This is the second part of a two-part paper on the application of computer-aided diagnosis to diffuse optical tomography (DOT) for diagnosing rheumatoid arthritis (RA). A comprehensive analysis of techniques for the classification of DOT images of proximal interphalangeal joints of subjects with and without RA is presented. A method for extracting heuristic features from DOT images was presented in Part 1. The ability of five classification algorithms to accurately label each DOT image as belonging to a subject with or without RA is analyzed here. The algorithms of interest are the k -nearest-neighbors, linear and quadratic discriminant analysis, self-organizing maps, and support vector machines (SVM). With a polynomial SVM classifier, we achieve 100.0% sensitivity and 97.8% specificity. Lower bounds for these results (at 95.0% confidence level) are 96.4% and 93.8%, respectively. Image features most predictive of RA are from the spatial variation of optical properties and the absolute range in feature values. The optimal classifiers are low-dimensional combinations (<7 features). These results underscore the high potential for DOT to become a clinically useful diagnostic tool and warrant larger prospective clinical trials to conclusively demonstrate the ultimate clinical utility of this approach.
This is the first part of a two-part paper on the application of computer-aided diagnosis to diffuse optical tomography (DOT). An approach for extracting heuristic features from DOT images and a method for using these features to diagnose rheumatoid arthritis (RA) are presented. Feature extraction is the focus of Part 1, while the utility of five classification algorithms is evaluated in Part 2. The framework is validated on a set of 219 DOT images of proximal interphalangeal (PIP) joints. Overall, 594 features are extracted from the absorption and scattering images of each joint. Three major findings are deduced. First, DOT images of subjects with RA are statistically different (p<0.05 ) from images of subjects without RA for over 90% of the features investigated. Second, DOT images of subjects with RA that do not have detectable effusion, erosion, or synovitis (as determined by MRI and ultrasound) are statistically indistinguishable from DOT images of subjects with RA that do exhibit effusion, erosion, or synovitis. Thus, this subset of subjects may be diagnosed with RA from DOT images while they would go undetected by reviews of MRI or ultrasound images. Third, scattering coefficient images yield better one-dimensional classifiers. A total of three features yield a Youden index greater than 0.8. These findings suggest that DOT may be capable of distinguishing between PIP joints that are healthy and those affected by RA with or without effusion, erosion, or synovitis.
In this paper we present two-dimensional phantom measurements of fluorescence light distribution in the frequency
domain and reconstruction of three-dimensional fluorophore distribution. An experimental set-up was built up with two
dimensional laser scanning, intensity modulation with frequencies up to 1 GHz, and two-dimensional imaging of
modulated fluorescence light. Stable phantoms were developed simulating mammary tissue to perform measurements in
a backscattering geometry for a variety of cylindrical fluorescence sources with different diameters, fluorophore
concentrations, and surface distances at different modulation frequencies. At first calculated fluorescence light
distributions from Monte-Carlo simulations was compared to measured data. In a second step from tomographic data sets
of calculated fluorescent light, three-dimensional tomographic reconstructions of fluorophore distribution were
performed. Finally three-dimensional tomographic reconstructions of fluorophore distribution were performed from
tomographic fluorescence measurements. We found good concurrence between measured and calculated fluorescence
distribution. Synthetic and real tomographic reconstruction showed good localization but underestimated the depth of
fluorophore distribution.
We present a study on the effectiveness of computer-aided diagnosis (CAD) of rheumatoid arthritis (RA) from frequency-domain diffuse optical tomographic (FDOT) images. FDOT is used to obtain the distribution of tissue optical properties. Subsequently, the non-parametric Kruskal-Wallis ANOVA test is employed to verify statistically significant differences between the optical parameters of patients affected by RA and healthy volunteers. Furthermore, quadratic discriminate analysis (QDA) of the absorption (μa) and scattering (μa or μ's) distributions is used to classify subjects as affected or not affected by RA.
We evaluate the classification efficiency by determining the sensitivity (Se), specificity (Sp), and the Youden index (Y). We find that combining features extracted from μa and μa or μ's images allows for more accurate classification than when μa or μa or μ's features are considered individually on their own. Combining μa and μa or μ's features yields values of up to Y = 0.75 (Se = 0.84 and Sp = 0.91). The best results when μa or μ's features are considered individually are Y = 0.65 (Se = 0.85 and Sp = 0.80) and Y = 0.70 (Se = 0.80 and Sp = 0.90), respectively.
A computer-aided interpretation approach is proposed to detect rheumatic arthritis (RA) in human finger joints using optical tomographic images. The image interpretation method employs a classification algorithm that makes use of a so-called self-organizing mapping scheme to classify fingers as either affected or unaffected by RA. Unlike in previous studies, this allows for combining multiple image features, such as minimum and maximum values of the absorption coefficient for identifying affected and not affected joints. Classification performances obtained by the proposed method were evaluated in terms of sensitivity, specificity, Youden index, and mutual information. Different methods (i.e., clinical diagnostics, ultrasound imaging, magnet resonance imaging, and inspection of optical tomographic images), were used to produce ground truth benchmarks to determine the performance of image interpretations. Using data from 100 finger joints, findings suggest that some parameter combinations lead to higher sensitivities, while others to higher specificities when compared to single parameter classifications employed in previous studies. Maximum performances are reached when combining the minimum/maximum ratio of the absorption coefficient and image variance. In this case, sensitivities and specificities over 0.9 can be achieved. These values are much higher than values obtained when only single parameter classifications were used, where sensitivities and specificities remained well below 0.8.
The intrinsic optical parameters absorption coefficient µa, scattering coefficient µs, anisotropy factor g, and effective scattering coefficient µ were determined for human red blood cell (RBC) suspensions of hematocrit 33.2% dependent on the oxygen saturation (SAT O2) in the wavelength range 250 to 2000 nm, including the range above 1100 nm, about which there are no data available in the literature. Integrating sphere measurements of light transmittance and reflectance in combination with inverse Monte Carlo simulation were carried out for SAT O2 levels of 100 and 0%. In the wavelength range up to 1200 nm, the absorption behavior is determined by the hemoglobin absorption. The spectral range above the cells' absorption shows no dependence on SAT O2 and approximates the absorption of water with values 20 to 30% below the respective values for water. Parameters µs and g are significantly influenced by the SAT O2-induced absorption changes. Above 600 nm, µs decreases continuously from values of 85 mm−1 to values of 30 mm−1 at 2000 nm. The anisotropy factor shows a slight decrease with wavelengths above 600 nm. In the spectral regions of 1450 and 1900 nm where water has local absorption maxima, g shows a significant decrease down to 0.85, whereas µ increases.
The frequency-domain experimental data is typically corrupted by noise and the measurement accuracy is compromised.
Assuming the widely used shot-noise model, it is well-known that the signal-to-noise ratio (SNR) of the amplitude signal
decreases with increasing frequency, whereas the SNR of phase measurement reaches a peak value in the range between
400 MHz and 800 MHz in tissue volumes typical for small animal imaging studies. As a consequence, it can be assumed
that there exists an optimal frequency for which the reconstruction accuracy would be best. To determine optimal
frequencies for FDOT, we investigate here the frequency dependence of optical tomographic reconstruction results using
the frequency-domain equation of radiative transfer. We present numerical and experimental studies with a focus on
small tissue geometries as encountered in small animal imaging and imaging of human finger joints affected by arthritis.
Best results were achieved in the 400-800 MHz frequency range, depending on the particular optical properties.
Novel methods that can help in the diagnosis and monitoring of joint disease are essential for
efficient use of novel arthritis therapies that are currently emerging. Building on previous
studies that involved continuous wave imaging systems we present here first clinical data obtained
with a new frequency-domain imaging system. Three-dimensional tomographic data sets of absorption and
scattering coefficients were generated for 107 fingers. The data were analyzed using ANOVA, MANOVA,
Discriminant Analysis DA, and a machine-learning algorithm that is based on self-organizing mapping
(SOM) for clustering data in 2-dimensional parameter spaces. Overall we found that the SOM algorithm
outperforms the more traditional analysis methods in terms of correctly classifying finger
joints. Using SOM, healthy and affected joints can now be separated with a sensitivity of 0.97 and
specificity of 0.91. Furthermore, preliminary results suggest that if a combination of multiple
image properties is used, statistical significant differences can be found between RA-affected finger joints that show different clinical features (e.g. effusion, synovitis or erosion).
A recent research study has shown that combining multiple parameters, drawn from optical tomographic images,
leads to better classification results to identifying human finger joints that are affected or not affected
by rheumatic arthritis RA. Building up on the research findings of the previous study, this article presents an
advanced computer-aided classification approach for interpreting optical image data to detect RA in finger joints.
Additional data are used including, for example, maximum and minimum values of the absorption coefficient
as well as their ratios and image variances. Classification performances obtained by the proposed method were
evaluated in terms of sensitivity, specificity, Youden index and area under the curve AUC. Results were compared
to different benchmarks ("gold standard"): magnet resonance, ultrasound and clinical evaluation. Maximum accuracies
(AUC=0.88) were reached when combining minimum/maximum-ratios and image variances and using
ultrasound as gold standard.
This research study explores the combined use of more than one parameter derived from optical tomographic images to increase diagnostic accuracy which is measured in terms of sensitivity and specificity. Parameters considered include, for example, smallest or largest absorption or scattering coefficients or the ratios thereof in an image region of interest. These parameters have been used individually in a previous study to determine if a finger joint is affected or not affected by rheumatoid arthritis. To combine these parameters in the analysis we employ here a vector quantization based classification method called Self-Organizing Mapping (SOM). This method allows producing multivariate ROC-curves from which sensitivity and specificities can be determined. We found that some parameter combinations can lead to higher sensitivities whereas others to higher specificities when compared to singleparameter classifications employed in previous studies. The best diagnostic accuracy, in terms of highest Youden index, was achieved by combining three absorption parameters [maximum(µa), minimum(µa), and the ratio of minimum(µa) and maximum(µa)], which result in a sensitivity of 0.78, a specificity of 0.76, a Youden index of 0.54, and an area under the curve (AUC) of 0.72. These values are higher than for previously reported single parameter classifications with a best sensitivity and specificity of 0.71, a Youden index of 0.41, and an AUC of 0.66.
We found that using more than one parameter derived from optical tomographic images can lead to better image classification results compared to cases when only one parameter is used.. In particular we present a multi-parameter classification approach, called self-organizing mapping (SOM), for detecting synovitis in arthritic finger joints based on sagittal laser optical tomography (SLOT). This imaging modality can be used to determine various physical parameters such as minimal absorption and scattering coefficients in an image of the proximal interphalengeal joint. Results were compared to different gold standards: magnet resonance imaging, ultra-sonography and clinical evaluation. When compared to classifications based on single-parameters, e.g., absorption minimum only, the study reveals that multi-parameter classifications lead to higher classification sensitivities and specificities and statistical significances with p-values <5 per cent. Finally, the data suggest that image analyses are more reliable and avoid ambiguous interpretations when using more than one parameter.
Inflammatory processes as they occur during rheumatoid arthritis (RA) lead to changes in the optical properties of joint tissues and fluids. These changes occur early on in the disease process and can potentially be used as diagnostic parameter. In this work we report on in vivo studies involving 12 human subjects, which show the potential of diffuse optical tomographic techniques for the diagnosis of inflammatory processes in proximal interphalangeal (PIP) joints.
Diffuse optical tomography is emerging as a viable new biomedical imaging modality. Using near-infrared light this technique probes absorption as well as scattering properties of biological tissues. First commercial instruments are now available that combined with appropriate image reconstruction scheme allow to obtain cross sectional views of various body parts. The main applications are currently brain, breast, limb and joint imaging. While the spatial resolution is limited compared to other imaging modalities such as MRI or X-ray tomography, diffuse optical tomography provides for a fast, inexpensive, acquisition of a variety of physiological parameters that are otherwise not accessible. We present here a brief overview over the current state-of-the-art technology and some of its main applications.
KEYWORDS: Scattering, Tissue optics, Sensors, Ultrasonography, Signal detection, Modulation transfer functions, Light scattering, Tissues, Modulation, Signal to noise ratio
Acousto-photonic imaging (API) is a new approach in biomedical imaging that combines diffuse imaging by photon density waves (PDW) and light "tagging" inside the tissue by focussed ultrasound. This light "tagging" enables 3D optical imaging with mm resolution in tissue limited only by the geometrical extent of the ultrasound focus and the signal to noise ratio.
We discuss some possible mechanisms of light "tagging" and its dependance of different parameters. We present several phantom measurements which investigate advantages and disadvantages of API against PDW. The main advantage of API is the possibility of real 3D imaging while its biggest disadvantage is the poor light intensity from deeper regions.
The system theory was developed from the combination of the operational calculus of N. Wiener and the transfer theory of K. Kuepfmueller. The system theory is the basic for the optical transfer function. With the introduction of a so- called transfer (or system) function, the question arose how to apply this theory to problems in optical tissue diagnostics.
The absorption coefficient (mu) a, the scattering coefficient (mu) s, and the scattering anisotropy factor g of porcine liver were studied in vitro using the integrating sphere technique and inverse Monte Carlo simulation in the wavelength range 450 to 700 nm. A reference preparation technique was developed using a dermatome providing specimens of 200 to 800 micrometers thickness without pre-freezing the tissue. The optical parameters as measured applying the reference preparation were compared to those measured after cryo-homogenization. We found significant deviations of the scattering coefficient and the anisotropy factor which were compensated when the reduced scattering coefficient (mu) s' was calculated. We also compared the effects of freezing reference specimens at - 20 degree(s)C and at 77 K without homogenization. For both freezing protocols noticeable deviations were found in all three optical parameters as well as in (mu) s'. The impact of tissue storage at 4 degree(s)C was measured in the range 4 to 48 hours post mortem and showed a clear reduction of (mu) a and a significant increase of (mu) s even after 24 hours of storage. Short time storage of the specimens in saline solution reduced all three optical parameters significantly. In conclusion, the tissue preparation must be controlled in order to provide in vitro optical parameters that sufficiently mimic the in vivo situation.
An experimental setup will be presented which provides the detection of photon density waves (PDW) in transmission through tissue samples of up to 4 cm thickness. A laser diode (825 nm) which was modulated at frequencies between 40 - 1000 MHz served as light source. Amplitude and phase of the PDW were measured at various positions on tissue phantoms as well as during interstitial laser coagulation of tissue samples (porcine liver) by scanning source and detector. A Nd:YAG laser was used in combination with a scattering applicator to produce thermal lesions. Due to a change of the optical properties the actual size of the coagulated volume could be monitored during the therapy. Several phantom and in vitro experiments have been performed to show that this monitoring technique is capable of visualizing the coagulated region with a clinically relevant precision.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.