Multimodal imaging device to comprehensively assess infection, oxygenation, and wound analytics-A pilot study.

Wound analytics, infection detection, and oxygenation measurement are the three critical prerequisites for appropriate wound care. Although devices that rapidly detect the above-mentioned parameters independently exist, there is no single point-of-care device that is enabled with all the three functionalities. Through this study, we are introducing and evaluating the performance of Illuminate Pro Max-a novel, rapid, hand-held non-contact, point-of-care multimodal imaging device that is equipped to measure the three wound assessment parameters. Here, a total of 60 diabetic foot ulcer patients were imaged using Illuminate Pro Max to detect bioburden and measure StO2 levels and wound dimensions (size and depth). The results were further evaluated against the current gold standard technique for each parameter, that is, culture test to detect bioburden, a transcutaneous oxygen pressure (TcPO2) measuring device-Perimed Periflux 5000 to measure oxygenation, and paper ruler to measure wound size. Culture tests reported 42 samples as infection-positive and 18 samples as infection-negative. On comparing with the culture report, the device showed 88% sensitivity and 86% PPV in detecting the bioburden. Wound dimensions (length and width) were comparable with the paper scale measurements. Wound depth was also reported by the device. The StO2 map generated by the device depicted the tissue oxygenation levels in various regions of the wound. In conclusion, this novel, comprehensive point-of-care multispectral imaging device can be an effective tool for rapid wound assessment which can help in prompt treatment.

Multispectral-derived genotypic similarities from budget cameras allow grain yield prediction and genomic selection augmentation in single and multi-environment scenarios in spring wheat.

With abundant available genomic data, genomic selection has become routine in many plant breeding programs. Multispectral data captured by UAVs showed potential for grain yield (GY) prediction in many plant species using machine learning; however, the possibilities of utilizing this data to augment genomic prediction models still need to be explored. We collected high-throughput phenotyping (HTP) multispectral data in a genotyped multi-environment large-scale field trial using two cost-effective cameras to fill this gap. We tested back to back the prediction ability of GY prediction models, including genomic (G matrix), multispectral-derived (M matrix), and environmental (E matrix) relationships using best linear unbiased predictor (BLUP) methodology in single and multi-environment scenarios. We discovered that M allows for GY prediction comparable to the G matrix and that models using both G and M matrices show superior accuracies and errors compared with G or M alone, both in single and multi-environment scenarios. We showed that the M matrix is not entirely environment-specific, and the genotypic relationships become more robust with more data capture sessions over the season. We discovered that the optimal time for data capture occurs during grain filling and that camera bands with the highest heritability are important for GY prediction using the M matrix. We showcased that GY prediction can be performed using only an RGB camera, and even a single data capture session can yield valuable data for GY prediction. This study contributes to a better understanding of multispectral data and its relationships. It provides a flexible framework for improving GS protocols without significant investments or software customization. Supplementary Information: The online version contains supplementary material available at 10.1007/s11032-024-01449-w.

Novel multispectral imaging to predict disease progression in pediatric morphea.

BACKGROUND: Morphea, or localized scleroderma, is an inflammatory, fibrosing skin disorder that can be progressive and debilitating. Infrared thermography frequently has false positive results. The aim of this study was to assess the ability of multispectral imaging to predict disease progression in children with morphea. METHODS: Children with morphea were recruited between 2016 and 2022. Multispectral images of affected and matched contralateral unaffected sites were obtained using the Antera™ 3D camera. Clinical assessment was performed using the Localized Scleroderma Assessment Tool (LoSCAT). Children were followed up every 3 months for imaging and clinical review. The main outcome measurement was correlation of hemoglobin gradient between affected and matched contralateral unaffected tissue and progression. RESULTS: Of 17 children, the average age was 12 years (range 6-18 years); most were female (76.5%) and white (94.1%). Nearly two-thirds (64.7%) had linear morphea, 35.2% had plaque morphea; 58.8% had been treated with systemic agents. The average LoSCAT score was 20.6 (range 5-73). The average hemoglobin gradient between affected and matched contralateral unaffected skin was four times higher in those who had progression (average differential 0.3, range 0.1-0.4) compared to those who did not (average differential 0.08, range 0.02-0.15). Using a cut off of a 0.18 hemoglobin gradient between affected and unaffected skin, the sensitivity of multispectral imaging for detecting progression in pediatric morphea is 90% with specificity of 100%. CONCLUSIONS: Multispectral imaging is a novel assessment tool with promising accuracy in predicting progression as an adjunct to clinical assessment in pediatric morphea. Further research should examine its performance against thermography.

Multispectral Demosaicing Based on Iterative-Linear-Regression Model for Estimating Pseudo-Panchromatic Image.

This paper proposes a method for demosaicing raw images captured by multispectral cameras. The proposed method estimates a pseudo-panchromatic image (PPI) via an iterative-linear-regression model and utilizes the estimated PPI for multispectral demosaicing. The PPI is estimated through horizontal and vertical guided filtering, with the subsampled multispectral-filter-array-(MSFA) image and low-pass-filtered MSFA as the guide image and filtering input, respectively. The number of iterations is automatically determined according to a predetermined criterion. Spectral differences between the estimated PPI and MSFA are calculated for each channel, and each spectral difference is interpolated using directional interpolation. The weights are calculated from the estimated PPI, and each interpolated spectral difference is combined using the weighted sum. The experimental results indicate that the proposed method outperforms the State-of-the-Art methods with regard to spatial and spectral fidelity for both synthetic and real-world images.

Chlorophyll Fluorescence Imaging for Environmental Stress Diagnosis in Crops.

The field of plant phenotype is used to analyze the shape and physiological characteristics of crops in multiple dimensions. Imaging, using non-destructive optical characteristics of plants, analyzes growth characteristics through spectral data. Among these, fluorescence imaging technology is a method of evaluating the physiological characteristics of crops by inducing plant excitation using a specific light source. Through this, we investigate how fluorescence imaging responds sensitively to environmental stress in garlic and can provide important information on future stress management. In this study, near UV LED (405 nm) was used to induce the fluorescence phenomenon of garlic, and fluorescence images were obtained to classify and evaluate crops exposed to abiotic environmental stress. Physiological characteristics related to environmental stress were developed from fluorescence sample images using the Chlorophyll ratio method, and classification performance was evaluated by developing a classification model based on partial least squares discrimination analysis from the image spectrum for stress identification. The environmental stress classification performance identified from the Chlorophyll ratio was 14.9% in F673/F717, 25.6% in F685/F730, and 0.209% in F690/F735. The spectrum-developed PLS-DA showed classification accuracy of 39.6%, 56.2% and 70.7% in Smoothing, MSV, and SNV, respectively. Spectrum pretreatment-based PLS-DA showed higher discrimination performance than the existing image-based Chlorophyll ratio.

Detailed Profiling of the Tumor Microenvironment in Ethnic Breast Cancer, Using Tissue Microarrays and Multiplex Immunofluorescence.

Breast cancer poses a global health challenge, yet the influence of ethnicity on the tumor microenvironment (TME) remains understudied. In this investigation, we examined immune cell infiltration in 230 breast cancer samples, emphasizing diverse ethnic populations. Leveraging tissue microarrays (TMAs) and core samples, we applied multiplex immunofluorescence (mIF) to dissect immune cell subtypes across TME regions. Our analysis revealed distinct immune cell distribution patterns, particularly enriched in aggressive molecular subtypes triple-negative and HER2-positive tumors. We observed significant correlations between immune cell abundance and key clinicopathological parameters, including tumor size, lymph node involvement, and patient overall survival. Notably, immune cell location within different TME regions showed varying correlations with clinicopathologic parameters. Additionally, ethnicities exhibited diverse distributions of cells, with certain ethnicities showing higher abundance compared to others. In TMA samples, patients of Chinese and Caribbean origin displayed significantly lower numbers of B cells, TAMs, and FOXP3-positive cells. These findings highlight the intricate interplay between immune cells and breast cancer progression, with implications for personalized treatment strategies. Moving forward, integrating advanced imaging techniques, and exploring immune cell heterogeneity in diverse ethnic cohorts can uncover novel immune signatures and guide tailored immunotherapeutic interventions, ultimately improving breast cancer management.

Preharvest phenotypic prediction of grain quality and yield of durum wheat using multispectral imaging.

Durum wheat is an important cereal that is widely grown in the Mediterranean basin. In addition to high yield, grain quality traits are of high importance for farmers. The strong influence of climatic conditions makes the improvement of grain quality traits, like protein content, vitreousness, and test weight, a challenging task. Evaluation of quality traits post-harvest is time- and labor-intensive and requires expensive equipment, such as near-infrared spectroscopes or hyperspectral imagers. Predicting not only yield but also important quality traits in the field before harvest is of high value for breeders aiming to optimize resource allocation. Implementation of efficient approaches for trait prediction, such as the use of high-resolution spectral data acquired by a multispectral camera mounted on unmanned aerial vehicles (UAVs), needs to be explored. In this study, we have acquired multispectral image data with an 11-band multispectral camera mounted on a UAV and analyzed the data with machine learning (ML) models to predict grain yield and important quality traits in breeding micro-plots. Combining 11-band multispectral data for 34 cultivars and 16 environments allowed to develop ML models with good prediction capability. Applying the trained models to test sets explained a considerable degree of phenotypic variance with good accuracy showing r squared values of 0.84, 0.69, 0.64, and 0.61 and normalized root mean squared errors of 0.17, 0.07, 0.14, and 0.03 for grain yield, protein content, vitreousness, and test weight, respectively.

P-MIRU, a Polarized Multispectral Imaging System, Reveals Reflection Information on the Biological Surface.

Reflection light forms the core of our visual perception of the world. We can obtain vast information by examining reflection light from biological surfaces, including pigment composition and distribution, tissue structure and surface microstructure. However, because of the limitations in our visual system, the complete information in reflection light, which we term 'reflectome', cannot be fully exploited. For example, we may miss reflection light information outside our visible wavelengths. In addition, unlike insects, we have virtually no sensitivity to light polarization. We can detect non-chromatic information lurking in reflection light only with appropriate devices. Although previous studies have designed and developed systems for specialized uses supporting our visual systems, we still do not have a versatile, rapid, convenient and affordable system for analyzing broad aspects of reflection from biological surfaces. To overcome this situation, we developed P-MIRU, a novel multispectral and polarization imaging system for reflecting light from biological surfaces. The hardware and software of P-MIRU are open source and customizable and thus can be applied for virtually any research on biological surfaces. Furthermore, P-MIRU is a user-friendly system for biologists with no specialized programming or engineering knowledge. P-MIRU successfully visualized multispectral reflection in visible/non-visible wavelengths and simultaneously detected various surface phenotypes of spectral polarization. The P-MIRU system extends our visual ability and unveils information on biological surfaces.

Galactosidase-catalyzed fluorescence amplification method (GAFAM): sensitive fluorescent immunohistochemistry using novel fluorogenic ß-galactosidase substrates and its application in multiplex immunostaining.

Multiplex immunohistochemistry/multiplex immunofluorescence (mIHC/mIF) enables the simultaneous detection of multiple markers in a single tissue section by visualizing the markers in different colors. Currently, tyramide signal amplification (TSA) is the most commonly used method because it is heat resistant to multiplexing. SPiDER-ßGal (6'-(diethylamino)-4'-(fluoromethyl)spiro[isobenzofuran-1(3H),9'-[9H]xanthen]-3'-yl ß-D-galactopyranoside), a novel fluorogenic substrate of ß-galactosidase (ß-gal) was reported recently. Its properties are favorable for application in sensitive mIF based on quinone methide chemistry. Combining SPiDER-ßGal with its related substrates, a novel, sensitive fluorescent IHC method for formalin-fixed paraffin-embedded (FFPE) sections was developed, named the galactosidase-catalyzed fluorescence amplification method (GAFAM). Evaluation of GAFAM indicated the following characteristics: (1) the entire GAFAM procedure was complete within a few hours; (2) the optimal working concentration of the substrates was 20 µM; (3) the fluorescent product was heat resistant; (4) the GAFAM exhibited sensitivity comparable with that of TSA, which was higher than that of conventional IF; and (5) the GAFAM was applicable to mIF and multispectral imaging. GAFAM is expected to be applicable to IF (or mIF in combination with TSA), and is a promising tool for facilitating morphological research in various fields of life science.

SUFI: an automated approach to spectral unmixing of fluorescent multiplex images captured in mouse and post-mortem human brain tissues.

BACKGROUND: Multispectral fluorescence imaging coupled with linear unmixing is a form of image data collection and analysis that allows for measuring multiple molecular signals in a single biological sample. Multiple fluorescent dyes, each measuring a unique molecule, are simultaneously measured and subsequently "unmixed" to provide a read-out for each molecular signal. This strategy allows for measuring highly multiplexed signals in a single data capture session, such as multiple proteins or RNAs in tissue slices or cultured cells, but can often result in mixed signals and bleed-through problems across dyes. Existing spectral unmixing algorithms are not optimized for challenging biological specimens such as post-mortem human brain tissue, and often require manual intervention to extract spectral signatures. We therefore developed an intuitive, automated, and flexible package called SUFI: spectral unmixing of fluorescent images. RESULTS: This package unmixes multispectral fluorescence images by automating the extraction of spectral signatures using vertex component analysis, and then performs one of three unmixing algorithms derived from remote sensing. We evaluate these remote sensing algorithms' performances on four unique biological datasets and compare the results to unmixing results obtained using ZEN Black software (Zeiss). We lastly integrate our unmixing pipeline into the computational tool dotdotdot, which is used to quantify individual RNA transcripts at single cell resolution in intact tissues and perform differential expression analysis, and thereby provide an end-to-end solution for multispectral fluorescence image analysis and quantification. CONCLUSIONS: In summary, we provide a robust, automated pipeline to assist biologists with improved spectral unmixing of multispectral fluorescence images.

Non-invasive imaging and clinical skin scores in juvenile localised scleroderma.

OBJECTIVES: To evaluate whether in juvenile localised scleroderma (JLS), non-invasive imaging can differentiate affected from non-affected skin and whether imaging correlates with a validated skin score (Localised Scleroderma Cutaneous Assessment Tool, LoSCAT). METHODS: 25 children with JLS were recruited into a prospective study and a single 'target' lesion selected. High frequency ultrasound (HFUS, measuring skin thickness), infrared thermography (IRT, skin temperature), laser Doppler imaging (LDI, skin blood flow) and multispectral imaging (MSI, oxygenation), were performed at four sites: two of affected skin (centre and inner edge of lesion) and two of non-affected skin (one cm from edge of lesion 'outer' and contralateral non-affected side), at 4 visits at 3 monthly intervals. RESULTS: Differences between affected and non-affected skin were detected with all 4 techniques. Compared with non-affected skin, affected skin was thinner (p< 0.001) with higher temperature (p< 0.001-0.006), perfusion (p< 0.001-0.039) and oxygenation (p< 0.001-0.028). Lesion skin activity (LoSCAT) was positively correlated with centre HFUS (r = 0.32; 95% CI [0.02, 0.61]; p= 0.036) and negatively correlated with centre LDI (r=-0.26; 95% CI [-0.49, -0.04]; p= 0.022). Lesion skin damage was positively correlated with centre and inner IRT (r = 0.43; 95% CI [0.19, 0.67]; p< 0.001, r = 0.36, 95% CI [0.12, 0.59]; p= 0.003, respectively) and with centre and inner LDI (r = 0.37; 95% CI [0.05, 0.69]; p= 0.024, r = 0.41; 95% CI [0.08, 0.74]; p= 0.015, respectively). CONCLUSION: Non-invasive imaging can detect differences between affected and non-affected skin in JLS and may help to differentiate between activity (thicker, less well perfused skin) and damage (thinner, highly perfused skin).

In vivo evaluation of sunscreen application by multispectral imaging: A new tool for educating sunscreen users.

BACKGROUND: The sun protection factor (SPF) of sunscreens is evaluated using standardized protocols based on the application of 2 mg/cm2 of product. However, the amount of product applied by sunscreen users in real life is likely to be much lower. OBJECTIVES: To evaluate a new multispectral imaging approach for determining the actual quantity of sunscreen applied by users and to assess the benefits of an application guide for the use of an SPF 50+ sunscreen. MATERIALS AND METHODS: Analyses of the reflectance spectra obtained from multispectral images were used to determine the actual dose of sunscreen that 26 healthy volunteers applied to their face following three application modalities: a single application, reapplication after 30 min, and application according to an instruction guide. RESULTS: Without the application guide, volunteers applied an average of 1.04 mg/cm2 of sunscreen during the single application and 1.23 mg/cm2 during the repeated application. With the application guide, the amount of sunscreen applied was 1.45 mg/cm2 : around 40% higher than during the single application. Spreading of the sunscreen was also less uniform with the unguided single application than with the other application modalities. CONCLUSIONS: This study showed that the multispectral imaging approach can be used to measure the amount of sunscreen applied in vivo. Our findings confirmed that the standard dose used for SPF measurements and other sunscreen tests is far higher than that applied by users in practice. Providing users with precise guidelines could increase the amount of sunscreen applied, resulting in more adequate photoprotection.

A Rehabilitation of Pixel-Based Spectral Reconstruction from RGB Images.

Recently, many deep neural networks (DNN) have been proposed to solve the spectral reconstruction (SR) problem: recovering spectra from RGB measurements. Most DNNs seek to learn the relationship between an RGB viewed in a given spatial context and its corresponding spectra. Significantly, it is argued that the same RGB can map to different spectra depending on the context with respect to which it is seen and, more generally, that accounting for spatial context leads to improved SR. However, as it stands, DNN performance is only slightly better than the much simpler pixel-based methods where spatial context is not used. In this paper, we present a new pixel-based algorithm called A++ (an extension of the A+ sparse coding algorithm). In A+, RGBs are clustered, and within each cluster, a designated linear SR map is trained to recover spectra. In A++, we cluster the spectra instead in an attempt to ensure neighboring spectra (i.e., spectra in the same cluster) are recovered by the same SR map. A polynomial regression framework is developed to estimate the spectral neighborhoods given only the RGB values in testing, which in turn determines which mapping should be used to map each testing RGB to its reconstructed spectrum. Compared to the leading DNNs, not only does A++ deliver the best results, it is parameterized by orders of magnitude fewer parameters and has a significantly faster implementation. Moreover, in contradistinction to some DNN methods, A++ uses pixel-based processing, which is robust to image manipulations that alter the spatial context (e.g., blurring and rotations). Our demonstration on the scene relighting application also shows that, while SR methods, in general, provide more accurate relighting results compared to the traditional diagonal matrix correction, A++ provides superior color accuracy and robustness compared to the top DNN methods.

Microbiological Quality Estimation of Meat Using Deep CNNs on Embedded Hardware Systems.

Spectroscopic sensor imaging of food samples meta-processed by deep machine learning models can be used to assess the quality of the sample. This article presents an architecture for estimating microbial populations in meat samples using multispectral imaging and deep convolutional neural networks. The deep learning models operate on embedded platforms and not offline on a separate computer or a cloud server. Different storage conditions of the meat samples were used, and various deep learning models and embedded platforms were evaluated. In addition, the hardware boards were evaluated in terms of latency, throughput, efficiency and value on different data pre-processing and imaging-type setups. The experimental results showed the advantage of the XavierNX platform in terms of latency and throughput and the advantage of Nano and RP4 in terms of efficiency and value, respectively.

Improving Generalizability of Spectral Reflectance Reconstruction Using L1-Norm Penalization.

Spectral reflectance reconstruction for multispectral images (such as Weiner estimation) may perform sub-optimally when the object being measured has a texture that is not in the training set. The accuracy of the reconstruction is significantly lower without training samples. We propose an improved reflectance reconstruction method based on L1-norm penalization to solve this issue. Using L1-norm, our method can provide the transformation matrix with the favorable sparse property, which can help to achieve better results when measuring the unseen samples. We verify the proposed method by reconstructing spectral reflection for four types of materials (cotton, paper, polyester, and nylon) captured by a multispectral imaging system. Each of the materials has its texture and there are 204 samples in each of the materials/textures in the experiments. The experimental results show that when the texture is not included in the training dataset, L1-norm can achieve better results compared with existing methods using colorimetric measure (i.e., color difference) and shows consistent accuracy across four kinds of materials.

Design of a Dual-Mode Multispectral Filter Array.

Multispectral imaging is valuable in many vision-related fields as it provides an additional modality to observe the world. Cameras equipped with multispectral filter arrays (MSFAs) are typically impractical for everyday use due to their intractable demosaicking and chromatic reproduction processes, which restrict their applicability beyond academic research. In this work, a novel MSFA design is proposed to enable dual-mode imaging for multispectral cameras. In addition to a conventional multispectral image, the camera is also able to produce a Bayer-formed RGB image from a single shot by grouping and merging adjacent pixels in the proposed MSFA, making it suitable for scenarios where display-ready RGB images are required. Furthermore, a two-stage optimization scheme is implemented to jointly optimize objective functions for both imaging modes. The evaluation results on multiple datasets suggest that the proposed MSFA design is able to simultaneously achieve competitive spectral reconstruction accuracy compared to elaborate multispectral cameras and chromatic accuracy compared to commercial RGB cameras.

Quasi Real-Time Apple Defect Segmentation Using Deep Learning.

Defect segmentation of apples is an important task in the agriculture industry for quality control and food safety. In this paper, we propose a deep learning approach for the automated segmentation of apple defects using convolutional neural networks (CNNs) based on a U-shaped architecture with skip-connections only within the noise reduction block. An ad-hoc data synthesis technique has been designed to increase the number of samples and at the same time to reduce neural network overfitting. We evaluate our model on a dataset of multi-spectral apple images with pixel-wise annotations for several types of defects. In this paper, we show that our proposal outperforms in terms of segmentation accuracy general-purpose deep learning architectures commonly used for segmentation tasks. From the application point of view, we improve the previous methods for apple defect segmentation. A measure of the computational cost shows that our proposal can be employed in real-time (about 100 frame-per-second on GPU) and in quasi-real-time (about 7/8 frame-per-second on CPU) visual-based apple inspection. To further improve the applicability of the method, we investigate the potential of using only RGB images instead of multi-spectral images as input images. The results prove that the accuracy in this case is almost comparable with the multi-spectral case.

High-Dynamic-Range Spectral Reflectance for the Segmentation of Paint Pigment: Application to Dalí's Oil Painting Dos Figuras (1926).

Restorers and curators in museums sometimes find it difficult to accurately segment areas of paintings that have been contaminated with other pigments or areas that need to be restored, and work on the painting needs to be carried out with minimum possible damage. It is therefore necessary to develop measurement systems and methods that facilitate this task in the least invasive way possible. The aim of this study was to obtain high-dynamic-range (HDR) spectral reflectance and spatial resolution for Dalí's painting entitled Two Figures (1926) in order to segment a small area of black and white pigment that was affected by the contact transfer of reddish pigment from another painting. Using Hypermatrixcam to measure the HDR spectral reflectance developed by this research team, an HDR multispectral cube of 12 images was obtained for the band 470-690 nm in steps of 20 nm. With the values obtained for the spectral reflectance of the HDR cube, the colour of the area of paint affected by the transfer was studied by calculating the a*b* components with the CIELab system. These a*b* values were then used to define two methods of segmenting the exact areas in which there was a transfer of reddish pigment. The area studied in the painting was originally black, and the contamination with reddish pigment occupied 13.87% to 32% of the total area depending on the selected method. These different solutions can be explained because the lower limit is segmentation based on pure pigment and the upper limit considers red as an exclusion of non-black pigment. Over- and under-segmentation is a common problem described in the literature related to pigment selection. In this application case, as red pigment is not original and should be removed, curators will choose the method that selects the highest red area.

Multispectral Imaging for Skin Diseases Assessment-State of the Art and Perspectives.

Skin optical inspection is an imperative procedure for a suspicious dermal lesion since very early skin cancer detection can guarantee total recovery. Dermoscopy, confocal laser scanning microscopy, optical coherence tomography, multispectral imaging, multiphoton laser imaging, and 3D topography are the most outstanding optical techniques implemented for skin examination. The accuracy of dermatological diagnoses attained by each of those methods is still debatable, and only dermoscopy is frequently used by all dermatologists. Therefore, a comprehensive method for skin analysis has not yet been established. Multispectral imaging (MSI) is based on light-tissue interaction properties due to radiation wavelength variation. An MSI device collects the reflected radiation after illumination of the lesion with light of different wavelengths and provides a set of spectral images. The concentration maps of the main light-absorbing molecules in the skin, the chromophores, can be retrieved using the intensity values from those images, sometimes even for deeper-located tissues, due to interaction with near-infrared light. Recent studies have shown that portable and cost-efficient MSI systems can be used for extracting skin lesion characteristics useful for early melanoma diagnoses. This review aims to describe the efforts that have been made to develop MSI systems for skin lesions evaluation in the last decade. We examined the hardware characteristics of the produced devices and identified the typical structure of an MSI device for dermatology. The analyzed prototypes showed the possibility of improving the specificity of classification between the melanoma and benign nevi. Currently, however, they are rather adjuvants tools for skin lesion assessment, and efforts are needed towards a fully fledged diagnostic MSI device.

Combining Feature Selection Techniques and Neurofuzzy Systems for the Prediction of Total Viable Counts in Beef Fillets Using Multispectral Imaging.

In the food industry, quality and safety issues are associated with consumers' health condition. There is a growing interest in applying various noninvasive sensorial techniques to obtain quickly quality attributes. One of them, hyperspectral/multispectral imaging technique has been extensively used for inspection of various food products. In this paper, a stacking-based ensemble prediction system has been developed for the prediction of total viable counts of microorganisms in beef fillet samples, an essential cause to meat spoilage, utilizing multispectral imaging information. As the selection of important wavelengths from the multispectral imaging system is considered as an essential stage to the prediction scheme, a features fusion approach has been also explored, by combining wavelengths extracted from various feature selection techniques. Ensemble sub-components include two advanced clustering-based neuro-fuzzy network prediction models, one utilizing information from average reflectance values, while the other one from the standard deviation of the pixels' intensity per wavelength. The performances of neurofuzzy models were compared against established regression algorithms such as multilayer perceptron, support vector machines and partial least squares. Obtained results confirmed the validity of the proposed hypothesis to utilize a combination of feature selection methods with neurofuzzy models in order to assess the microbiological quality of meat products.