Logo of the Physikalisch-Technische Bundesanstalt

Optical Medical Imaging

Working Group 8.31

Fluorescence guided rheuma imaging

Development of a novel method detecting inflammatory rheumatic changes in finger joints by NIR fluorescence imaging


Development of a fluorescence based measuring and imaging system for early detection of rheumatoid arthritis. Derivation of yet unknown observables for the quantitative interpretation of the (dynamical) frame sequences for the development of algorithms resulting in discriminators supporting diagnostic decision making.

Relevance of rheumatoid diseases

Rheumatoid arthritis has a prevalence of about 2% and is one of the most common diseases in Germany. Currently, about 1.3 million people are affected by rheumatoid arthritis. Furthermore, 35000 adults and 500 to 900 children and teenagers under 16 years come along every year. Economically, the high incidence of the disease is often leading to a high non-productive time as well as high costs for therapy and rehabilitation have to be considered.

An early and appropriate treatment may stop the disease for a long time. Consequently, quality of life can be maintained and the follow-up costs can be minimized if the disease is diagnosed as early as possible. Obviously a cost-efficient, low impact method for early diagnosis is desirable. As the pharmaceuticals for therapy of rheumatoid arthritis may have severe side-effects, the detection method should also provide information on the success of the chosen therapy.


Metrological aspect

The metrological task with respect to fluorescence measurements and imaging is to assure the comparability of measurements intra-individually between different joints of one subject as well as inter-individually between subjects for uniform evaluation of rheumatoid arthritis. The comparability is achieved by the development and investigation of different normalizing procedures. The quantification of the fluorescence intensity by referencing to fluorescent scattering materials or LED based phantoms represents a fundamental aspect in securing accuracy of measurement in clinical applications.

Another metrological challenge is the identification and definition of new dynamical measurands, e.g. to describe the recorded kinetics of a given contrast agent or the spatial-temporal enhancement of the dye, which is suited for an objective diagnosis or even a quantitative clinical monitoring.

To top

The detection methode

Near-Infrared Fluorescence Imaging

Near-infrared imaging (NIR imaging) uses light in the spectral range from 650 nm to 1000 nm for spectral discrimination of tissue. Compared with visible light, NIR light is less absorbed by endogenous chromophores and less scattered by tissue. Thus the penetration depth of NIR photons reaches up to several centimeters depending on the scattering and absorption coefficient. NIR imaging is well suited for non-invasive diagnostics because the involved NIR irradiance is harmless for the human body and  screening is possible.

Intrinsic optical parameters like absorption and scattering coefficients often permit no sufficient discrimination between healthy and diseased tissue. Advantageously fluorescent contrast agents are used. Until now no commercial detection method takes advantages of NIR fluorescence imaging for early detection of rheumatoid arthritis.


Near-Infrared Contrast Agents

Cyanine dyes of the class of the indotricarbocyanines with an absorption and fluorescence emission in the spectral range from 750 nm to 850 nm are well-suited as optical contrast agents. Since the 1970s Indocyanine Green (ICG) has already been used for cardiovascular diagnostics and hepatic functional tests. By the synthesis of derivatives with modified functional groups, in particular carboxyl groups, so-called NIR diagnostic agents have been created, in which both, molecular weight and the degree of hydrophilicity can be varied systematically.



In the current project, which is based on the results of a previous project for diagnosis of rheumatoid arthritis supported by the Investitionsbank Berlin (IBB) and co-funded by the European Comission (ERD fund), an imaging equipment has been developed, where fluorescence emission is imaged from two hands simultaneously. The detection of consecutive frames enables the kinetic evaluation of the dye bolus. A frame rate of 4 images/s has been realized to account for the influence of the heart beat. Two different imaging procedures have been investigated.

  1. 1. Broad beam excitation of fluorescence radiation with an expanded cw laser beam and detection of fluorescence emission with a sensitive electron multiplying CCD (EMCCD) camera.
  2. Point by point excitation of tissue by scanning with a collimated cw laser beam and detection of fluorescence using an avalanche photodiode.

For both imaging principles demonstrator-prototypes were build. They were compared with respect to detection sensitivity, detector linearity, measurement dynamic, spatial resolution and homogeneity of illumination. The demonstrators are approved for use in clinical environment.

To top


The origin for establishment of this novel procedure to detect rheumatoid arthritis were trials in animal models. Therefore, the dye ICG was intravenously administered to mice with lyme arthritis and the enrichment of ICG was imaged at different times. Figure 1 shows a typical result. Plotting the fluorescence intensity of the ankle joints over time yields characteristic plots for discrimination between healthy and diseased animals (Figure 2). Details can be found Fischer et al. 2006.

  Fluoreszenzbilder in Falschfarbendarstellung bei einer gesunden und einer erkrankten Maus

Figure 1: Fluorescence images in false color representation of an a) healthy and b) diseased mouse 6 s, 20 s, 60 s, and 120 s after intravenous application of ICG. The fluorescence intensities are related to an external reference to make images of different animals comparable.

  Zeitliche Entwicklung der Fluoreszenzintensität im Sprunggelenk von arthritischen und gesunden Mäusen nach intravenöser Gabe von ICG.

Figure 2: Temporal development of the fluorescence intensity in ankle joints (averaged over 10 ankle joints) of arthritic and healthy mice after intravenous administration of ICG. Each curve is normalized to its maximum. The accuracy of measurement due to the biological variability (inter-individual variation) is indicated by the standard deviation, shown for one point only for the sake of clarity.


Figure 3: Fluorescence images of hands after an intravenously administered bolus of ICG. 2-hand-imager examples (upper row from left to right) for a test subject without pathological findings, a patient with rheumatoid arthritis and one with Rhiz-arthrosis. 1-hand-scanner examples (lower row from left to right) for a test subject without pathological findings, a patient with rheumatoid arthritis and one with osteoarthrosis.

The developed imaging systems have been successfully used in a clinical trial. The two above mentioned approaches have been applied, the laser based scanner and the EMCCD imager system. Both approaches provided the evidence that the IR-fluorescence of an unspecific dye can be used to monitor rheumatoid arthritis in finger joints (see Fig. 3).

Preliminary results of the detailed data analysis are very promising and allow a good discrimination between healthy and inflamed joints. There are indications so far, that the scanning system may have a better discriminative power than the imager.

The development of the 2-hand-imager and the clinical study are supported by the Investitionsbank Berlin (IBB) and co-financed by the European Regional Development Fund (EFRE).  

To top


Wunder A, Ebert B,
Rheumatology: Noninvasive Fluorescence Imaging in Rheumatoid Arthritis-Animal studies and Clinical Applications
in: Handbook of Biophotonics Vol. 2, 545-60; ed. By J. Popp, V. Tuchin, A. Chiou, and St. Heinemann, WILEY-VCH, Weinheim, Germany (2012).

Gemeinhardt I, Puls D, Gemeinhardt O, Taupitz M, Wagner S, Schnorr B, Licha K, Schirner M, Ebert B, Petzelt D, Macdonald R, Schnorr J,
Near-infrared fluorescence imaging of experimentally collagen-induced arthritis in rats using the nonspecific dye tetrasulfocyanine in comparison with gadolinium-based contrast-enhanced magnetic resonance imaging, histology, and clinical score
J. Biomed. Opt. 17 (10), 106008 (2012) [doi: 10.1117/1.JBO.17.10.106008].

Ebert B, Riefke B, Sukowski U, Licha K,
Cyanine dyes as contrast agents for near-infrared imaging in vivo: acute tolerance, pharmacokinetics, and fluorescence imaging
J. Biomed. Opt. 16(6),066003 (2011) [doi: 10.1117/1.358567]

Licha K, Welker P, Weinhart M, Wegner N, Kern S, Reichert S, Gemeinhardt I, Weissbach C, Ebert B, Haag R, Schirner M (2011),
Fluorescence imaging with multifunctional polyglycerol sulfates: novel polymeric near-IR probes targeting inflammation.
Bioconjug. Chem. 22(12), 2453-2460 (2011) [doi: 10.1021/bc200272].

Dziekan T, Weissbach C, Voigt J, Ebert B, Macdonald R, Bahner ML, Mahler M, Schirner M, Berliner M, Berliner B, Osel J, Osel I,
Detection of rheumatoid arthritis by evaluation of normalized variances of fluorescence time correlation functions
J. Biomed. Opt. 16(7), 076015 (2011) [doi: 10.1117/1.359995].

Fischer T, Ebert B, Voigt J, Macdonald R, Schneider U, Thomas, A, Hamm, B, Hermann, K-G A,
Detection of Rheumatoid Arthritis Using Non-specific Contrast Enhanced Fluorescence Imaging

Acad. Radiol. 17, 375-381 (2010) [doi: 10.1016/j.acra.2009.09.016].

Vollmer S, Vater A, Licha K, Gemeinhardt I, Gemeinhardt O, Voigt J, Ebert B, Schnorr J, Taupitz M, Macdonald R, Schirner M,
ED-B Fibronectin as Target for Near Infrared Fluorescence Imaging of Rheumatoid Arthritis Affected Joints in vivo
Molecular Imaging, 8, 330-340 (2009) [doi: 10.2310/7290.2009.00030].

Fischer T, Gemeinhardt I, Wagner S, v. Stieglitz D, Schnorr J, Hermann K-G A, Ebert B, Petzelt D, Macdonald R, Licha K, Schirner M, Krenn V, Kamradt T, Taupitz M,
Assessment of Unspecific Near-Infrared Dyes in Laser-Induced Fluorescence Imaging of Experimental Arthritis
Acad. Radiol. 13, 4-13 (2006) [doi: 10.1016/j.acra.2005.07.010]

To top