Lukasz A. Paluchowski

Hyperspectral imaging as a diagnostic tool for chronic skin ulcers

The healing process of chronic wounds is complex, and the complete pathogenesis is not known. Diagnosis is currently based on visual inspection, biopsies and collection of samples from the wound surface. This is often time consuming, expensive and to some extent subjective procedures. Hyperspectral imaging has been shown to be a promising modality for optical diagnostics. The main objective of this study was to identify a suitable technique for reproducible classification of hyperspectral data from a wound and the surrounding tissue. Two statistical classification methods have been tested and compared to the performance of a dermatologist. Hyperspectral images (400-1000 nm) were collected from patients with venous leg ulcers using a pushbroom-scanning camera (VNIR 1600, Norsk Elektro Optikk AS).Wounds were examined regularly over 4 - 6 weeks. The patients were evaluated by a dermatologist at every appointment. One patient has been selected for presentation in this paper (female, age 53 years). The oxygen saturation of the wound area was determined by wavelength ratio metrics. Spectral angle mapping (SAM) and k-means clustering were used for classification. Automatic extraction of endmember spectra was employed to minimize human interaction. A comparison of the methods shows that k-means clustering is the most stable method over time, and shows the best overlap with the dermatologist’s assessment of the wound border. The results are assumed to be affected by the data preprocessing and chosen endmember extraction algorithm. Results indicate that it is possible to develop an automated method for reliable classification of wounds based on hyperspectral data.

Can spectral-spatial image segmentation be used to discriminate experimental burn wounds?

Abstract. Hyperspectral imaging (HSI) is a noncontact and noninvasive optical modality emerging the field of medical research. The goal of this study was to determine the ability of HSI and image segmentation to discriminate burn wounds in a preclinical porcine model. A heated brass rod was used to introduce burn wounds of graded severity in a pig model and a sequence of hyperspectral data was recorded up to 8-h postinjury. The hyperspectral images were processed by an unsupervised spectral–spatial segmentation algorithm. Segmentation was validated using results from histology. The proposed algorithm was compared to K-means segmentation and was found superior. The obtained segmentation maps revealed separated zones within the burn sites, indicating a variation in burn severity. The suggested image-processing scheme allowed mapping dynamic changes of spectral properties within the burn wounds over time. The results of this study indicate that unsupervised spectral–spatial segmentation applied on hyperspectral images can discriminate burn injuries of varying severity.

Identification of inflammation sites in arthritic joints using hyperspectral imaging

Inflammatory arthritic diseases have prevalence between 2 and 3% and may lead to joint destruction and deformation resulting in a loss of function. Patient’s quality of life is often severely affected as the disease attacks hands and finger joints. Pathology involved in arthritis includes angiogenesis, hyper-vascularization, hyper-metabolism and relative hypoxia. We have employed hyperspectral imaging to study the hemodynamics of affected- and non-affected joints and tissue. Two hyperspectral, push-broom cameras were used (VNIR-1600, SWIR-320i, Norsk Elektro Optikk AS, Norway). Optical spectra (400nm – 1700nm) of high spectral resolution were collected from 15 patients with visible symptoms of arthritic rheumatic diseases in at least one joint. The control group consisted of 10 healthy individuals. Concentrations of dominant chromophores were calculated based on analytical calculations of light transport in tissue. Image processing was used to analyze hyperspectral data and retrieve information, e.g. blood concentration and tissue oxygenation maps. The obtained results indicate that hyperspectral imaging can be used to quantify changes within affected joints and surrounding tissue. Further improvement of this method will have positive impact on diagnosis of arthritic joints at an early stage. Moreover it will enable development of fast, noninvasive and noncontact diagnostic tool of arthritic joints

A combined 3D and hyperspectral method for surface imaging of wounds

Information about the size and depth of a wound and how it is developing is an important prognostic tool in wound diagnostics. In this study a two-camera vision system has been developed to collect optical properties, shape and volume of chronic skin ulcers as tool for diagnostic assistance. This system combines the functionality of 2D imaging spectroscopy and 3D stereo-photogrammetry. A high resolution hyperspectral camera and a monochromatic video frame camera were mounted on the same scanning system. Stereo images were acquired to obtain information about the wound surface geometry. A Digital Surface Model (DSM) of the wound surface was reconstructed by applying stereophotogrammetric methods. The hyperspectral image was co-registered to the monochromatic frame image and the wound border was extracted by applying spectroscopic analysis (e.g. tissue oxygenation, pigmentation, classification). The resulting DSM of the undamaged surroundings of the wound was used to reconstruct the top surface above the wound and thus the wound volume. The analyses can, if desired, be limited to a certain depth of interest like the wound bed or wound border. Simultaneous analysis of the hyperspectral data and the surface model gives a promising, new, non-invasive tool for characterization of chronic wounds. Future work will concentrate on implementation of real time analysis and improvement of the accuracy of the system.

Vessel contrast enhancement in hyperspectral images

Imaging of vessel structures can be useful for investigation of endothelial function, angiogenesis and hyper-vascularization. This can be challenging for hyperspectral tissue imaging due to photon scattering and absorption in other parts of the tissue. Real-time processing techniques for enhancement of vessel contrast in hyperspectral tissue images were investigated. Wavelet processing and an inverse diffusion model were employed, and compared to band ratio metrics and statistical methods. A multiscale vesselness filter was applied for further enhancement. The results show that vessel structures in hyperspectral images can be enhanced and characterized using a combination of statistical, numerical and more physics informed models.

Hyperspectral imaging for detection of arthritis: feasibility and prospects

Abstract. Rheumatoid arthritis (RA) is a disease that frequently leads to joint destruction. It has a high incidence rate worldwide, and the disease significantly reduces patients’ quality of life. Detecting and treating inflammatory arthritis before structural damage to the joint has occurred is known to be essential for preventing patient disability and pain. Existing diagnostic technologies are expensive, time consuming, and require trained personnel to collect and interpret data. Optical techniques might be a fast, noninvasive alternative. Hyperspectral imaging (HSI) is a noncontact optical technique which provides both spectral and spatial information in one measurement. In this study, the feasibility of HSI in arthritis diagnostics was explored by numerical simulations and optimal imaging parameters were identified. Hyperspectral reflectance and transmission images of RA and normal human joint models were simulated using the Monte Carlo method. The spectral range was 600 to 1100 nm. Characteristic spatial patterns for RA joints and two spectral windows with transmission were identified. The study demonstrated that transmittance images of human joints could be used as one parameter for discrimination between arthritic and unaffected joints. The presented work shows that HSI is a promising imaging modality for the diagnostics and follow-up monitoring of arthritis in small joints.

Simulation of light transport in arthritic- and non-arthritic human fingers

Rheumatoid arthritis is a disease that frequently leads to joint destruction. It has high incidence rates worldwide, and the disease significantly reduces patient’s quality of life due to pain, swelling and stiffness of the affected joints. Early diagnosis is necessary to improve course of the disease, therefore sensitive and accurate diagnostic tools are required. Optical imaging techniques have capability for early diagnosis and monitoring of arthritis. As compared to conventional diagnostic techniques optical technique is a noninvasive, noncontact and fast way of collecting diagnostic information. However, a realistic model of light transport in human joints is needed for understanding and developing of such optical diagnostic tools. The aim of this study is to develop a 3D numerical model of light transport in a human finger. The model will guide development of a hyperspectral imaging (HSI) diagnostic modality for arthritis in human fingers. The implemented human finger geometry is based on anatomical data. Optical data of finger tissues are adjusted to represent either an arthritic or an unaffected finger. The geometry and optical data serve as input into a 3D Monte Carlo method, which calculate diffuse reflectance, transmittance and absorbed energy distributions. The parameters of the model are optimized based on HIS-measurements of human fingers. The presented model serves as an important tool for understanding and development of HSI as an arthritis diagnostic modality. Yet, it can be applied to other optical techniques and finger diseases.

Spectral-spatial classification combined with diffusion theory based inverse modeling of hyperspectral images

Hyperspectral imagery opens a new perspective for biomedical diagnostics and tissue characterization. High spectral resolution can give insight into optical properties of the skin tissue. However, at the same time the amount of collected data represents a challenge when it comes to decomposition into clusters and extraction of useful diagnostic information. In this study spectral-spatial classification and inverse diffusion modeling were employed to hyperspectral images obtained from a porcine burn model using a hyperspectral push-broom camera. The implemented method takes advantage of spatial and spectral information simultaneously, and provides information about the average optical properties within each cluster. The implemented algorithm allows mapping spectral and spatial heterogeneity of the burn injury as well as dynamic changes of spectral properties within the burn area. The combination of statistical and physics informed tools allowed for initial separation of different burn wounds and further detailed characterization of the injuries in short post-injury time.

Hyperspectral characterization of an in vitro wound model

Wound healing is a complex process not fully understood. There is a need of better methods to evaluate the different stages of healing, and optical characterization is a promising tool in this respect. In this study hyperspectral imaging was employed to characterize an in vitro wound model. The wound model was established by first cutting circular patches of human abdominal skin using an 8mm punch biopsy tool, and then creating dermal wounds in the center of the skin patches using a 5mm tool. The wounds were incubated in medium with 10% serum and antibiotics. Hyperspectral images were collected every three days using a push broom hyper spectral camera (Hyspex VNIR1600). The camera had a spectral resolution of 3.7 nm and was fitted with a close up lens giving a FOV of 2.5 cm and a spatial resolution of 29 micrometer. Samples for histology were collected throughout the measurement period, which was 21 days in total. Data were processed in ENVI and Matlab. A successful classification based on hyperspectral imaging of the implemented model is presented. It was not possible to see the healing zone in the in vitro model with the naked eye without dying. The hyperspectral results showed that newly formed epithelium could be imaged without any additional contrast agents or dyes. It was also possible to detect non-viable tissue. In vitro wound models and hyperspectral imaging can thus be employed to gain further insight in the complicated process of healing in different kinds of wounds.

Combined hyperspectral and 3D characterization of non-healing skin ulcers

Current diagnosis of chronic wounds is based on visual inspection, collection of biopsies and analysis of extracted wound fluid. These methods are expensive, invasive, time consuming and often subjective. In this study a hyperspectral imaging system and a stereo vision system were combined to characterize and classify venous ulcers. It was shown that the combination of 2D imaging spectroscopy and 3D stereo-photogrammetry can provide valuable diagnostic information based on the optical properties of the tissue and the surface geometry of the wound.