Search

Contains fulltext : 305368.pdf (Publisher’s version ) (Closed access), Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. In biomedical image analysis, chosen performance metrics often do not reflect the domain interest, and thus fail to adequately measure scientific progress and hinder translation of ML techniques into practice. To overcome this, we created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Developed by a large international consortium in a multistage Delphi process, it is based on the novel concept of a problem fingerprint-a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), dataset and algorithm output. On the basis of the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as classification tasks at image, object or pixel level, namely image-level classification, object detection, semantic segmentation and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. Its applicability is demonstrated for various biomedical use cases.

Contains fulltext : 305370.pdf (Publisher’s version ) (Closed access), Validation metrics are key for tracking scientific progress and bridging the current chasm between artificial intelligence research and its translation into practice. However, increasing evidence shows that, particularly in image analysis, metrics are often chosen inadequately. Although taking into account the individual strengths, weaknesses and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multistage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides a reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. Although focused on biomedical image analysis, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. The work serves to enhance global comprehension of a key topic in image analysis validation.

Item does not contain fulltext, Amidst the ongoing pandemic, the assessment of computed tomography (CT) images for COVID-19 presence can exceed the workload capacity of radiologists. Several studies addressed this issue by automating COVID-19 classification and grading from CT images with convolutional neural networks (CNNs). Many of these studies reported initial results of algorithms that were assembled from commonly used components. However, the choice of the components of these algorithms was often pragmatic rather than systematic and systems were not compared to each other across papers in a fair manner. We systematically investigated the effectiveness of using 3-D CNNs instead of 2-D CNNs for seven commonly used architectures, including DenseNet, Inception, and ResNet variants. For the architecture that performed best, we furthermore investigated the effect of initializing the network with pretrained weights, providing automatically computed lesion maps as additional network input, and predicting a continuous instead of a categorical output. A 3-D DenseNet-201 with these components achieved an area under the receiver operating characteristic curve of 0.930 on our test set of 105 CT scans and an AUC of 0.919 on a publicly available set of 742 CT scans, a substantial improvement in comparison with a previously published 2-D CNN. This article provides insights into the performance benefits of various components for COVID-19 classification and grading systems. We have created a challenge on grand-challenge.org to allow for a fair comparison between the results of this and future research.

Contains fulltext : 282663.pdf (Publisher’s version ) (Open Access), An increasing number of artificial intelligence (AI) systems are being proposed in ophthalmology, motivated by the variety and amount of clinical and imaging data, as well as their potential benefits at the different stages of patient care. Despite achieving close or even superior performance to that of experts, there is a critical gap between development and integration of AI systems in ophthalmic practice. This work focuses on the importance of trustworthy AI to close that gap. We identify the main aspects or challenges that need to be considered along the AI design pipeline so as to generate systems that meet the requirements to be deemed trustworthy, including those concerning accuracy, resiliency, reliability, safety, and accountability. We elaborate on mechanisms and considerations to address those aspects or challenges, and define the roles and responsibilities of the different stakeholders involved in AI for ophthalmic care, i.e., AI developers, reading centers, healthcare providers, healthcare institutions, ophthalmological societies and working groups or committees, patients, regulatory bodies, and payers. Generating trustworthy AI is not a responsibility of a sole stakeholder. There is an impending necessity for a collaborative approach where the different stakeholders are represented along the AI design pipeline, from the definition of the intended use to post-market surveillance after regulatory approval. This work contributes to establish such multi-stakeholder interaction and the main action points to be taken so that the potential benefits of AI reach real-world ophthalmic settings.

Item does not contain fulltext, IMPORTANCE: Treatments for geographic atrophy (GA), a late stage of age-related macular degeneration (AMD), are currently under development. Understanding the natural course is needed for optimal trial design. Although enlargement rates of GA and visual acuity (VA) in the short term are known from clinical studies, knowledge of enlargement in the long term, life expectancy, and visual course is lacking. OBJECTIVE: To determine long-term enlargement of GA. DESIGN, SETTING, AND PARTICIPANTS: In this study, participant data were collected from 4 population-based cohort studies, with up to 25 years of follow-up and eye examinations at 5-year intervals: the Rotterdam Study cohorts 1, 2, and 3 and the Blue Mountains Eye Study. Data were collected from 1990 to 2015, and data were analyzed from January 2019 to November 2020. MAIN OUTCOMES AND MEASURES: Area of GA was measured pixel by pixel using all available imaging. Area enlargement and enlargement of the square root-transformed area, time until GA reached the central fovea, and time until death were assessed, and best-corrected VA, smoking status, macular lesions according to the Three Continent AMD Consortium classification, a modified version of the Wisconsin age-related maculopathy grading system, and AMD genetic variants were covariates in Spearman, Pearson, or Mann-Whitney analyses. RESULTS: Of 171 included patients, 106 (62.0%) were female, and the mean (SD) age at inclusion was 82.6 (7.1) years. A total of 147 of 242 eyes with GA (60.7%) were newly diagnosed in our study. The mean area of GA at first presentation was 3.74 mm2 (95% CI, 3.11-4.67). Enlargement rate varied widely between persons (0.02 to 4.05 mm2 per year), with a mean of 1.09 mm2 per year (95% CI, 0.89-1.30). Stage of AMD in the other eye was correlated with GA enlargement (Spearman ρ = 0.34; P = .01). Foveal involvement was already present in incident GA in 55 of 147 eyes (37.4%); 23 of 42 eyes (55%) developed this after a mean (range) period of 5

Contains fulltext : 235480.pdf (Publisher’s version ) (Open Access), PURPOSE: We sought to develop and validate a deep learning model for segmentation of 13 features associated with neovascular and atrophic age-related macular degeneration (AMD). DESIGN: Development and validation of a deep-learning model for feature segmentation. METHODS: Data for model development were obtained from 307 optical coherence tomography volumes. Eight experienced graders manually delineated all abnormalities in 2712 B-scans. A deep neural network was trained with these data to perform voxel-level segmentation of the 13 most common abnormalities (features). For evaluation, 112 B-scans from 112 patients with a diagnosis of neovascular AMD were annotated by 4 independent observers. The main outcome measures were Dice score, intraclass correlation coefficient, and free-response receiver operating characteristic curve. RESULTS: On 11 of 13 features, the model obtained a mean Dice score of 0.63 +/- 0.15, compared with 0.61 +/- 0.17 for the observers. The mean intraclass correlation coefficient for the model was 0.66 +/- 0.22, compared with 0.62 +/- 0.21 for the observers. Two features were not evaluated quantitatively because of a lack of data. Free-response receiver operating characteristic analysis demonstrated that the model scored similar or higher sensitivity per false positives compared with the observers. CONCLUSIONS: The quality of the automatic segmentation matches that of experienced graders for most features, exceeding human performance for some features. The quantified parameters provided by the model can be used in the current clinical routine and open possibilities for further research into treatment response outside clinical trials.

Contains fulltext : 238599.pdf (Publisher’s version ) (Open Access), Adversarial attacks are considered a potentially serious security threat for machine learning systems. Medical image analysis (MedIA) systems have recently been argued to be vulnerable to adversarial attacks due to strong financial incentives and the associated technological infrastructure. In this paper, we study previously unexplored factors affecting adversarial attack vulnerability of deep learning MedIA systems in three medical domains: ophthalmology, radiology, and pathology. We focus on adversarial black-box settings, in which the attacker does not have full access to the target model and usually uses another model, commonly referred to as surrogate model, to craft adversarial examples that are then transferred to the target model. We consider this to be the most realistic scenario for MedIA systems. Firstly, we study the effect of weight initialization (pre-training on ImageNet or random initialization) on the transferability of adversarial attacks from the surrogate model to the target model, i.e., how effective attacks crafted using the surrogate model are on the target model. Secondly, we study the influence of differences in development (training and validation) data between target and surrogate models. We further study the interaction of weight initialization and data differences with differences in model architecture. All experiments were done with a perturbation degree tuned to ensure maximal transferability at minimal visual perceptibility of the attacks. Our experiments show that pre-training may dramatically increase the transferability of adversarial examples, even when the target and surrogate's architectures are different: the larger the performance gain using pre-training, the larger the transferability. Differences in the development data between target and surrogate models considerably decrease the performance of the attack; this decrease is further amplified by difference in the model architecture. We believe these factors should be consider

Contains fulltext : 228667.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 232933.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 221442.pdf (publisher's version ) (Open Access)

Contains fulltext : 221444.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 222173.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 220771.pdf (Publisher’s version ) (Open Access), PURPOSE: To validate the performance of a commercially available, CE-certified deep learning (DL) system, RetCAD v.1.3.0 (Thirona, Nijmegen, The Netherlands), for the joint automatic detection of diabetic retinopathy (DR) and age-related macular degeneration (AMD) in colour fundus (CF) images on a dataset with mixed presence of eye diseases. METHODS: Evaluation of joint detection of referable DR and AMD was performed on a DR-AMD dataset with 600 images acquired during routine clinical practice, containing referable and non-referable cases of both diseases. Each image was graded for DR and AMD by an experienced ophthalmologist to establish the reference standard (RS), and by four independent observers for comparison with human performance. Validation was furtherly assessed on Messidor (1200 images) for individual identification of referable DR, and the Age-Related Eye Disease Study (AREDS) dataset (133 821 images) for referable AMD, against the corresponding RS. RESULTS: Regarding joint validation on the DR-AMD dataset, the system achieved an area under the ROC curve (AUC) of 95.1% for detection of referable DR (SE = 90.1%, SP = 90.6%). For referable AMD, the AUC was 94.9% (SE = 91.8%, SP = 87.5%). Average human performance for DR was SE = 61.5% and SP = 97.8%; for AMD, SE = 76.5% and SP = 96.1%. Regarding detection of referable DR in Messidor, AUC was 97.5% (SE = 92.0%, SP = 92.1%); for referable AMD in AREDS, AUC was 92.7% (SE = 85.8%, SP = 86.0%). CONCLUSION: The validated system performs comparably to human experts at simultaneous detection of DR and AMD. This shows that DL systems can facilitate access to joint screening of eye diseases and become a quick and reliable support for ophthalmological experts.

Contains fulltext : 220801.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 203359.pdf (publisher's version ) (Open Access), PURPOSE: Currently, no outcome measures are clinically validated and accepted as clinical endpoints by regulatory agencies for drug development in intermediate age-related macular degeneration (iAMD). The MACUSTAR Consortium, a public-private research group funded by the European Innovative Medicines Initiative intends to close this gap. PROCEDURES: Development of study protocol and statistical analysis plan including predictive modelling of multimodal endpoints based on a review of the literature and expert consensus. RESULTS: This observational study consists of a cross-sectional and a longitudinal part. Functional outcome measures assessed under low contrast and low luminance have the potential to detect progression of visual deficit within iAMD and to late AMD. Structural outcome measures will be multimodal and investigate topographical relationships with function. Current patient-reported outcome measures (PROMs) are not acceptable to regulators and may not capture the functional deficit specific to iAMD with needed precision, justifying development of novel PROMs for iAMD. The total sample size will be n = 750, consisting mainly of subjects with iAMD (n = 600). CONCLUSIONS: As clinical endpoints currently accepted by regulators cannot detect functional loss or patient-relevant impact in iAMD, we will clinically validate novel candidate endpoints for iAMD.

Contains fulltext : 204002.pdf (publisher's version ) (Open Access)

Contains fulltext : 207373.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 212599.pdf (publisher's version ) (Closed access)

Contains fulltext : 203617.pdf (publisher's version ) (Closed access)

Contains fulltext : 201368.pdf (publisher's version ) (Open Access)

Contains fulltext : 191291.pdf (publisher's version ) (Open Access), We developed a deep learning algorithm for the automatic segmentation and quantification of intraretinal cystoid fluid (IRC) in spectral domain optical coherence tomography (SD-OCT) volumes independent of the device used for acquisition. A cascade of neural networks was introduced to include prior information on the retinal anatomy, boosting performance significantly. The proposed algorithm approached human performance reaching an overall Dice coefficient of 0.754 ± 0.136 and an intraclass correlation coefficient of 0.936, for the task of IRC segmentation and quantification, respectively. The proposed method allows for fast quantitative IRC volume measurements that can be used to improve patient care, reduce costs, and allow fast and reliable analysis in large population studies.

Contains fulltext : 179538.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 174703.pdf (Publisher’s version ) (Open Access), Purpose: To evaluate a machine learning algorithm that automatically grades age-related macular degeneration (AMD) severity stages from optical coherence tomography (OCT) scans. Methods: A total of 3265 OCT scans from 1016 patients with either no signs of AMD or with signs of early, intermediate, or advanced AMD were randomly selected from a large European multicenter database. A machine learning system was developed to automatically grade unseen OCT scans into different AMD severity stages without requiring retinal layer segmentation. The ability of the system to identify high-risk AMD stages and to assign the correct severity stage was determined by using receiver operator characteristic (ROC) analysis and Cohen's kappa statistics (kappa), respectively. The results were compared to those of two human observers. Reproducibility was assessed in an independent, publicly available data set of 384 OCT scans. Results: The system achieved an area under the ROC curve of 0.980 with a sensitivity of 98.2% at a specificity of 91.2%. This compares favorably with the performance of human observers who achieved sensitivities of 97.0% and 99.4% at specificities of 89.7% and 87.2%, respectively. A good level of agreement with the reference was obtained (kappa = 0.713) and was in concordance with the human observers (kappa = 0.775 and kappa = 0.755, respectively). Conclusions: A machine learning system capable of automatically grading OCT scans into AMD severity stages was developed and showed similar performance as human observers. The proposed automatic system allows for a quick and reliable grading of large quantities of OCT scans, which could increase the efficiency of large-scale AMD studies and pave the way for AMD screening using OCT.

Contains fulltext : 177930.pdf (publisher's version ) (Open Access), We developed a fully automated system using a convolutional neural network (CNN) for total retina segmentation in optical coherence tomography (OCT) that is robust to the presence of severe retinal pathology. A generalized U-net network architecture was introduced to include the large context needed to account for large retinal changes. The proposed algorithm outperformed qualitative and quantitatively two available algorithms. The algorithm accurately estimated macular thickness with an error of 14.0 +/- 22.1 microm, substantially lower than the error obtained using the other algorithms (42.9 +/- 116.0 microm and 27.1 +/- 69.3 microm, respectively). These results highlighted the proposed algorithm's capability of modeling the wide variability in retinal appearance and obtained a robust and reliable retina segmentation even in severe pathological cases.

Contains fulltext : 181865.pdf (publisher's version ) (Open Access), We propose a method for automatic detection of the foveal center in optical coherence tomography (OCT). The method is based on a pixel-wise classification of all pixels in an OCT volume using a fully convolutional neural network (CNN) with dilated convolution filters. The CNN-architecture contains anisotropic dilated filters and a shortcut connection and has been trained using a dynamic training procedure where the network identifies its own relevant training samples. The performance of the proposed method is evaluated on a data set of 400 OCT scans of patients affected by age-related macular degeneration (AMD) at different severity levels. For 391 scans (97.75%) the method identified the foveal center with a distance to a human reference less than 750 mum, with a mean (+/- SD) distance of 71 mum +/- 107 mum. Two independent observers also annotated the foveal center, with a mean distance to the reference of 57 mum +/- 84 mum and 56 mum +/- 80 mum, respectively. Furthermore, we evaluate variations to the proposed network architecture and training procedure, providing insight in the characteristics that led to the demonstrated performance of the proposed method.

Contains fulltext : 181640.pdf (publisher's version ) (Open Access)

Contains fulltext : 173146.pdf (publisher's version ) (Open Access), Modern Computed Tomography (CT) scanners are capable of acquiring contrast dynamics of the whole brain, adding functional to anatomical information. Soft tissue segmentation is important for subsequent applications such as tissue dependent perfusion analysis and automated detection and quantification of cerebral pathology. In this work a method is presented to automatically segment white matter (WM) and gray matter (GM) in contrast- enhanced 4D CT images of the brain. The method starts with intracranial segmentation via atlas registration, followed by a refinement using a geodesic active contour with dominating advection term steered by image gradient information, from a 3D temporal average image optimally weighted according to the exposures of the individual time points of the 4D CT acquisition. Next, three groups of voxel features are extracted: intensity, contextual, and temporal. These are used to segment WM and GM with a support vector machine. Performance was assessed using cross validation in a leave-one-patient-out manner on 22 patients. Dice coefficients were 0.81 +/- 0.04 and 0.79 +/- 0.05, 95% Hausdorff distances were 3.86 +/- 1.43 and 3.07 +/- 1.72 mm, for WM and GM, respectively. Thus, WM and GM segmentation is feasible in 4D CT with good accuracy.

Contains fulltext : 173029.pdf (Publisher’s version ) (Open Access), Recent advances in machine learning yielded new techniques to train deep neural networks, which resulted in highly successful applications in many pattern recognition tasks such as object detection and speech recognition. In this paper we provide a head-to-head comparison between a state-of-the art in mammography CAD system, relying on a manually designed feature set and a Convolutional Neural Network (CNN), aiming for a system that can ultimately read mammograms independently. Both systems are trained on a large data set of around 45,000 images and results show the CNN outperforms the traditional CAD system at low sensitivity and performs comparable at high sensitivity. We subsequently investigate to what extent features such as location and patient information and commonly used manual features can still complement the network and see improvements at high specificity over the CNN especially with location and context features, which contain information not available to the CNN. Additionally, a reader study was performed, where the network was compared to certified screening radiologists on a patch level and we found no significant difference between the network and the readers.

Contains fulltext : 175031.pdf (Publisher’s version ) (Open Access), PURPOSE: Symmetry is an important feature of human anatomy and the absence of symmetry in medical images can indicate the presence of pathology. Quantification of image symmetry can then be used to improve the automatic analysis of medical images. METHODS: A method is presented that computes both local and global symmetry in 2D medical images. A symmetry axis is determined to define for each position p in the image a mirrored position p' on the contralateral side of the axis. In the neighborhood of p', an optimally corresponding position ps is determined by minimizing a cost function d that combines intensity differences in a patch around p and the mirrored patch around ps and the spatial distance between p' and ps. The optimal value of d is used as a measure of local symmetry s. The average of all values of s, indicated as S, quantifies global symmetry. Starting from an initial approximation of the symmetry axis, the optimal orientation and position of the axis is determined by greedy minimization of S. RESULTS: The method was evaluated in three experiments concerning abnormality detection in frontal chest radiographs. In the first experiment, global symmetry S was used to discriminate between 174 normal images and 174 images containing diffuse textural abnormalities from the publicly available CRASS database of tuberculosis suspects. Performance, measured as area under the receiver operating characteristic curve Az was 0.838. The second experiment investigated whether adding the local symmetry s as an additional feature to a set of 106 texture features resulted in improvements in classifying local patches in the same image database. We found that Az increased from 0.878 to 0.891 (P = 0.001). In the third experiment, it was shown that the contrast of pulmonary nodules, obtained from the publicly available JSRT database, increased significantly in the local symmetry map compared to the original image. CONCLUSIONS: We conclude that the proposed algorithm for symmetry

Contains fulltext : 168084.pdf (publisher's version ) (Open Access), Lack of human resources and radiological interpretation expertise impair tuberculosis (TB) screening programmes in TB-endemic countries. Computer-aided detection (CAD) constitutes a viable alternative for chest radiograph (CXR) reading. However, no automated techniques that exploit the additional clinical information typically available during screening exist. To address this issue and optimally exploit this information, a machine learning-based combination framework is introduced. We have evaluated this framework on a database containing 392 patient records from suspected TB subjects prospectively recruited in Cape Town, South Africa. Each record comprised a CAD score, automatically computed from a CXR, and 12 clinical features. Comparisons with strategies relying on either CAD scores or clinical information alone were performed. Our results indicate that the combination framework outperforms the individual strategies in terms of the area under the receiving operating characteristic curve (0.84 versus 0.78 and 0.72), specificity at 95% sensitivity (49% versus 24% and 31%) and negative predictive value (98% versus 95% and 96%). Thus, it is believed that combining CAD and clinical information to estimate the risk of active disease is a promising tool for TB screening.

Contains fulltext : 167707.pdf (publisher's version ) (Open Access), Pathologists face a substantial increase in workload and complexity of histopathologic cancer diagnosis due to the advent of personalized medicine. Therefore, diagnostic protocols have to focus equally on efficiency and accuracy. In this paper we introduce 'deep learning' as a technique to improve the objectivity and efficiency of histopathologic slide analysis. Through two examples, prostate cancer identification in biopsy specimens and breast cancer metastasis detection in sentinel lymph nodes, we show the potential of this new methodology to reduce the workload for pathologists, while at the same time increasing objectivity of diagnoses. We found that all slides containing prostate cancer and micro- and macro-metastases of breast cancer could be identified automatically while 30-40% of the slides containing benign and normal tissue could be excluded without the use of any additional immunohistochemical markers or human intervention. We conclude that 'deep learning' holds great promise to improve the efficacy of prostate cancer diagnosis and breast cancer staging.

Contains fulltext : 168003.pdf (Publisher’s version ) (Open Access), The major advantage of multiple-instance learning (MIL) applied to a computer-aided detection (CAD) system is that it allows optimizing the latter with case-level labels instead of accurate lesion outlines as traditionally required for a supervised approach. As shown in previous work, a MIL-based CAD system can perform comparably to its supervised counterpart considering complex tasks such as chest radiograph scoring in tuberculosis (TB) detection. However, despite this remarkable achievement, the uncertainty inherent to MIL can lead to a less satisfactory outcome if analysis at lower levels (e.g., regions or pixels) is needed. This issue may seriously compromise the applicability of MIL to tasks related to quantification or grading, or detection of highly localized lesions. In this paper, we propose to reduce uncertainty by embedding a MIL classifier within an active learning (AL) framework. To minimize the labeling effort, we develop a novel instance selection mechanism that exploits the MIL problem definition through one-class classification. We adapt this mechanism to provide meaningful regions instead of individual instances for expert labeling, which is a more appropriate strategy given the application domain. In addition, and contrary to usual AL methods, a single iteration is performed. To show the effectiveness of our approach, we compare the output of a MIL-based CAD system trained with and without the proposed AL framework. The task is to detect textural abnormalities related to TB. Both quantitative and qualitative evaluations at the pixel level are carried out. In the former case, receiver operating characteristic and precision-recall analyses are used. For completeness, we also compare with a CAD system following a supervised approach and related baselines. The quantitative results show that the proposed method significantly improves the MIL-based classification and narrows the gap with the supervised approach in \%50 and \%75, depending on if 100 or 3

Contains fulltext : 167823.pdf (Publisher’s version ) (Open Access), PURPOSE: Age-related macular degeneration is a common form of vision loss affecting older adults. The etiology of AMD is multifactorial and is influenced by environmental and genetic risk factors. In this study, we examine how 19 common risk variants contribute to drusen progression, a hallmark of AMD pathogenesis. METHODS: Exome chip data was made available through the International AMD Genomics Consortium (IAMDGC). Drusen quantification was carried out with color fundus photographs using an automated drusen detection and quantification algorithm. A genetic risk score (GRS) was calculated per subject by summing risk allele counts at 19 common genetic risk variants weighted by their respective effect sizes. Pathway analysis of drusen progression was carried out with the software package Pathway Analysis by Randomization Incorporating Structure. RESULTS: We observed significant correlation with drusen baseline area and the GRS in the age-related eye disease study (AREDS) dataset (rho = 0.175, P = 0.006). Measures of association were not statistically significant between drusen progression and the GRS (P = 0.54). Pathway analysis revealed the cell adhesion molecules pathway as the most highly significant pathway associated with drusen progression (corrected P = 0.02). CONCLUSIONS: In this study, we explored the potential influence of known common AMD genetic risk factors on drusen progression. Our results from the GRS analysis showed association of increasing genetic burden (from 19 AMD associated loci) to baseline drusen load but not drusen progression in the AREDS dataset while pathway analysis suggests additional genetic contributors to AMD risk.

Item does not contain fulltext, Automated detection of Tuberculosis (TB) using chest radiographs (CXRs) is gaining popularity due to the lack of trained human readers in resource limited countries with a high TB burden. The majority of the computer-aided detection (CAD) systems for TB focus on detection of parenchymal abnormalities and ignore other important manifestations such as pleural effusion (PE). The costophrenic angle is a commonly used measure for detecting PE, but has limitations. In this work, an automatic method to detect PE in the left and right hemithoraces is proposed and evaluated on a database of 638 CXRs. We introduce a robust way to localize the costophrenic region using the chest wall contour as a landmark structure, in addition to the lung segmentation. Region descriptors are proposed based on intensity and morphology information in the region around the costophrenic recess. Random forest classifiers are trained to classify left and right hemithoraces. Performance of the PE detection system is evaluated in terms of recess localization accuracy and area under the receiver operating characteristic curve (AUC). The proposed method shows significant improvement in the AUC values as compared to systems which use lung segmentation and the costophrenic angle measurement alone.

Contains fulltext : 167797.pdf (Publisher’s version ) (Open Access), Convolutional neural networks (CNNs) are deep learning network architectures that have pushed forward the state-of-the-art in a range of computer vision applications and are increasingly popular in medical image analysis. However, training of CNNs is time-consuming and challenging. In medical image analysis tasks, the majority of training examples are easy to classify and therefore contribute little to the CNN learning process. In this paper, we propose a method to improve and speed-up the CNN training for medical image analysis tasks by dynamically selecting misclassified negative samples during training. Training samples are heuristically sampled based on classification by the current status of the CNN. Weights are assigned to the training samples and informative samples are more likely to be included in the next CNN training iteration. We evaluated and compared our proposed method by training a CNN with (SeS) and without (NSeS) the selective sampling method. We focus on the detection of hemorrhages in color fundus images. A decreased training time from 170 epochs to 60 epochs with an increased performance-on par with two human experts-was achieved with areas under the receiver operating characteristics curve of 0.894 and 0.972 on two data sets. The SeS CNN statistically outperformed the NSeS CNN on an independent test set.

Contains fulltext : 168320.pdf (publisher's version ) (Open Access), We developed an automatic system to identify and differentiate color fundus images containing no lesions, drusen or exudates. Drusen and exudates are lesions with a bright appearance, associated with age-related macular degeneration and diabetic retinopathy, respectively. The system consists of three lesion detectors operating at pixel-level, combining their outputs using spatial pooling and classification with a random forest classifier. System performance was compared with ratings of two independent human observers using human-expert annotations as reference. Kappa agreements of 0.89, 0.97 and 0.92 and accuracies of 0.93, 0.98 and 0.95 were obtained for the system and observers, respectively.

Contains fulltext : 154605.pdf (publisher's version ) (Open Access)

Contains fulltext : 154701.pdf (Publisher’s version ) (Open Access), In order to reach performance levels comparable to those of human experts, computer-aided detection (CAD) systems are typically optimized by means of a supervised learning approach that relies on large training databases comprising manually annotated lesions. However, manually outlining those lesions constitutes a difficult and time-consuming process that renders detailedly annotated data often difficult to obtain. In this paper, we investigate an alternative pattern classification approach, namely multiple-instance learning (MIL), that does not require such detailed information for a CAD system to be optimized. We have applied MIL to a CAD system aimed at detecting textural lesions associated with tuberculosis. Only the case (or image) condition (normal or abnormal), which was determined by radiological means, was required during training. Based upon the well-known miSVM technique, we propose a novel algorithm, specifically designed for our CAD application, that overcomes serious drawbacks of the former related to underestimation of the positive instances and costly iteration. The key of the proposed method is to use probability estimates instead of decision values to guide the MIL procedure. In addition, we include countermeasures that deal with the uncertainty resulting from instance relabeling. To show the advantages of our MIL-based approach as compared with a traditional supervised one, experiments with three different image databases were conducted. The area under the receiver operating characteristic curve was utilized as a performance measure. With the first database, for which training lesion annotations were available, the supervised system was not much better than our MILbased method (0:88 vs. 0:86). Thus, the proposed approach achieved highly competitive results without resorting to lesionlevel information and the associated annotation process. When evaluating the remaining databases, given their large difference with respect to the previous image set

Contains fulltext : 154610.pdf (Publisher’s version ) (Open Access), PURPOSE: To examine human performance and agreement on reticular pseudodrusen (RPD) detection and quantification by using single- and multimodality grading protocols and to describe and evaluate a machine learning system for the automatic detection and quantification of reticular pseudodrusen by using single- and multimodality information. METHODS: Color fundus, fundus autofluoresence, and near-infrared images of 278 eyes from 230 patients with or without presence of RPD were used in this study. All eyes were scored for presence of RPD during single- and multimodality setups by two experienced observers and a developed machine learning system. Furthermore, automatic quantification of RPD area was performed by the proposed system and compared with human delineations. RESULTS: Observers obtained a higher performance and better interobserver agreement for RPD detection with multimodality grading, achieving areas under the receiver operating characteristic (ROC) curve of 0.940 and 0.958, and a kappa agreement of 0.911. The proposed automatic system achieved an area under the ROC of 0.941 with a multimodality setup. Automatic RPD quantification resulted in an intraclass correlation (ICC) value of 0.704, which was comparable with ICC values obtained between single-modality manual delineations. CONCLUSIONS: Observer performance and agreement for RPD identification improved significantly by using a multimodality grading approach. The developed automatic system showed similar performance as observers, and automatic RPD area quantification was in concordance with manual delineations. The proposed automatic system allows for a fast and accurate identification and quantification of RPD, opening the way for efficient quantitative imaging biomarkers in large data set analysis.

Contains fulltext : 137637.pdf (publisher's version ) (Open Access), PURPOSE: We describe the differences and similarities in clinical characteristics and phenotype of familial and sporadic patients with age-related macular degeneration (AMD). METHODS: We evaluated data of 1828 AMD patients and 1715 controls enrolled in the European Genetic Database. All subjects underwent ophthalmologic examination, including visual acuity testing and fundus photography. Images were graded and fundus photographs were used for automatic drusen quantification by a machine learning algorithm. Data on disease characteristics, family history, medical history, and lifestyle habits were obtained by a questionnaire. RESULTS: The age at first symptoms was significantly lower in AMD patients with a positive family history (68.5 years) than in those with no family history (71.6 years, P = 1.9 x 10(-5)). Risk factors identified in sporadic and familial subjects were increasing age (odds ratio [OR], 1.08 per year; P = 3.0 x 10(-51), and OR, 1.15; P = 5.3 x 10(-36), respectively) and smoking (OR, 1.01 per pack year; P = 1.1 x 10(-6) and OR, 1.02; P = 0.005). Physical activity and daily red meat consumption were significantly associated with AMD in sporadic subjects only (OR, 0.49; P = 3.7 x 10(-10) and OR, 1.81; P = 0.001). With regard to the phenotype, geographic atrophy and cuticular drusen were significantly more prevalent in familial AMD (17.5% and 21.7%, respectively) compared to sporadic AMD (9.8% and 12.1%). CONCLUSIONS: Familial AMD patients become symptomatic at a younger age. The higher prevalence of geographic atrophy and cuticular drusen in the familial AMD cases may be explained by the contribution of additional genetic factors segregating within families.

Contains fulltext : 118034.pdf (Publisher’s version ) (Open Access), PURPOSE: To evaluate a machine learning algorithm that allows for computer aided diagnosis (CAD) of non-advanced age-related macular degeneration (AMD) by providing an accurate detection and quantification of drusen location, area and size. METHODS: Color fundus photographs of 407 eyes without AMD or with early to moderate AMD were randomly selected from a large European multicenter database. A machine learning system was developed to automatically detect and quantify drusen on each image. Based on detected drusen, the CAD software provided a risk assessment to develop advanced AMD. Evaluation of the CAD system was performed using annotations made by two blinded human graders. RESULTS: Free-response Receiver Operating Characteristics (FROC) analysis showed that the proposed system approaches the performance of human observers in detecting drusen. The estimated drusen area showed excellent agreement with both observers, with mean intra-class correlation coefficients (ICC) larger than 0.85. Maximum druse diameter agreement was lower with a maximum ICC of 0.69 but comparable to the interobserver agreement (ICC=0.79). For automatic AMD risk assessment, the system achieved areas under the Receiver Operating Characteristic (ROC) curve of 0.948 and 0.954, reaching similar performance as human observers. CONCLUSIONS: A machine learning system, capable of separating high risk from low risk patients with non-advanced AMD by providing accurate detection and quantification of drusen, was developed. The proposed method allows for quick and reliable diagnosis of AMD, opening the way for large dataset analysis within population studies and genotype-phenotype correlation analysis.