Lossy Image Compression in a Preclinical Multimodal Imaging Study.

The growing use of multimodal high-resolution volumetric data in pre-clinical studies leads to challenges related to the management and handling of the large amount of these datasets. Contrarily to the clinical context, currently there are no standard guidelines to regulate the use of image compression in pre-clinical contexts as a potential alleviation of this problem. In this work, the authors study the application of lossy image coding to compress high-resolution volumetric biomedical data. The impact of compression on the metrics and interpretation of volumetric data was quantified for a correlated multimodal imaging study to characterize murine tumor vasculature, using volumetric high-resolution episcopic microscopy (HREM), micro-computed tomography (µCT), and micro-magnetic resonance imaging (µMRI). The effects of compression were assessed by measuring task-specific performances of several biomedical experts who interpreted and labeled multiple data volumes compressed at different degrees. We defined trade-offs between data volume reduction and preservation of visual information, which ensured the preservation of relevant vasculature morphology at maximum compression efficiency across scales. Using the Jaccard Index (JI) and the average Hausdorff Distance (HD) after vasculature segmentation, we could demonstrate that, in this study, compression that yields to a 256-fold reduction of the data size allowed to keep the error induced by compression below the inter-observer variability, with minimal impact on the assessment of the tumor vasculature across scales.

Melanoma classification using light-Fields with morlet scattering transform and CNN: Surface depth as a valuable tool to increase detection rate.

Medical image classification through learning-based approaches has been increasingly used, namely in the discrimination of melanoma. However, for skin lesion classification in general, such methods commonly rely on dermoscopic or other 2D-macro RGB images. This work proposes to exploit beyond conventional 2D image characteristics, by considering a third dimension (depth) that characterises the skin surface rugosity, which can be obtained from light-field images, such as those available in the SKINL2 dataset. To achieve this goal, a processing pipeline was deployed using a morlet scattering transform and a CNN model, allowing to perform a comparison between using 2D information, only 3D information, or both. Results show that discrimination between Melanoma and Nevus reaches an accuracy of 84.00, 74.00 or 94.00% when using only 2D, only 3D, or both, respectively. An increase of 14.29pp in sensitivity and 8.33pp in specificity is achieved when expanding beyond conventional 2D information by also using depth. When discriminating between Melanoma and all other types of lesions (a further imbalanced setting), an increase of 28.57pp in sensitivity and decrease of 1.19pp in specificity is achieved for the same test conditions. Overall the results of this work demonstrate significant improvements over conventional approaches.

Skin lesion classification using features of 3D border lines.

Machine learning algorithms are progressively assuming important roles as computational tools to support clinical diagnosis, namely in the classification of pigmented skin lesions using RGB images. Most current classification methods rely on common 2D image features derived from shape, colour or texture, which does not always guarantee the best results. This work presents a contribution to this field, by exploiting the lesions' border line characteristics using a new dimension - depth, which has not been thoroughly investigated so far. A selected group of features is extracted from the depth information of 3D images, which are then used for classification using a quadratic Support Vector Machine. Despite class imbalance often present in medical image datasets, the proposed algorithm achieves a top geometric mean of 94.87%, comprising 100.00% sensitivity and 90.00% specificity, using only depth information for the detection of Melanomas. Such results show that potential gains can be achieved by extracting information from this often overlooked dimension, which provides more balanced results in terms of sensitivity and specificity than other settings.

Light Field Image Dataset of Skin Lesions.

Light field imaging technology has been attracting increasing interest because it enables capturing enriched visual information and expands the processing capabilities of traditional 2D imaging systems. Dense multiview, accurate depth maps and multiple focus planes are examples of different types of visual information enabled by light fields. This technology is also emerging in medical imaging research, like dermatology, allowing to find new features and improve classification algorithms, namely those based on machine learning approaches. This paper presents a contribution for the research community, in the form of a publicly available light field image dataset of skin lesions (named SKINL2 v1.0). This dataset contains 250 light fields, captured with a focused plenoptic camera and classified into eight clinical categories, according to the type of lesion. Each light field is comprised of 81 different views of the same lesion. The database also includes the dermatoscopic image of each lesion. A representative subset of 17 central view images of the light fields is further characterised in terms of spatial information (SI), colourfulness (CF) and compressibility. This dataset has high potential for advancing medical imaging research and development of new classification algorithms based on light fields, as well as in clinically-oriented dermatology studies.