José Manuel Prats-Montalbán

Automatic Individual Arterial Input Functions Calculated From PCA Outperform Manual and Population-Averaged Approaches for the Pharmacokinetic Modeling of DCE-MR Images

To introduce a segmentation method to calculate an automatic arterial input function (AIF) based on principal component analysis (PCA) of dynamic contrast enhanced MR (DCE‐MR) imaging and compare it with individual manually selected and population‐averaged AIFs using calculated pharmacokinetic parameters.

Assessment of cluster yield components by image analysis

BACKGROUND Berry weight, berry number and cluster weight are key parameters for yield estimation for wine and tablegrape industry. Current yield prediction methods are destructive, labour-demanding and time-consuming. In this work, a new methodology, based on image analysis was developed to determine cluster yield components in a fast and inexpensive way. RESULTS Clusters of seven different red varieties of grapevine (Vitis vinifera L.) were photographed under laboratory conditions and their cluster yield components manually determined after image acquisition. Two algorithms based on the Canny and the logarithmic image processing approaches were tested to find the contours of the berries in the images prior to berry detection performed by means of the Hough Transform. Results were obtained in two ways: by analysing either a single image of the cluster or using four images per cluster from different orientations. The best results (R(2) between 69% and 95% in berry detection and between 65% and 97% in cluster weight estimation) were achieved using four images and the Canny algorithm. The models capability based on image analysis to predict berry weight was 84%. CONCLUSION The new and low-cost methodology presented here enabled the assessment of cluster yield components, saving time and providing inexpensive information in comparison with current manual methods.

Computer vision for automatic quality inspection of dried figs (Ficus carica L.) in real-time

Dried fig skin colour was assessed comparing image analysis and colourimetry.PCA and PLS-DA distinguished between high quality figs and deteriorated ones.A system based on computer vision for sorting of figs in real-time was developed.The browning index and X colour coordinate guaranteed an accurate sorting. This work reports the development of automated systems based on computer vision to improve the quality control and sorting of dried figs of Cosenza (protected denomination of origin) focusing on two research issues. The first was based on qualitative discrimination of figs through colour assessment comparing the analysis of colour images obtained using a digital camera with those obtained according to conventional instrumental methods, i.e. colourimetry currently done in laboratories. Data were expressed in terms of CIE XYZ, CIELAB and HunterLab colour spaces, as well as the browning index measurement of each fruit, and then, analysed using PCA and PLS-DA based methods. The results showed that both chroma meter and image analysis allowed a complete distinction between high quality and deteriorated figs, according to colour attributes. The second research issue had the purpose of developing image processing algorithms to achieve real-time sorting of figs using an experimental prototype based on machine vision, simulating an industrial application. An extremely high 99.5% of deteriorated figs were classified correctly as well as 89.0% of light coloured good quality figs A lower percentage was obtained for dark good quality figs but results were acceptable since the most of the confusion was among the two classes of good product.

Multivariate statistical projection methods to perform robust feature extraction and classification in surface grading

We present an innovative way to simultaneously perform feature extraction and classification for the quality-control issue of surface grading by applying two multivariate statistical projection methods: SIMCA and PLS-DA. These tools have been applied to compress the color texture data that describe the visual appearance of surfaces (soft color texture descriptors) and to directly perform classification using statistics and predictions from the projection models. Experiments have been carried out using an extensive ce- ramic images database (VxC TSG) comprised of 14 different mod- els, 42 surface classes, and 960 pieces. A factorial experimental design evaluated all the combinations of several factors affecting the accuracy rate. These factors include the tile model, color repre- sentation scheme (CIE Lab, CIE Luv, and RGB), and compression/ classification approach (SIMCA and PLS-DA). Moreover, a logistic regression model is fitted from the experiments to compute accu- racy estimates and study the effect of the factors on the accuracy rate. Results show that PLS-DA performs better than SIMCA, achieving a mean accuracy rate of 98.95%. These results outper- form those obtained in a previous work where the soft color texture descriptors in combination with the CIE Lab color space and the k-NN classifier achieved an accuracy rate of 97.36%.

N‐way modeling for wavelet filter determination in multivariate image analysis

When trying to analyze spatial relationships in image analysis, wavelets appear as one of the state‐of‐the‐art tools. However, image analysis is a problem‐dependent issue, and different applications might require different wavelets in order to gather the main sources of variation in the acquired images with respect to the specific task to be performed. This paper provides a methodology based on N‐way modeling for properly selecting the best wavelet choice to use or at least to provide a range of possible wavelet choices (in terms of families, filters, and decomposition levels), for each image and problem at hand. The methodology has been applied on two different data sets with exploratory and monitoring objectives. Copyright

Prostate functional magnetic resonance image analysis using multivariate curve resolution methods

This paper discusses the potential of multivariate curve resolution models to extract physiological dynamics behaviors from dynamic contrast‐enhanced magnetic resonance imaging prostate perfusion studies for cancer diagnosis. A relationship with biomarkers (“hidden” parameters for assessing the possible existence of a tumor) from pharmacokinetic models is also studied. Copyright

Segmentation techniques in image analysis: A comparative study: Segmentation techniques in image analysis: a comparative study

Nowadays, the detection, localization, and quantification of different kinds of features in an RGB image (segmentation) is extremely helpful for, e.g., process monitoring or customer product acceptance. In this article, some of the most commonly used RGB image segmentation approaches are compared in an orange quality control case study. Analysis of variance and correspondence analysis are combined for determining their most relevant differences and highlighting their pros and cons.

Multivariate image analysis and near infrared chemical imaging for characterisation of micro-mixing in polymeric thin films

Introduction P harmaceutical regulations require sampling and testing of in-process materials and drug products to evaluate the adequacy of mixing so as to assure uniformity and homogeneity. The optimal determination of the distribution of the drug and excipients affects blend homogeneity, content uniformity, and may also affect dissolution. These issues are related, not only to the manufacturing process, but also to the solubility of the active pharmaceutical ingredient (API). One possible way for improving the solubility of these drugs (40% of all new active drug candidates have very low solubility) is dispersion in a polymeric film. Therefore, the ability to visualise and assess the compositional heterogeneity and structure of the end-products is extremely important for the design, development and manufacture of polymeric films. Process understanding and product design are essential for process analytical technology (PAT) and quality by design (QbD). Spectroscopic chemical imaging methods, such as near infrared chemical imaging (NIR-CI) permit an intra-unit definition of drug distribution by providing reliable chemical and spatial information on the distribution of drug and excipients which has an influence on the micro-mixing properties. Micro-mixing describes how particles from different ingredients interact with each other to form a blend with certain properties such as degree of agglomeration, cohesion, hydrophobicity and electric conductivity. Regarding this issue, multivariate curve resolution (MCR) segregates the information linked to each of the chemical compounds in the mixture, by using the (hyper)spectral correlation, in a chemical sense. This way, we obtain full chemically-interpretable images (the socalled chemical distribution maps, CDMs), in which the chemical compounds of the mixture appear in each separated “chemical channel (or band)” and are distributed according to their corresponding chemical concentration, converted into a grey level intensity. CDMs are afterwards analysed in a univariate way. However, the information extracted this way might not be sufficient for process monitoring or final quality prediction purposes, since it is not only the different distribution and concentration of the chemical compounds in the image that counts but also the way in which they combine. In other words, the correlation structure between and within the CDM’s segregated chemical compounds, both in terms of chemical and spatial (textural and physicochemical mixture properties) information is not used. In order to analyse these correlation structures of the mixtures, thereby obtaining information about the micro-mixing properties, multivariate image analysis (MIA) of the CDMs may be useful, unravelling the different behaviours in separate PCs. Properties may depend on the correlation structure of the distribution of the different chemical compounds. The importance of each type of information can be assessed and linked to the final quality properties. Thus, the aforementioned PAT and QbD goals (process understanding, process and product design, and final quality) can be more effectively achieved, obtaining better monitoring and predictive models. The present work reports a three-step methodology to analyse the chemical composition and spatial relationships between API and different excipients. The first step consists of the application of MCR to separate the chemical information in the hyperspectral images. In a second, MIA is applied to obtain meaningful and complementary improved information from the images related to each chemical compound in the mixtures. Finally, a third step uses these MIA score images to extract features able to characterise quality micromixing properties of the images.

Multivariate Curve Resolution for Magnetic Resonance Image Analysis: Applications in Prostate Cancer Biomarkers Development

Abstract One current trending topic in medical imaging is imaging biomarkers, which are being developed for early cancer detection in several organs such as lung, breast, liver, colon, prostate, or brain. These imaging biomarkers, which act as indicators of a normal biological process, a disease or a response to a therapeutic intervention, rely on (hard) physiological models; making the interpretation difficult in some cases. In this work, multivariate curve resolution (MCR) is applied on magnetic resonance images from perfusion (dynamic contrast enhanced-magnetic resonance) and diffusion (diffusion weighted-magnetic resonance) sequences, in the case of prostate cancer. MCR provides new imaging biomarkers, complementary to those obtained from theoretical models, to improve clinical diagnosis, as well as an evaluation tool for assessing the appropriateness of theoretical models.

Comparison of latent variable-based and artificial intelligence methods for impurity detection in PET recycling from NIR hyperspectral images: DoE and MIA for impurities detection optimization in PET recycling

In polyethylene terephthalates (PET)s recycling processes, separation from polyvinyl chloride (PVC) is of prior relevance due to its toxicity, which degrades the final quality of recycled PET. Moreover, the potential presence of some polymers in mixed plastics (such as PVC in PET) is a key aspect for the use of recycled plastic in products such as medical equipment, toys, or food packaging.