Search

Acne affects 85% of adolescents at some time during their lives. Dermatologists use manual methods such as direct visual assessment and ordinary flash photography to assess the acne. However, these manual methods are time consuming and may result in intra-observer and inter-observer variations, even by experienced dermatologists. The objective of this research is to develop a computational imaging method for automated acne grading. The first step in the proposed method is pre-processing which involves lighting compensation. The CIE La*b* color space is used to measure any dissimilarity between skin colors. Acne segmentation has been performed using automated modified K-means clustering algorithm and support vector machines (SVM) classifier. Color and diameter are the main features extracted to classify acne blobs into different acne classes; papule, pustule, nodule or cyst. Finally, the severity level is determined such as mild, moderate, severe and very severe.

With the widespread adoption of Cloud Computing, the need for trustworthy service providers becomes more important particularly in medical and health related areas. Service terms as one of trust factors are normally defined in the Service Level Agreement (SLA) binding both providers and customers. This paper presents a framework to perform proactive SLA renegotiation during service runtime that aims to maintain trust by customers on the cloud provider. Requirements and important basis for renegotiation such as detecting SLA violation and assessing certain service level boundaries are discussed. Preliminary experimental work shows that by using historical data, the framework is able to provide suitable recommendation on the SLO values that ensures trust is maintained.

Diabetic Retinopathy (DR) is vision loss impairment due to complication arising from diabetic condition affecting the retina. It is known that the foveal avascular zone (FAZ) enlarges with progression of DR due to the loss of capillaries in perifoveal capillary network. However, normal retinal fundus images suffer from low and varied contrast problems. A non-invasive digital image enhancement technique called RETICA has been developed that overcomes the problem of varied and low contrast in fundus images. RETICA first normalises the varied contrast using a Retinex-based method that separates the illumination from the reflectance part of the image followed by ICA that forms the original retinal pigment makeup namely the macular, haemoglobin and melanin retinal pigment. The technique has been applied on our FINDeRS dataset contained 175 fundus images and another 35 fundus image pairs obtained from an earlier study containing colour fundus images and their corresponding fluorescein fundus angiogram (FFA) images. For the 35 image pairs, RETICA achieved an average contrast improvement factor (CIF) of up to 5.46 compared to the invasive FFA at 5.12. For the FINDeRS images, RETICA achieved an average CIF of 5.63 with denoising. The RETICA image enhancement technique potentially reduces the need for the invasive fluorescein angiogram in DR assessment.

The onset of osteoarthritis (OA), a most common knee joint disease, can be characterized by the degeneration of articular cartilage (AC). Degenerative changes in AC have been assessed by the morphological and physiological measurements using non-invasive modality such as Magnetic Resonance Imaging (MRI) to obtain MRI images of the knee. However, visualization and quantification of AC from MR images is difficult due to the low visibility contrast of AC compared to surrounding tissues, low and varying signal intensities in cartilage region and variable intensities in different slices of single dataset. In this work, we present a method to fuse multinuclear ((23)Na and (1)H) MR images acquired in the same plane without changing the position of the human knee as well as the Radio Frequency (RF) coil. This work is performed towards our hypothesis that fusion of sodium and proton images will provide an enhanced image that can be used for an accurate assessment of cartilage morphology. Our result shows that merging of sodium knee MR image with proton knee MR image resulting in enhanced contrast information in the cartilage region and resolves low visibility and varying intensities issue with 2D/3D proton MR. We conclude that the proposed method can further be utilized for the accurate assessment of cartilage morphology.

Psoriasis is a common skin disorder with a prevalence of 0.6 – 4.8% around the world. The most common is plaques psoriasis and it appears as red scaling plaques. Psoriasis is incurable but treatable in a long term treatment. Although PASI (Psoriasis Area and Severity Index) scoring is recognised as gold standard for psoriasis assessment, this method is still influenced by inter and intra-rater variation. An imaging and analysis system called α-PASI is developed to perform PASI scoring objectively. Percentage of lesion area to the body surface area is one of PASI parameter. In this paper, enhanced imaging methods are developed to improve the determination of body surface area (BSA) and lesion area. BSA determination method has been validated on medical mannequin. BSA accuracies obtained at four body regions are 97.80% (lower limb), 92.41% (trunk), 87.72% (upper limb), and 83.82% (head). By applying fuzzy c-means clustering algorithm, the membership functions of lesions area for PASI area scoring have been determined. Performance of scoring result has been tested with double assessment by α-PASI area algorithm on body region images from 46 patients. Kappa coefficients for α-PASI system are greater than or equal to 0.72 for all body regions (Head - 0.76, Upper limb - 0.81, Trunk - 0.85, Lower limb - 0.72). The overall kappa coefficient for the α-PASI area is 0.80 that can be categorised as substantial agreement. This shows that the α-PASI area system has a high reliability and can be used in psoriasis area assessment.

Diabetic Retinopathy (DR) is a vision loss impairment due to complications arising from the diabetic condition and affects the retina and resulting pathologies can be monitored by analysing the colour fundus image. However, in retinal fundus images, the contrast between the retinal vasculature and the background is very low and varies within the image making visualisation and analysis of small retinal vasculatures difficult. Therefore, enhancement of the fundus image is important to provide the best visualization of the retinal blood vessels. Fluorescein angiogram overcomes this imaging problem but it is invasive and leads to other physiological problems. In this research work, a non-invasive digital image enhancement technique called RETICA has been developed that overcomes the problem of varied and low contrast in fundus images. RETICA first normalises the varied contrast using a Retinex based method that separates the illumination from the reflectance part of the image followed by ICA that forms the original retinal pigment makeup namely the macular, haemoglobin and melanin retinal pigment. The haemoglobin image exhibits the highest contrast for retinal vessels. Results based on a dataset of 13 fundus images show that RETICA successfully normalises the low and varied contrast and enhances the retinal vessels. It achieved a better average contrast improvement factor of up to 5.56 compared to the invasive FFA with 5.34. This improvement in contrast reduces the need for fluorescein angiogram in DR assessment.

Fringe Projection Profilometry (FPP) is the most widely used method for 3D reconstruction. In FPP, the elevation information is deduced from the deformation of the pattern induced by the object. In the shadow areas, no fringe patterns are formed and as a result, the elevation information in these areas cannot be determined resulting in errors in the 3D reconstruction. This paper proposes a new method to detect the occurrences of shadow using the Haar Wavelet for a typical single camera with one projector configuration. The Haar wavelet emphasises the sudden change caused by boundary of the shadow and also shows the non-existence of the pattern in the shadow area. The method detects shadow boundary and differentiates shadow and non-shadow area based on the threshold defined from the statistics of the signal. The method is found more suitable for deformed fringe pattern than Fourier-based shadow detection, as the nature of the signal is non-stationary. The proposed method correctly detects shadow with robustness to the noise power up to 16.9 dB.

Psoriasis is a skin disease that causes the appearance of reddish and scaly skin lesions. Lesion erythema, which refers to the inflammation (colour) of psoriasis lesion, is defined as one of Psoriasis Area and Severity Index (PASI) parameters. However, visual assessment by dermatologists is subjective and results in inter-rater variations. In this paper, an objective PASI erythema-scoring algorithm has been developed. The colour of lesion erythema was found to be dependent on the normal skin tone of the affected person. Normal skin tones are categorised into four groups (dark, brown, light brown and fair skins). A soft clustering is applied to solve the ambiguity problems at cluster boundaries. CIE L*a*b* data of lesions and their surrounding normal skin are used to calculate lesion erythema. The hue difference between lesion and normal skin corresponds to the lesion erythema. Two dedicated fuzzy c-means (FCM) algorithms are applied consecutively to classify normal skin tone and to score PASI erythema. 2,322 normal skin and 1,462 lesions samples from 204 recruited patients at Hospital Kuala Lumpur are used to build skin tone and PASI erythema score classifiers respectively. Agreement values between first and second assessments of 430 lesions for PASI erythema are determined to evaluate scoring performance. Kappa coefficients are found ≥ 0.70 for all skin tones (fair - 0.70, light brown - 0.8, brown - 0.79, and dark skin - 0.90). These agreement results show that the proposed method is reliable and objective, and thus can be used for clinical practices.

In dermatology, skin is assessed for its response to UV light and is categorised under the Fitzpatrick skin type (FST). The FST is commonly applied as a predictor of skin cancer. The FST classification however is subjective. To enable an objective assessment, a pattern model of skin colour taken from normal subjects is required. In this paper, a model for skin phototype (SPT) V is developed and presented. We conducted an observational study involving 41 participants with FST V. The study analysed the reflectance spectra of facultative skin and constitutive skin which was captured using spectrophotometer and multispectral camera. Using piecewise linear regression, we modelled the reflectance spectra over different intervals of wavelength ranging from 430nm to 740nm. It is found that the modelled reflectance achieved a good fit (R-squared >0.995) with the measured reflectance spectra of the 82 skin samples obtained in the study. The proposed method offers an objective determination of skin type SPT V.

Wounds that do not follow a predictable course of healing within a specified period of time develop into ulcers causing severe pain and discomfort to the patients. One of the most prominent changes during wound healing is the colour of the tissues. Describing the tissues in terms of percentages of each tissue colour is an approved clinical method of wound healing assessment. The growth of the red granulation tissue marks the beginning of ulcer healing. Granulation tissue appears red in colour due to haemoglobin content in the blood capillaries. An approach based on utilizing haemoglobin content in chronic ulcers as an image marker to detect the growth of granulation tissue is investigated in this study. Independent Component Analysis is employed to extract grey-level haemoglobin images from RGB colour images of chronic ulcers. Extracted haemoglobin images indicate areas of haemoglobin distribution reflecting detected regions of granulation tissue. Data clustering techniques are implemented to classify and segment detected regions of granulation tissue from the extracted haemoglobin images. Results obtained indicate that the developed algorithm performs fairly well with an average sensitivity of 88.24% and specificity of 98.82% when compared to the dermatologist's assessment. The ultimate aim of this research work is to develop an objective non-invasive wound healing assessment system capable of evaluating the healing status of chronic ulcers in a more precise and reliable way.

A surface roughness algorithm has been developed and validated for determining roughness of psoriasis lesions. The algorithm extracts an estimated waviness surface from 3D rough surface of psoriasis lesion by applying high order polynomial surface fitting. Vertical deviations of the lesion are determined by subtracting its 3D surface from the estimated waviness surface. However, the performance of the algorithm is dependent on the area of skin surface. The objective of this paper is to determine the minimum area for optimal performance of the skin surface roughness algorithm. In the determined sample area, all significant roughness components must be covered for surface roughness determination. To find the minimum size of sampled area, skin surface roughness has been determined at several sampling area variations. Normal skin surfaces are used as input data in this evaluation. By referring to the plot of surface roughness dependency on sampled area variation, it can be shown that the threshold area is found to be 4.9×4.9 mm2 for skin surface roughness stability. Skin surface roughness variation is less for the sample areas larger than this threshold. However, there is a small surface roughness increment after the surface roughness stability. It is caused by fitting error at border regions of very large sample size.

Various methods have been proposed to analyse fringe patterns in Fringe Projection Profilometry (FPP) used for 3D Acquisition systems. FPP is a full frame and high-speed acquisition system that enables fast 3D acquisition of large areas with high accuracy. Most of fringe analysis methods assume that the captured pattern is perfect sinusoidal. However, due to projector response and cross talk between colours, optical aberration and the presence of noise, it is difficult to obtain a perfect sinusoidal signal. In this paper, Spline Inverse Function Analysis (SIFA) is proposed to analyse both sinusoidal and non-sinusoidal patterns in FPP. This method is based on Inverse Function Analysis (IFA) method, but instead using polynomial fitting, this method uses spline to avoid Runge phenomenon and heavy computation on determining the polynomial degree and minimize the error. The proposed method is compared with Fourier Transform Profilometry (FTP) method and results show that SIFA gives 50 times better result than FTP.

Abnormality of melanin production causes skin pigmentation disorders. Many simulation models of light interaction with human skin have been reported. The most accurate skin model is based on Monte Carlo approach. An inverse model is used to derive optical properties of skin by fitting the spectral reflectance obtained from the model of light interaction to the actual spectral reflectance obtained from empirical data. Currently there is no an inverse Monte Carlo model which is designed to analyse skin chromophores especially melanin type and its concentration. In this paper, a development of an inverse Monte Carlo model to analyse skin pigmentation model is presented. The inverse model is first validated with 37 spectral reflectance data obtained from normal participants. An observational clinical study involving 11 melasma patients is then conducted. As expected, result shows that the concentrations of eumelanin and pheomelanin of all melasma lesions are higher than concentration of eumelanin and pheomelanin of normal skin, resulting in higher light photon absorption. It can be inferred that the developed inverse model is able to determine relative concentrations of eumelanin and pheomelanin.

Acne is chronic disorder of the pilosebaceous units with excess sebum production, follicular epidermal hyper proliferation, inflammation and P acnes activity. It affects 85% of adolescents at some time during their lives. Dermatologists use manual methods such as direct visual assessment and ordinary flash photography to assess the acne. These methods are very time consuming and tedious. To address these issues, researchers in recent years have proposed computational imaging methods for aiding in the acne diagnosis. To develop algorithm with an automated acne grading method is the objective of this proposed method. This work presents an image segmentation method for acne lesions based on color features with K-means clustering. The segmentation results from randomly selected images show the sensitivity, specificity, positive predictive value and negative predictive value greater than 81%.

Ulcers are chronic wounds that cause severe pain and discomfort to the patients. Describing the ulcer tissues in terms of percentages of each tissue colour is an approved method of wound healing assessment. The first indication of ulcer healing is the growth of the red granulation tissue from the base of ulcers. Granulation tissue appears red in colour due the hemoglobin content in the blood capillaries. The main objective of this research work is to investigate the optical characteristics of hemoglobin content in ulcers as a possible image marker in detecting granulation tissue from RGB colour images of chronic ulcers. Independent Component Analysis is implemented to extract grey-level hemoglobin images from RGB colour images of chronic ulcers. Extracted hemoglobin images reflect areas of hemoglobin distribution representing detected regions of granulation tissue. K-means clustering is implemented to classify and segment detected regions of granulation tissue. Preliminary analysis of the results indicates an overall accuracy of 97.31% of the algorithm performance when compared to manual segmentation. The ultimate aim of this research work is to produce an objective non-invasive approach that aid physicians to assess the healing status of chronic ulcers in a more precise and reliable way.

Abnormality of melanin production causes skin pigmentation disorders. Currently, assessment of treatment efficacy (under Physician's Global Assessment framework) only refers to visual conditions of skin surface and not the condition of the underlying skin layers and pigments. Albeit researches on models and simulations of light interaction with human skin have been reported, none has been specifically developed for pigmentation analysis of melanin types - eumelanin and pheomelanin. Therefore, our research objectives are to develop image analysis of skin pigmentation for classification and quantification of eumelanin and pheomelanin pigment types in human skin. In this research, the model is developed using data collected from clinical study. It is hypothesised that the multispectral approach will provide an accurate characterisation of skin layers to determine melanin types namely eumelanin and pheomelanin. Monte Carlo method is then used to determine skin model parameters such as concentration of eumelanin and pheomelanin.

Acne affects 85% of adolescents at some time during their lives. There are various causes for acne including genetic, hormonal, sebaceous activity, bacteria, climate, chemical and psychological. Till now, dermatologists use manual methods such as direct visual assessment and ordinary flash photography to assess the acne. These methods are very time consuming and tedious. To address these issues, researchers in recent years have proposed computational imaging methods for aiding in the acne diagnosis. This paper proposes an algorithm to identify acne lesions, scars and normal skin features from photographs taken by Digital Single-Lens Reflex (DSLR) cameras. The images are converted from RGB to CIELAB color space, thresholded to three clusters and segmented using minimum Euclidean distance. The segmentation results from randomly selected images show sensitivity and specificity of greater than 80%.

Diabetic retinopathy (DR) is a sight threatening complication due to diabetes mellitus that affects the retina. Research shows that the foveal avascular zone (FAZ) could be used to diagnose DR as it enlarges in DR cases due to the loss of capillaries in the perifoveal capillary network. A computerised DR monitoring and grading system based on analysis of FAZ enlargement in colour fundus image has been developed and has high values of sensitivity, specificity and accuracy. However the system is still semi automated. A user interruption is needed in the selection of retinal vessel endpoints to determine the FAZ. Therefore, the objective of this paper is to improve the FAZ determination so that the computerised DR grading system is fully automated. Results show the fully automated system consistently maintains high sensitivity (>73%), specificity (>77%) and accuracy (>77%) for all DR stages. This indicates that the fully automated DR monitoring and grading system has a potential to be used for early detection of DR and for effective treatment of severe cases.

In skin care industry and dermatology, analysis of human skin tone is an important parameter for evaluating the current condition of the skin. Research has shown that the human skin tones are due to the combination of skin chromophores (pigments) such as melanin, haemoglobin, bilirubin and beta-carotene. Several works on skin modelling have been reported but none has been specifically developed to classify and measure types of melanin. In this research, the spectral responses of different human skin phototypes are investigated for melanin pigment (pheomelanin and eumelanin) analysis. We propose skin pigmentation model based on a modified Beer-Lambert model of skin diffuse reflectance to measure types of melanin. A clinical study involving 118 participants with different skin phototypes (SPTs) is conducted where the skin reflectance data of participants are measured using Spectrophotometer Konica Minolta 2600c. Applying the data to the proposed skin model, it was found the pheomelanin concentration is -4.6E5± 5.4E-6 moles/l for SPT III, -5.9E-5±6.4E- 6 moles/l for SPT IV, and -8.2E-5±9.8E-6moles/l for SPT V and the eumelanin concentration is 9.7E-5±7.3E-6 moles/l for SPT III, 1.2E-4±1.03E-5 moles/l for SPT IV, and 1.6E-4±1.7E-5 moles/l for SPT V. Results show that proposed model can be used for pigmentation analysis to measure melanin pigments types in skin.

The ability to measure objectively wound healing is important for an effective wound management. Describing wound tissues in terms of percentages of each tissue colour is an approved clinical method of wound assessment. Wound healing is indicated by the growth of the red granulation tissue, which is rich in small blood capillaries that contain haemoglobin pigment reflecting the red colour of the tissue. A novel approach based on utilizing haemoglobin pigment content in chronic ulcers as an image marker to detect the growth of granulation tissue is investigated in this study. Independent Component Analysis is employed to convert colour images of chronic ulcers into images due to haemoglobin pigment only. K-means clustering is implemented to classify and segment regions of granulation tissue from the extracted haemoglobin images. Results obtained indicate an overall accuracy of 96.88% of the algorithm performance when compared to the manual segmentation.

Scaliness of psoriasis lesions is one of the parameters to be determined during Psoriasis Area and Severity Index (PASI) scoring. Dermatologists typically use their visual and tactile senses to assess PASI scaliness. However, it is known that the scores are subjective resulting in inter- and intra-rater variability. In this paper, an objective 3D imaging method is proposed to assess PASI scaliness parameter of psoriasis lesions. As scales on the lesion invariably causes roughness, a surface-roughness measurement method is proposed for 3D curved surfaces. The method applies a polynomial surface fitting to the lesion surface to extract the estimated waviness from the actual lesion surface. Surface roughness is measured from the vertical deviations of the lesion surface from the estimated waviness surface. The surface roughness algorithm has been validated against 328 lesion models of known roughness on a medical mannequin. The proposed algorithm is found to have an error 0.0013 ± 0.0022 mm giving an accuracy of 89.30%. The algorithm is invariant to rotation of the measured surface. Accuracy of the rotated lesion models is found to be greater than 95%. System repeatability has been evaluated to successive measurements of 456 psoriasis lesions. The system repeatability can be accepted since 95.27% of the measurement differences are less than two standard deviation of measurement difference.

A signal subspace approach for extracting visual evoked potentials (VEPs) from the background electroencephalogram (EEG) colored noise without the need for a prewhitening stage is proposed. Linear estimation of the clean signal is performed by minimizing signal distortion while maintaining the residual noise energy below some given threshold. The generalized eigendecomposition of the covariance matrices of a VEP signal and brain background EEG noise is used to transform them jointly to diagonal matrices. The generalized subspace is then decomposed into signal subspace and noise subspace. Enhancement is performed by nulling the components in the noise subspace and retaining the components in the signal subspace. The performance of the proposed algorithm is tested with simulated and real data, and compared with the recently proposed signal subspace techniques. With the simulated data, the algorithms are used to estimate the latencies of P(100), P(200), and P(300) of VEP signals corrupted by additive colored noise at different values of SNR. With the real data, the VEP signals are collected at Selayang Hospital, Kuala Lumpur, Malaysia, and the capability of the proposed algorithm in detecting the latency of P(100) is obtained and compared with other subspace techniques. The ensemble averaging technique is used as a baseline for this comparison. The results indicated significant improvement by the proposed technique in terms of better accuracy and less failure rate.

Data from medical imaging system need to be analysed for diagnostics and clinical purposes. In a computerized system, the analysis normally involves classification process to determine disease and its condition. In an earlier work based on a database of 315 fundus images (FINDeRS), it is found that the foveal avascular zone (FAZ) enlargement strongly correlates with diabetic retinopathy (DR) progression having a correlation factor up to 0.883 at significant levels better than 0.01. However, it is also found that the FAZ areas can belong to different DR severity but with different levels of certainty having a Gaussian distribution. In this research work, the suitability of the Gaussian Bayes classifier in determining DR severity level is investigated. A v-fold cross-validation (VFCF) process is applied to the FINDeRS database to evaluate the performance of the classifier. It is shown that the classifier achieved sensitivity of >84%, specificity of >97% and accuracy of >95% for all DR stages. At high values of sensitivity (>95%), specificity (>97%) and accuracy (>98%) obtained for No DR and Severe NPDR/PDR stages, the Gaussian Bayes classifier is suitable as part of a computerised DR grading and monitoring system for early detection of DR and for effective treatment of severe cases.

Thickness is one of the morphological characteristic of skin lesion that represents severity condition. Dermatologists use tactile inspection to subjectively assess the thickness by feeling the alteration of the lesion from its surrounding normal skin. In this paper, a method to objectively measure the abnormal elevation occurs in skin lesions is presented. A 3D fringe projection scanner is used to obtain 3D surface profile of the lesion. Thickness of a lesion is defined as the elevations of lesion surface from its lesion base. The lesion base is determined from the neighboring normal skin using a 3D surface interpolation technique. The lesion elevations are determined in a 3D space grid by subtracting the elevation of the lesion surface profile from the interpolated lesion base profile at all corresponding locations thus giving lesion thickness as the average value of the elevations. The algorithm has been validated using 3D surface samples with an error of 0.031 mm ± SD 0.014 mm (95% Confidence Interval: ±0.0011 mm). The validated algorithm has been successfully applied to measure thicknesses of 450 psoriasis plaque lesions with severity level ranging from mild to severe and thickness ranging from 0.021 mm to 0.883 mm. From the measured thicknesses, Psoriasis Area and Severity Index (PASI) thickness scores 0 to 4 are then determined using unsupervised K-means Clustering.

Measuring changes in ulcer during treatment can indicate the effectiveness of a treatment regime. Identifying appropriate treatment regime will reduce healing time. Current methods for measuring wound size are subjective and require manual contact with the wound. With the availability of techniques which can provide accurate depth measurement and surface profile, computer programs that can construct solids out of the wounds and quantitatively compute volume are indispensable. The development of volume estimation algorithms from 3D skin surface images to monitor wounds progress throughout treatment is discussed. The performance of midpoint projection and convex hull approximation (Delaunay tetrahedralization) methods, for solid reconstruction and volume computation, is detailed. Results of calculating wound models volume indicate that convex hull reconstruction preceded by surface division outperforms midpoint projection in case of regular boundary models. The error ranged from (0–2.8%) for convex hull reconstruction, while midpoint projection error range was from (0–6.5%).

Skin surface roughness can be changed due to skin lesion progression. The evaluation of the damage of the skin is assessed on the base of clinical experience of dermatologists. Currently, the assessment of skin surface is made by running a finger across the skin. It is found that this tactile evaluation may result in inter and intra variations. The objective of our research is to develop an objective assessment tool in evaluation skin roughness. Psoriasis is used as a sample of skin lesion that can change skin surface roughness as its extents develop. 3D surface images of 120 psoriasis lesions scanned from 29 patients are analysed in the study. The lesions are previously grouped into four categorises based on its tactile characteristics by dermatologists. The proposed method converts the 3D images into 2D height maps. The 2D height maps are then fitted and filtered using averaging filter. Finally, Mean Square Error (MSE) between the fitted and filtered 2D height maps are computed and utilized as skin roughness descriptors. Results show that the proposed method has been able to differentiate each lesion according to its group perfectly.

Early detection of several diseases related to the retina can be analyzed from fundus images. However, in fundus images the contrast between retinal vasculature and the background is very low. Therefore, analyzing these tiny retinal blood vessels is difficult. Fluorescein angiogram overcomes this imaging problem; however, it is an invasive procedure that leads to other physiological problems. In this work, we develop a fundus image model based on probability distribution function of melanin, haemoglobin and macular pigment to represent melanin, retinal vasculature and macular region, respectively. Enhancement of the low contrast of retinal vasculature in the retinal fundus image is performed by separating the retinal pigments makeup, namely macular pigment, haemoglobin and melanin, using independent component analysis. Independent component image due to haemoglobin obtained exhibits higher contrast retinal blood vessels. Results show that this approach outperforms other non-invasive enhancement methods and can be beneficial for retinal vasculature segmentation. Contrast enhancement factor of 2.62 for a digital retinal fundus image model is achieved. This improvement in contrast reduces the need of applying contrasting agent on patients.

In this work 3D surface scans of wounds are used to obtain several measurement including wound top area, true surface area (rue area), depth, and volume for the purpose of assessing the progress of ulcer wounds throughout treatment. KONICA MINOLTA 910 laser scanner is used to obtain the surface scans. The algorithm for estimating top area and true surface area from surface scan can reduce the inaccuracy that might result when using manual method. Two methods for solid construction and volume computation were considered; namely mid-point projection and convex hull approximation (Delaunay tetrahedralization). The performance of convex hull approximation method for volume estimation is improved by performing surface subdivision prior to the approximation. The performance of these algorithms on different patterns of simulated wound models is presented. Furthermore the algorithms are tested in two molded wounds printed using rapid prototyping (RP) technique.

The quantitative computation of surface roughness has received increasing attention due to its importance in numerical surface study. A number of algorithms have been employed to compute surface roughness. These algorithms, which operate at singular scales of measurement, provide scale-dependant roughness parameters. In this paper, the computation of a scale-independent roughness parameter is performed via the generation of multiscale digital elevation models (DEMs).The proposed procedure employs the lifting scheme to generate multiscale DEMs. Using the mask of pixels modified at each scale, granulometric analysis is employed to compute the average size of convex and concave regions in the DEM, and the scale-independent average roughness of the terrain of the DEM due the distribution of convex and concave regions in the terrain. The proposed procedure provides a surface roughness parameter that is realistic with respect to the amplitudes and frequencies of the terrain, invariant with respect to rotation and translation, and has intuitive meaning.

The 3D surface textures of skin are important for both visual and tactile inspection. The information of surface textures can aid the inclusion of haptic technology in tele-dermatology as well as enhance the effectiveness of current 2D based applications. This work aims to analyse and model surface textures of skin including diseased skin. For this purpose the multiple pairwise pixel interaction model of Markov-Gibbs Random Field (MGRF) has been used. Surface textures are in the form of 3D mesh which are converted to 2D height maps and modeled as grayscale textures. The results demonstrate that homogenous stochastic textures of skin can be modeled successfully. The successfully modeled textures represent the surface irregularities on scale of 0.1 mm which can create the tactile perception of roughness for skin in haptic applications.