Suchergebnisse

Deep learning diagnostic algorithms are proving comparable results with human experts in a wide variety of tasks, and they still require a huge amount of well-annotated data for training, which is often non affordable. Metric learning techniques have allowed a reduction in the required annotated data allowing few-shot learning over deep learning architectures. Aims and Objectives: In this work, we analyze the state-of-the-art loss functions such as triplet loss, contrastive loss, and multi-class N-pair loss for the visual embedding extraction of hematoxylin and eosin (H&E) microscopy images and we propose a novel constellation loss function that takes advantage of the visual distances of the embeddings of the negative samples and thus, performing a regularization that increases the quality of the extracted embeddings. Materials and Methods: To this end, we employed the public H&E imaging dataset from the University Medical Center Mannheim (Germany) that contains tissue samples from low-grade and high-grade primary tumors of digitalized colorectal cancer tissue slides. These samples are divided into eight different textures (1. tumour epithelium, 2. simple stroma, 3. complex stroma, 4. immune cells, 5. debris and mucus, 6. mucosal glands, 7. adipose tissue and 8. background,). The dataset was divided randomly into train and test splits and the training split was used to train a classifier to distinguish among the different textures with just 20 training images. The process was repeated 10 times for each loss function. Performance was compared both for cluster compactness and for classification accuracy on separating the aforementioned textures. Results: Our results show that the proposed loss function outperforms the other methods by obtaining more compact clusters (Davis-Boulding: 1.41 ± 0.08, Silhouette: 0.37 ± 0.02) and better classification capabilities (accuracy: 85.0 ± 0.6) over H and E microscopy images. We demonstrate that the proposed constellation loss can be successfully used in the medical domain in situations of data scarcity. ; This work was partially supported by PICCOLO project. This project has received funding from the European Union's Horizon2020 Research and Innovation Programme under grant agreement No. 732111.

Estimating the temperature of hot emissive samples (e.g. liquid slag) in the context of harsh industrial environments such as steelmaking plants is a crucial yet challenging task, which is typically addressed by means of methods that require physical contact. Current remote methods require information on the emissivity of the sample. However, the spectral emissivity is dependent on the sample composition and temperature itself, and it is hardly measurable unless under controlled laboratory procedures. In this work, we present a portable device and associated probabilistic model that can simultaneously produce quasi real-time estimates for temperature and spectral emissivity of hot samples in the [0.2, 12.0μm ] range at distances of up to 20m . The model is robust against variable atmospheric conditions, and the device is presented together with a quick calibration procedure that allows for in field deployment in rough industrial environments, thus enabling in line measurements. We validate the temperature and emissivity estimates by our device against laboratory equipment under controlled conditions in the [550, 850∘C ] temperature range for two solid samples with well characterized spectral emissivity's: alumina ( α−Al2O3 ) and hexagonal boron nitride ( h−BN ). The analysis of the results yields Root Mean Squared Errors of 32.3∘C and 5.7∘C respectively, and well correlated spectral emissivity's. ; This work was supported in part by the Basque Government (Hazitek AURRERA B: Advanced and Useful REdesign of CSP process for new steel gRAdes) under Grant ZE-2017/00009.

Although deep learning pathology diagnostic algorithms are proving comparable results with human experts in a wide variety of tasks, they still require a huge amount of well annotated data for training. Generating such extensive and well labelled datasets is time consuming and is not feasible for certain tasks and so, most of the medical datasets available are scarce in images and therefore, not enough for training. In this work we validate that the use of few shot learning techniques can transfer knowledge from a well defined source domain from Colon tissue into a more generic domain composed by Colon, Lung and Breast tissue by using very few training images. Our results show that our few-shot approach is able to obtain a balanced accuracy (BAC) of 90% with just 60 training images, even for the Lung and Breast tissues that were not present on the training set. This outperforms the finetune transfer learning approach that obtains 73% BAC with 60 images and requires 600 images to get up to 81% BAC. ; This study has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 732111 (PICCOLO project).

Plant fungal diseases are one of the most important causes of crop yield losses. Therefore, plant disease identification algorithms have been seen as a useful tool to detect them at early stages to mitigate their effects. Although deep-learning based algorithms can achieve high detection accuracies, they require large and manually annotated image datasets that is not always accessible, specially for rare and new diseases. This study focuses on the development of a plant disease detection algorithm and strategy requiring few plant images (Few-shot learning algorithm). We extend previous work by using a novel challenging dataset containing more than 100,000 images. This dataset includes images of leaves, panicles and stems of five different crops (barley, corn, rape seed, rice, and wheat) for a total of 17 different diseases, where each disease is shown at different disease stages. In this study, we propose a deep metric learning based method to extract latent space representations from plant diseases with just few images by means of a Siamese network and triplet loss function. This enhances previous methods that require a support dataset containing a high number of annotated images to perform metric learning and few-shot classification. The proposed method was compared over a traditional network that was trained with the cross-entropy loss function. Exhaustive experiments have been performed for validating and measuring the benefits of metric learning techniques over classical methods. Results show that the features extracted by the metric learning based approach present better discriminative and clustering properties. Davis-Bouldin index and Silhouette score values have shown that triplet loss network improves the clustering properties with respect to the categorical-cross entropy loss. Overall, triplet loss approach improves the DB index value by 22.7% and Silhouette score value by 166.7% compared to the categorical cross-entropy loss model. Moreover, the F-score parameter obtained from the Siamese network with the triplet loss performs better than classical approaches when there are few images for training, obtaining a 6% improvement in the F-score mean value. Siamese networks with triplet loss have improved the ability to learn different plant diseases using few images of each class. These networks based on metric learning techniques improve clustering and classification results over traditional categorical cross-entropy loss networks for plant disease identification. ; This project was partially supported by the Spanish Government through CDTI Centro para el Desarrollo Tecnológico e Industrial project AI4ES (ref CER-20211030), by the University of the Basque Country (UPV/EHU) under grant COLAB20/01 and by the Basque Government through grant IT1229-19.

Colorectal cancer presents one of the most elevated incidences of cancer worldwide. Colonoscopy relies on histopathology analysis of hematoxylin-eosin (H&E) images of the removed tissue. Novel techniques such as multi-photon microscopy (MPM) show promising results for performing real-time optical biopsies. However, clinicians are not used to this imaging modality and correlation between MPM and H&E information is not clear. The objective of this paper is to describe and make publicly available an extensive dataset of fully co-registered H&E and MPM images that allows the research community to analyze the relationship between MPM and H&E histopathological images and the effect of the semantic gap that prevents clinicians from correctly diagnosing MPM images. The dataset provides a fully scanned tissue images at 10x optical resolution (0.5 µm/px) from 50 samples of lesions obtained by colonoscopies and colectomies. Diagnostics capabilities of TPF and H&E images were compared. Additionally, TPF tiles were virtually stained into H&E images by means of a deep-learning model. A panel of 5 expert pathologists evaluated the different modalities into three classes (healthy, adenoma/hyperplastic, and adenocarcinoma). Results showed that the performance of the pathologists over MPM images was 65% of the H&E performance while the virtual staining method achieved 90%. MPM imaging can provide appropriate information for diagnosing colorectal cancer without the need for H&E staining. However, the existing semantic gap among modalities needs to be corrected. ; This work was supported by the PICCOLO project. This project has received funding from the European Union's Horizon 2020 Research and Innovation Programme under grant agreement No. 732111. The sole responsibility of this publication lies with the authors. The European Union is not responsible for any use that may be made of the information contained therein. This research has also been supported by the project ONKOTOOLS (KK2020/00069) funded by the Basque Government Industry Department under the ELKARTEK program.