RESUMO
Challenges have become the state-of-the-art approach to benchmark image analysis algorithms in a comparative manner. While the validation on identical data sets was a great step forward, results analysis is often restricted to pure ranking tables, leaving relevant questions unanswered. Specifically, little effort has been put into the systematic investigation on what characterizes images in which state-of-the-art algorithms fail. To address this gap in the literature, we (1) present a statistical framework for learning from challenges and (2) instantiate it for the specific task of instrument instance segmentation in laparoscopic videos. Our framework relies on the semantic meta data annotation of images, which serves as foundation for a General Linear Mixed Models (GLMM) analysis. Based on 51,542 meta data annotations performed on 2,728 images, we applied our approach to the results of the Robust Medical Instrument Segmentation Challenge (ROBUST-MIS) challenge 2019 and revealed underexposure, motion and occlusion of instruments as well as the presence of smoke or other objects in the background as major sources of algorithm failure. Our subsequent method development, tailored to the specific remaining issues, yielded a deep learning model with state-of-the-art overall performance and specific strengths in the processing of images in which previous methods tended to fail. Due to the objectivity and generic applicability of our approach, it could become a valuable tool for validation in the field of medical image analysis and beyond.
Assuntos
Algoritmos , Laparoscopia , Humanos , Processamento de Imagem Assistida por Computador/métodosRESUMO
Primary Central Nervous System Lymphoma (PCNSL) is an aggressive neoplasm with a poor prognosis. Although therapeutic progresses have significantly improved Overall Survival (OS), a number of patients do not respond to HD-MTX-based chemotherapy (15-25%) or experience relapse (25-50%) after an initial response. The reasons underlying this poor response to therapy are unknown. Thus, there is an urgent need to develop improved predictive models for PCNSL. In this study, we investigated whether radiomics features can improve outcome prediction in patients with PCNSL. A total of 80 patients diagnosed with PCNSL were enrolled. A patient sub-group, with complete Magnetic Resonance Imaging (MRI) series, were selected for the stratification analysis. Following radiomics feature extraction and selection, different Machine Learning (ML) models were tested for OS and Progression-free Survival (PFS) prediction. To assess the stability of the selected features, images from 23 patients scanned at three different time points were used to compute the Interclass Correlation Coefficient (ICC) and to evaluate the reproducibility of each feature for both original and normalized images. Features extracted from Z-score normalized images were significantly more stable than those extracted from non-normalized images with an improvement of about 38% on average (p-value < 10-12). The area under the ROC curve (AUC) showed that radiomics-based prediction overcame prediction based on current clinical prognostic factors with an improvement of 23% for OS and 50% for PFS, respectively. These results indicate that radiomics features extracted from normalized MR images can improve prognosis stratification of PCNSL patients and pave the way for further study on its potential role to drive treatment choice.
RESUMO
Lesion segmentation in medical images is difficult yet crucial for proper diagnosis and treatment. Identifying lesions in medical images is costly and time-consuming and requires highly specialized knowledge. For this reason, supervised and semi-supervised learning techniques have been developed. Nevertheless, the lack of annotated data, which is common in medical imaging, is an issue; in this context, interesting approaches can use unsupervised learning to accurately distinguish between healthy tissues and lesions, training the network without using the annotations. In this work, an unsupervised learning technique is proposed to automatically segment coronavirus disease 2019 (COVID-19) lesions on 2D axial CT lung slices. The proposed approach uses the technique of image translation to generate healthy lung images based on the infected lung image without the need for lesion annotations. Attention masks are used to improve the quality of the segmentation further. Experiments showed the capability of the proposed approaches to segment the lesions, and it outperforms a range of unsupervised lesion detection approaches. The average reported results for the test dataset based on the metrics: Dice Score, Sensitivity, Specificity, Structure Measure, Enhanced-Alignment Measure, and Mean Absolute Error are 0.695, 0.694, 0.961, 0.791, 0.875, and 0.082 respectively. The achieved results are promising compared with the state-of-the-art and could constitute a valuable tool for future developments.
Assuntos
COVID-19 , COVID-19/diagnóstico por imagem , Humanos , Processamento de Imagem Assistida por Computador/métodos , Pulmão/diagnóstico por imagem , Cintilografia , Tórax , Tomografia Computadorizada por Raios X/métodosRESUMO
The coronavirus disease 19 (COVID-19) pandemic is having a dramatic impact on society and healthcare systems. In this complex scenario, lung computerized tomography (CT) may play an important prognostic role. However, datasets released so far present limitations that hamper the development of tools for quantitative analysis. In this paper, we present an open-source lung CT dataset comprising information on 50 COVID-19-positive patients. The CT volumes are provided along with (i) an automatic threshold-based annotation obtained with a Gaussian mixture model (GMM) and (ii) a scoring provided by an expert radiologist. This score was found to significantly correlate with the presence of ground glass opacities and the consolidation found with GMM. The dataset is freely available in an ITK-based file format under the CC BY-NC 4.0 license. The code for GMM fitting is publicly available, as well. We believe that our dataset will provide a unique opportunity for researchers working in the field of medical image analysis, and hope that its release will lay the foundations for the successfully implementation of algorithms to support clinicians in facing the COVID-19 pandemic.
RESUMO
Background and ObjectivesOver the last decade, Deep Learning (DL) has revolutionized data analysis in many areas, including medical imaging. However, there is a bottleneck in the advancement of DL in the surgery field, which can be seen in a shortage of large-scale data, which in turn may be attributed to the lack of a structured and standardized methodology for storing and analyzing surgical images in clinical centres. Furthermore, accurate annotations manually added are expensive and time consuming. A great help can come from the synthesis of artificial images; in this context, in the latest years, the use of Generative Adversarial Neural Networks (GANs) achieved promising results in obtaining photo-realistic images. MethodsIn this study, a method for Minimally Invasive Surgery (MIS) image synthesis is proposed. To this aim, the generative adversarial network pix2pix is trained to generate paired annotated MIS images by transforming rough segmentation of surgical instruments and tissues into realistic images. An additional regularization term was added to the original optimization problem, in order to enhance realism of surgical tools with respect to the background. Results Quantitative and qualitative (i.e., human-based) evaluations of generated images have been carried out in order to assess the effectiveness of the method. ConclusionsExperimental results show that the proposed method is actually able to translate MIS segmentations to realistic MIS images, which can in turn be used to augment existing data sets and help at overcoming the lack of useful images; this allows physicians and algorithms to take advantage from new annotated instances for their training.
Assuntos
Processamento de Imagem Assistida por Computador , Laparoscopia , Algoritmos , Humanos , Redes Neurais de ComputaçãoRESUMO
Artificial intelligence (AI) is a major branch of computer science that is fruitfully used for analyzing complex medical data and extracting meaningful relationships in datasets, for several clinical aims. Specifically, in the brain care domain, several innovative approaches have achieved remarkable results and open new perspectives in terms of diagnosis, planning, and outcome prediction. In this work, we present an overview of different artificial intelligent techniques used in the brain care domain, along with a review of important clinical applications. A systematic and careful literature search in major databases such as Pubmed, Scopus, and Web of Science was carried out using "artificial intelligence" and "brain" as main keywords. Further references were integrated by cross-referencing from key articles. 155 studies out of 2696 were identified, which actually made use of AI algorithms for different purposes (diagnosis, surgical treatment, intra-operative assistance, and postoperative assessment). Artificial neural networks have risen to prominent positions among the most widely used analytical tools. Classic machine learning approaches such as support vector machine and random forest are still widely used. Task-specific algorithms are designed for solving specific problems. Brain images are one of the most used data types. AI has the possibility to improve clinicians' decision-making ability in neuroscience applications. However, major issues still need to be addressed for a better practical use of AI in the brain. To this aim, it is important to both gather comprehensive data and build explainable AI algorithms.
RESUMO
Accurate diagnoses of specific diseases require, in general, the review of the whole medical history of a patient. Currently, even though many advances have been made for disease monitoring, domain experts are still requested to perform direct analyses in order to get a precise classification, thus implying significant efforts and costs. In this work we present a framework for automated diagnosis based on high-dimensional gene expression and clinical data. Given that high-dimensional data can be difficult to analyze and computationally expensive to process, we first perform data reduction to transform high-dimensional representations of data into a lower dimensional space, yet keeping them meaningful for our purposes. We used then different data visualization techniques to embed complex pieces of information in 2-D images, that are in turn used to perform diagnosis relying on deep learning approaches. Experimental analyses show that the proposed method achieves good performance, featuring a prediction Recall value between 91% and 99%.
Assuntos
Visualização de Dados , Redes Neurais de Computação , Expressão Gênica , HumanosRESUMO
Recent advances in image acquisition and processing techniques, along with the success of novel deep learning architectures, have given the opportunity to develop innovative algorithms capable to provide a better characterization of neurological related diseases. In this work, we introduce a neural network based approach to classify Multiple Sclerosis (MS) patients into four clinical profiles. Starting from their structural connectivity information, obtained by diffusion tensor imaging and represented as a graph, we evaluate the classification performances using unweighted and weighted connectivity matrices. Furthermore, we investigate the role of graph-based features for a better characterization and classification of the pathology. Ninety MS patients (12 clinically isolated syndrome, 30 relapsing-remitting, 28 secondary-progressive, and 20 primary-progressive) along with 24 healthy controls, were considered in this study. This work shows the great performances achieved by neural networks methods in the classification of the clinical profiles. Furthermore, it shows local graph metrics do not improve the classification results suggesting that the latent features created by the neural network in its layers have a much important informative content. Finally, we observe that graph weights representation of brain connections preserve important information to discriminate between clinical forms.
RESUMO
Prediction of disability progression in multiple sclerosis patients is a critical component of their management. In particular, one challenge is to identify and characterize a patient profile who may benefit of efficient treatments. However, it is not yet clear whether a particular relation exists between the brain structure and the disability status.This work aims at producing a fully automatic model for the expanded disability status score estimation, given the brain structural connectivity representation of a multiple sclerosis patient. The task is addressed by first extracting the connectivity graph, obtained by combining brain grey matter parcellation and tractography extracted from Diffusion and T1-weighted Magnetic Resonance (MR) images, and then processing it via a convolutional neural network (CNN) in order to compute the predicted score. Experiments show that the herein proposed approach achieves promising results, thus resulting as an important step forward on the road to better predict the evolution of the disease.
Assuntos
Avaliação da Deficiência , Esclerose Múltipla/fisiopatologia , Redes Neurais de Computação , Encéfalo/diagnóstico por imagem , Encéfalo/patologia , Substância Cinzenta/diagnóstico por imagem , Substância Cinzenta/patologia , Humanos , Processamento de Imagem Assistida por Computador , Imageamento por Ressonância Magnética , Esclerose Múltipla/diagnóstico por imagemRESUMO
Morphological analysis of the retinal vessels by fundoscopy provides noninvasive means for detecting and staging systemic microvascular damage. However, full exploitation of fundoscopy in clinical settings is limited by paucity of quantitative, objective information obtainable through the observer-driven evaluations currently employed in routine practice. Here, we report on the development of a semiautomated, computer-based method to assess retinal vessel morphology. The method allows simultaneous and operator-independent quantitative assessment of arteriole-to-venule ratio, tortuosity index, and mean fractal dimension. The method was implemented in two conditions known for being associated with retinal vessel changes: hypertensive retinopathy and Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL). The results showed that our approach is effective in detecting and quantifying the retinal vessel abnormalities. Arteriole-to-venule ratio, tortuosity index, and mean fractal dimension were altered in the subjects with hypertensive retinopathy or CADASIL with respect to age- and gender-matched controls. The interrater reliability was excellent for all the three indices (intraclass correlation coefficient ≥ 85%). The method represents simple and highly reproducible means for discriminating pathological conditions characterized by morphological changes of retinal vessels. The advantages of our method include simultaneous and operator-independent assessment of different parameters and improved reliability of the measurements.