in time through blood tests: ature readings under different boundary conditions over the test —h— Fastingglucose [102 mg/dl]; period, and the rapid assessment of the response to the leprosy Platelet count [1011/l]; Erythrocyte count treatment through normalized infrared imaging, while undergoing Hematocrit [10]; Leucocyte count [109/l]; Absolute neutrophil count [109/l].
simultaneous hepatitis C treatment.

Author's personal copy
J.V.C. Vargas et al. / Infrared Physics & Technology 52 (2009) 42–47 hot and cold spots, or asymmetries; (ii) static images on occasionsseparated by substantial length of time to monitor disease progress A recent study [31] still points out the need for an environmen- or treatment or to detect and estimate the severity of inflamma- tally controlled laboratory and a protocol to ensure consistency for tion, and (iii) series of images over a period of minutes to monitor comparing infrared images captured on different occasions (e.g., the recovery of skin temperature following a provocation such as patient thermal equilibration with the ambient, absence of skin thermal, mechanical, or chemical stress. The proposed normalized creams). The protocol recommends the thermal equilibration of a methodology has the potential to simplify and give more objectiv- patient at a temperature too high for vasoconstriction to occur, ity to infrared imaging protocols in all modes, thus improving the but low enough not to mask the effects of inflammation or angio- accuracy of image medical interpretation.
genesis, which adds a considerable deal of subjectivity to the exam, The experimental results are shown in Figs. 2–4. The experi- since no specific temperature levels are established. In time, pro- mentally measured points were interpolated with cubic–splines, gress has been made on producing open systems [32,33] for cap- for a better visualization of the physical phenomena. The results are organized in a logical sequence, i.e.: (i) regular blood tests to thermal images relating a patient, suggesting three modes of image assess the response to hepatitis C treatment; (ii) the physical and treatment: (i) single static image at an instant in time to identify infrared images comparison of the monitored leprosy skin lesion, Fig. 3. The correspondence between actual photos and infrared images of the leprosy skin lesion captured on different occasions during treatment follow-up. From top tobottom: day 0, day 59, day 87, and day 257.
Author's personal copy
J.V.C. Vargas et al. / Infrared Physics & Technology 52 (2009) 42–47 the same value ðh  2%Þ on the day the image was taken, showingthat the value was indeed normalized in the range of tested ambi-ent temperatures. The procedure was repeated in different occa-sions (0, 15, 28, 46, 59, 73, 87, 144, 182, 204, 257, 367 and 587 days). The 2% uncertainty was the highest value obtained in allmeasurements, therefore it was the value used to compute the er-ror bars shown in Fig. 4, where the results are shown graphically. It is observed clearly that after the single dose of the leprosy ROMtreatment the average dimensionless temperature started to drop,stabilizing at a lower level after day 87, approximately, and remaining at that plateau until the end of the observation period.
The normalized infrared results therefore demonstrate the favor-able leprosy treatment outcome in spite of the concurrent hepatitisC treatment, since day 87, whereas repigmentation was visually as- sessed only after day 182, and corroborated with a skin biopsy on day 390. Since the hepatitis C treatment causes a significant drop in leucocyte and absolute neutrophil count, it was not obvious Fig. 4. The patient's average dimensionless temperature response at the skin whether or not the leprosy treatment outcome would be affected affected region in time.
by that. Although more concurrent leprosy and hepatitis C casesneed to be studied, in the absence of other documented cases,the single case herein studied represents a first assessment that and (iii) the skin lesion average dimensionless temperature in hepatitis C and indeterminate leprosy treatments did not affect Fig. 2 presents a selection of the most representative blood tests monitored during the hepatitis C treatment and the remaining per-iod of observation. Abnormal liver enzyme levels (AST and ALAT) 4. Discussion and conclusions were present in the beginning of the hepatitis C treatment (day30), which dropped initially, elevated again (day 100), later show- In this study, it was proposed and investigated the utilization of ing a trend to normal ranges (day 180), i.e., the treatment end an average dimensionless temperature for infrared imaging analy- point, and indeed stabilizing at normal levels afterwards. Other sis. For that, the methodology was theoretically presented and ap- tests (platelet, erythrocyte, hematocrit, leucocyte, and absolute plied to one case study of indeterminate leprosy and hepatitis C.
neutrophil) showed substantial drop during treatment, and stabi- The bulk of the experimental results showed that normalized infra- lizing at normal levels after the end of the treatment. Those results, red imaging was able to detect the effectiveness of the leprosy together with the negative quantitative HCV–RNA demonstrate the treatment in 87 days, in spite of variable environmental conditions, hepatitis C treatment success in spite of concurrent leprosy whereas repigmentation started visually only in 182 days. Com- bined peginterferon alfa-2a and ribavirin did not affect the patient Actual photos and infrared images are shown in Fig. 3. The response to indeterminate leprosy treatment and vice versa. How- infrared images were obtained with the ambient temperature at ever, for a better assessment of the effects of the leprosy and hep- 20 °C, and the central body temperature was 36.5 °C in all occa- atitis C medications on each other, a larger sample of patients sions to ensure consistency for comparing images captured on dif- being treated simultaneously for both conditions should be ferent days. Four occasions were selected to compare the direct visual and infrared image assessments of the region of interest.
The key conclusion is that with a normalized methodology, On day 0, i.e., when the ROM single dose was taken by the patient, there is potential for production of analysis criteria for high resolu- an elevated temperature distribution was observed on the skin le- tion infrared imaging temperature readings for the diagnosis and sion, in comparison with the cervical region around it. Since there follow-up of skin lesions in leprosy treatment, independently of is immunocytochemical evidence that the initial damage in leprosy central body and local environmental temperatures, and possibly is directed at distal, small, unmyelinated nerve fibers, peripheral to all other skin-detectable pathologies, with no need for an envi- autonomic dysfunction is expected, resulting in local hyperemia ronmentally controlled laboratory and complex protocols.
and elevated skin temperature [34]. On day 15, it was seen quali-tatively that the temperature distribution on the region of interest dropped to lower values, closer to the surrounding normal skin val-ues, but the lesion remained pale. On day 87, the infrared image Additional research has been planned to show the effectiveness shows even lower values than the previous ones, but no lesion of the normalized thermography methodology for diagnosis and repigmentation was observed. On day 257, the temperature distri- follow-up of all forms of leprosy and possibly other skin-detectable bution was very similar to the previous one, and repigmentation pathologies, and to produce comprehensive dimensionless temper- was also in progress. In sum, although a stabilized and fairly uni- ature analysis criteria for them.
form temperature distribution was observed from day 87, the skinlesion did not show signs of repigmentation until day 182, which Conflicts of interest statement was the day when the skin repigmentation process was detectedin the laboratory (not shown in Fig. 3).
We declare that we have no conflict of interests.
Although Fig. 3 documented the decrease in temperature values in the region of interest qualitatively, it would be more instructive for clear medical interpretation if precise quantitative informationwere available. For that, the region of interest average dimension- The authors acknowledge with gratitude the support of the Bra- less temperature was calculated at 3 different ambient tempera- zilian National Council of Scientific and Technological Develop- ture levels, i.e., T 1 = 15.5,20 and 26 °C, obtaining approximately Author's personal copy
J.V.C. Vargas et al. / Infrared Physics & Technology 52 (2009) 42–47 [18] K. Inoue, J. Nishioka, T. Kobori, Y. Nakatani, S. Hukuda, The use of thermographic index and the heat distribution index, Ryumachi 30 (5) [1] J.D. Wallace, C.M. Cade, Clinical Thermography, CRC, Cleveland, OH, 1975.
(1990) 356–361.
[2] R.N. Lawson, Implications of surface temperatures in the diagnosis of breast [19] D. Giansanti, Improving spatial resolution in skin-contact thermography: cancer, Can. Med. Assoc. J. 75 (1956) 309–315.
comparison between a spline based and linear interpolation, Med. Eng. Phys.
[3] K.L. Williams, F.L. Williams, R.S. Handley, Infrared radiation thermometry in 30 (2008) 733–738.
clinical practice, Lancet 2 (OCT29) (1960) 958–959.
[20] H. Rosa, R. Martins, B. Vanderborght, Short report: association between leprosy [4] K.L. Williams, R.S. Handley, F.L. Williams, F. Infrared thermometry in the and hepatitis C infection: a survey in a region of central Brazil, Am. J. Trop.
diagnosis of breast disease, Lancet 2 (721) (1961) 1378–1379.
Med. Hyg. 55 (1) (1996) 22–23.
[5] A.J. Collins, E.F.J. Ring, J.A. Cosh, P.A. Bacon, Quantitation of thermography in [21] A.C.M. Braga, I.J.M. Reason, E.C.P. Maluf, E.R. Vieira, Leprosy and confinement arthritis using multi-isothermal analysis. I. The thermographic index, Ann.
due to leprosy show high association with hepatitis C in southern Brazil, Acta Rheum. Dis. 33 (2) (1974) 113–115.
Tropica 97 (2006) 88–93.
[6] E.F.J. Ring, A.J. Collins, P.A. Bacon, J.A. Cosh, Quantitation of thermography in [22] I. Prigogine, D. Kondepudi, Thermodynamique, Éditions Odile Jacob, 1999. ch.
arthritis using multi-isothermal analysis. II. Effect of nonsteroidal anti- inflammatory therapy on the thermographic index, Ann. Rheum. Dis. 33 (4) (1974) 353–356.
flüssigkeiten mit kleiner reibung und kleiner wärmeleitung, Z. Angew Math.
[7] A.J. Collins, Anti-inflammatory drug assessment by the thermographic index, Mech. 1 (1921) 115–121.
Acta Thermographica 1 (2) (1976) 73–79.
[24] J. Steketee, Spectral emissivity of skin and pericardium, Phys. Med. Biol. 18 (5) [8] E.F.J. Ring, Thermographic evaluation of calcitonin therapy in Paget's disease of (1973) 686–694.
the tibia, Acta Thermographica 1 (2) (1976) 67–72.
[25] T. Togawa, Non-contact skin emissivity: measurement from reflectance using [9] E.F.J. Ring, Computerised thermography for osteo-articular diseases, Acta step change in ambient radiation temperature, Clin. Phys. Physiol. Meas 10 (1) Thermographica 1 (3) (1976) 166–172.
(1989) 39–48.
[10] E.F.J. Ring, P.A. Bacon, Quantitative thermographic assessment of inositol [26] B.F. Jones, A re-appraisal of the use of infrared thermal image analysis in nicotinate therapy in Raynaud's phenomena, J. Int. Med. Res. 5 (4) (1977) 217– medicine, IEEE Trans. Med. Imaging 17 (6) (1998) 1019–1027.
[27] D. Kincaid, Cheney, Numerical Analysis Mathematics of Scientific Computing, Springer, Berlin, 1991.
corticosteroid injection in rheumatoid arthritis, Ann. Rheum. Dis. 40 (2) [28] Editorial. Journal of heat transfer policy on reporting uncertainties in (1981) 132–135.
experimental measurements and results, ASME J. Heat Transfer 115 (1993) [12] S. Zieniuk, B. Stawarz, K. Zieniuk, Infra-red thermography testicular studies of patients with short spermatic cords and quality of sperm morphology, [29] D.B. Strader, T. Wright, D.L. Thomas, L.B. Seef, Diagnosis, management and Ginekol. Pol. 69 (6) (1998) 551–554.
treatment of hepatitis C, Hepatology 39 (4) (2004) 1147–1171.
[13] R.S. Salisbury, G. Parr, M. De Silva, B.L. Hazleman, D.P. Page-Thomas, Heat [30] J.L. Dienstag, J.G. McHutchison, American gastroenterological association distribution over normal and abnormal joints: thermal pattern and technical review on the management of hepatitis C, Gastroenterology 130 quantification, Ann. Rheum. Dis. 42 (5) (1983) 494–499.
(1) (2006) 231–264.
[14] G.M. Papadimitriou, P.A. Bacon, N.D. Hall, Circulating activated lymphocytes in [31] B.F. Jones, P. Plassmann, Digital infrared thermal imaging of human skin, IEEE rheumatoid arthritis: a marker of synovial inflammation, J. Rheumatol. 9 (2) Eng. Med. Biol. 21 (6) (2002) 41–48.
(1982) 217–223.
[32] P. Plassmann, E.F.J. Ring, An open system for the acquisition and evaluation of [15] G.J. Alexander, P.A. Bacon, F. Cavalcanti, D.R. Blake, N.D. Hall, Low density medical thermological images, Eur. J. Thermol. 7 (4) (1997) 216–220.
lymphocytes: their relationship to disease activity and to antirheumatic [33] E.F.J. Ring, The historical development of temperature measurement in therapy, Br. J. Rheumatol. 23 (1) (1984) 6–14.
medicine, Infrared Phys. Technol. 49 (2007) 297–301.
[16] H. Warashina, Y. Hasegawa, H. Tsuchiya, S. Kitamura, K. Yamauchi, Y. Tori, M.
[34] A.F. Hoeksma, W.R. Faber, Assessment of skin temperature by palpation in Kawasaki, S. Sakano, Clinical, radiographic, and thermographic assessment of leprosy patients: interobserver reliability and correlation with infrared osteoarthritis in the knee joints, Ann. Rheum. Dis. 61 (9) (2002) 852–854.
thermometry, Int. J. Lepr. Other Mycobact. Dis. 68 (1) (2000) 65–67.
[17] T. Yahara, T. Koga, S. Yoshida, S. Nakagawa, H. Deguchi, K. Shirouzu, Relationship between microvessel density and thermographic hot areas inbreast cancer, Surg. Today 33 (4) (2003) 243–248.

Source: http://www.thermografie-centrum.nl/forms/m_infrared_physics_and_technology.pdf