Eike Petersen

Eike Petersen, Aasa Feragen, Maria Luise da Costa Zemsch et al.

Convolutional neural networks have enabled significant improvements in medical image-based diagnosis. It is, however, increasingly clear that these models are susceptible to performance degradation when facing spurious correlations and dataset shift, leading, e.g., to underperformance on underrepresented patient groups. In this paper, we compare two classification schemes on the ADNI MRI dataset: a simple logistic regression model using manually selected volumetric features, and a convolutional neural network trained on 3D MRI data. We assess the robustness of the trained models in the face of varying dataset splits, training set sex composition, and stage of disease. In contrast to earlier work in other imaging modalities, we do not observe a clear pattern of improved model performance for the majority group in the training dataset. Instead, while logistic regression is fully robust to dataset composition, we find that CNN performance is generally improved for both male and female subjects when including more female subjects in the training dataset. We hypothesize that this might be due to inherent differences in the pathology of the two sexes. Moreover, in our analysis, the logistic regression model outperforms the 3D CNN, emphasizing the utility of manual feature specification based on prior knowledge, and the need for more robust automatic feature selection.

Eike Petersen, Melanie Ganz, Sune Hannibal Holm et al.

Recent work on algorithmic fairness has largely focused on the fairness of discrete decisions, or classifications. While such decisions are often based on risk score models, the fairness of the risk models themselves has received considerably less attention. Risk models are of interest for a number of reasons, including the fact that they communicate uncertainty about the potential outcomes to users, thus representing a way to enable meaningful human oversight. Here, we address fairness desiderata for risk score models. We identify the provision of similar epistemic value to different groups as a key desideratum for risk score fairness. Further, we address how to assess the fairness of risk score models quantitatively, including a discussion of metric choices and meaningful statistical comparisons between groups. In this context, we also introduce a novel calibration error metric that is less sample size-biased than previously proposed metrics, enabling meaningful comparisons between groups of different sizes. We illustrate our methodology - which is widely applicable in many other settings - in two case studies, one in recidivism risk prediction, and one in risk of major depressive disorder (MDD) prediction.

Nina Weng, Siavash Bigdeli, Eike Petersen et al.

While many studies have assessed the fairness of AI algorithms in the medical field, the causes of differences in prediction performance are often unknown. This lack of knowledge about the causes of bias hampers the efficacy of bias mitigation, as evidenced by the fact that simple dataset balancing still often performs best in reducing performance gaps but is unable to resolve all performance differences. In this work, we investigate the causes of gender bias in machine learning-based chest X-ray diagnosis. In particular, we explore the hypothesis that breast tissue leads to underexposure of the lungs and causes lower model performance. Methodologically, we propose a new sampling method which addresses the highly skewed distribution of recordings per patient in two widely used public datasets, while at the same time reducing the impact of label errors. Our comprehensive analysis of gender differences across diseases, datasets, and gender representations in the training set shows that dataset imbalance is not the sole cause of performance differences. Moreover, relative group performance differs strongly between datasets, indicating important dataset-specific factors influencing male/female group performance. Finally, we investigate the effect of breast tissue more specifically, by cropping out the breasts from recordings, finding that this does not resolve the observed performance gaps. In conclusion, our results indicate that dataset-specific factors, not fundamental physiological differences, are the main drivers of male--female performance gaps in chest X-ray analyses on widely used NIH and CheXpert Dataset.

Kilian Zepf, Eike Petersen, Jes Frellsen et al.

Segmentation uncertainty models predict a distribution over plausible segmentations for a given input, which they learn from the annotator variation in the training set. However, in practice these annotations can differ systematically in the way they are generated, for example through the use of different labeling tools. This results in datasets that contain both data variability and differing label styles. In this paper, we demonstrate that applying state-of-the-art segmentation uncertainty models on such datasets can lead to model bias caused by the different label styles. We present an updated modelling objective conditioning on labeling style for aleatoric uncertainty estimation, and modify two state-of-the-art-architectures for segmentation uncertainty accordingly. We show with extensive experiments that this method reduces label style bias, while improving segmentation performance, increasing the applicability of segmentation uncertainty models in the wild. We curate two datasets, with annotations in different label styles, which we will make publicly available along with our code upon publication.

Nina Weng, Paraskevas Pegios, Eike Petersen et al.

Shortcut learning is when a model -- e.g. a cardiac disease classifier -- exploits correlations between the target label and a spurious shortcut feature, e.g. a pacemaker, to predict the target label based on the shortcut rather than real discriminative features. This is common in medical imaging, where treatment and clinical annotations correlate with disease labels, making them easy shortcuts to predict disease. We propose a novel detection and quantification of the impact of potential shortcut features via a fast diffusion-based counterfactual image generation that can synthetically remove or add shortcuts. Via a novel inpainting-based modification we spatially limit the changes made with no extra inference step, encouraging the removal of spatially constrained shortcut features while ensuring that the shortcut-free counterfactuals preserve their remaining image features to a high degree. Using these, we assess how shortcut features influence model predictions. This is enabled by our second contribution: An efficient diffusion-based counterfactual explanation method with significant inference speed-up at comparable image quality as state-of-the-art. We confirm this on two large chest X-ray datasets, a skin lesion dataset, and CelebA. Our code is publicly available at fastdime.compute.dtu.dk.

Dishantkumar Sutariya, Eike Petersen

Analyzing machine learning model performance stratified by patient and recording properties is becoming the accepted norm and often yields crucial insights about important model failure modes. Performing such analyses in a statistically rigorous manner is non-trivial, however. Appropriate performance metrics must be selected that allow for valid comparisons between groups of different sample sizes and base rates; metric uncertainty must be determined and multiple comparisons be corrected for, in order to assess whether any observed differences may be purely due to chance; and in the case of intersectional analyses, mechanisms must be implemented to find the most `interesting' subgroups within combinatorially many subgroup combinations. We here present a statistical toolbox that addresses these challenges and enables practitioners to easily yet rigorously assess their models for potential subgroup performance disparities. While broadly applicable, the toolbox is specifically designed for medical imaging applications. The analyses provided by the toolbox are illustrated in two case studies, one in skin lesion malignancy classification on the ISIC2020 dataset and one in chest X-ray-based disease classification on the MIMIC-CXR dataset.

Dishantkumar Sutariya, Eike Petersen

Deep learning models can identify racial identity with high accuracy from chest X-ray (CXR) recordings. Thus, there is widespread concern about the potential for racial shortcut learning, where a model inadvertently learns to systematically bias its diagnostic predictions as a function of racial identity. Such racial biases threaten healthcare equity and model reliability, as models may systematically misdiagnose certain demographic groups. Since racial shortcuts are diffuse - non-localized and distributed throughout the whole CXR recording - image preprocessing methods may influence racial shortcut learning, yet the potential of such methods for reducing biases remains underexplored. Here, we investigate the effects of image preprocessing methods including lung masking, lung cropping, and Contrast Limited Adaptive Histogram Equalization (CLAHE). These approaches aim to suppress spurious cues encoding racial information while preserving diagnostic accuracy. Our experiments reveal that simple bounding box-based lung cropping can be an effective strategy for reducing racial shortcut learning while maintaining diagnostic model performance, bypassing frequently postulated fairness-accuracy trade-offs.

Martina Mattioli, Eike Petersen, Aasa Feragen et al.

Artificial intelligence (AI) is being applied in almost every field. At the same time, the currently dominant deep learning methods are fundamentally black-box systems that lack explanations for their inferences, significantly limiting their trustworthiness and adoption. Explainable AI (XAI) methods aim to overcome this challenge by providing explanations of the models' decision process. Such methods are often proposed and developed by engineers and scientists with a predominantly technical background and incorporate their assumptions about the existence, validity, and explanatory utility of different conceivable explanatory mechanisms. However, the basic concept of an explanation -- what it is, whether we can know it, whether it is absolute or relative -- is far from trivial and has been the subject of deep philosophical debate for millennia. As we point out here, the assumptions incorporated into different XAI methods are not harmless and have important consequences for the validity and interpretation of AI explanations in different domains. We investigate ontological and epistemological assumptions in explainability methods when they are applied to AI systems, meaning the assumptions we make about the existence of explanations and our ability to gain knowledge about those explanations. Our analysis shows how seemingly small technical changes to an XAI method may correspond to important differences in the underlying assumptions about explanations. We furthermore highlight the risks of ignoring the underlying onto-epistemological paradigm when choosing an XAI method for a given application, and we discuss how to select and adapt appropriate XAI methods for different domains of application.

Nikolette Pedersen, Regitze Sydendal, Andreas Wulff et al.

Deep learning has been reported to achieve high performances in the detection of skin cancer, yet many challenges regarding the reproducibility of results and biases remain. This study is a replication (different data, same analysis) of a previous study on Alzheimer's disease detection, which studied the robustness of logistic regression (LR) and convolutional neural networks (CNN) across patient sexes. We explore sex bias in skin cancer detection, using the PAD-UFES-20 dataset with LR trained on handcrafted features reflecting dermatological guidelines (ABCDE and the 7-point checklist), and a pre-trained ResNet-50 model. We evaluate these models in alignment with the replicated study: across multiple training datasets with varied sex composition to determine their robustness. Our results show that both the LR and the CNN were robust to the sex distribution, but the results also revealed that the CNN had a significantly higher accuracy (ACC) and area under the receiver operating characteristics (AUROC) for male patients compared to female patients. The data and relevant scripts to reproduce our results are publicly available (https://github.com/ nikodice4/Skin-cancer-detection-sex-bias).

Vincent Olesen, Nina Weng, Aasa Feragen et al.

Machine learning models have achieved high overall accuracy in medical image analysis. However, performance disparities on specific patient groups pose challenges to their clinical utility, safety, and fairness. This can affect known patient groups - such as those based on sex, age, or disease subtype - as well as previously unknown and unlabeled groups. Furthermore, the root cause of such observed performance disparities is often challenging to uncover, hindering mitigation efforts. In this paper, to address these issues, we leverage Slice Discovery Methods (SDMs) to identify interpretable underperforming subsets of data and formulate hypotheses regarding the cause of observed performance disparities. We introduce a novel SDM and apply it in a case study on the classification of pneumothorax and atelectasis from chest x-rays. Our study demonstrates the effectiveness of SDMs in hypothesis formulation and yields an explanation of previously observed but unexplained performance disparities between male and female patients in widely used chest X-ray datasets and models. Our findings indicate shortcut learning in both classification tasks, through the presence of chest drains and ECG wires, respectively. Sex-based differences in the prevalence of these shortcut features appear to cause the observed classification performance gap, representing a previously underappreciated interaction between shortcut learning and model fairness analyses.

Eike Petersen, Enzo Ferrante, Melanie Ganz et al.

Medical imaging models have been shown to encode information about patient demographics such as age, race, and sex in their latent representation, raising concerns about their potential for discrimination. Here, we ask whether requiring models not to encode demographic attributes is desirable. We point out that marginal and class-conditional representation invariance imply the standard group fairness notions of demographic parity and equalized odds, respectively. In addition, however, they require matching the risk distributions, thus potentially equalizing away important group differences. Enforcing the traditional fairness notions directly instead does not entail these strong constraints. Moreover, representationally invariant models may still take demographic attributes into account for deriving predictions, implying unequal treatment - in fact, achieving representation invariance may require doing so. In theory, this can be prevented using counterfactual notions of (individual) fairness or invariance. We caution, however, that properly defining medical image counterfactuals with respect to demographic attributes is fraught with challenges. Finally, we posit that encoding demographic attributes may even be advantageous if it enables learning a task-specific encoding of demographic features that does not rely on social constructs such as 'race' and 'gender.' We conclude that demographically invariant representations are neither necessary nor sufficient for fairness in medical imaging. Models may need to encode demographic attributes, lending further urgency to calls for comprehensive model fairness assessments in terms of predictive performance across diverse patient groups.

Eike Petersen, Yannik Potdevin, Esfandiar Mohammadi et al.

Machine learning is expected to fuel significant improvements in medical care. To ensure that fundamental principles such as beneficence, respect for human autonomy, prevention of harm, justice, privacy, and transparency are respected, medical machine learning systems must be developed responsibly. Many high-level declarations of ethical principles have been put forth for this purpose, but there is a severe lack of technical guidelines explicating the practical consequences for medical machine learning. Similarly, there is currently considerable uncertainty regarding the exact regulatory requirements placed upon medical machine learning systems. This survey provides an overview of the technical and procedural challenges involved in creating medical machine learning systems responsibly and in conformity with existing regulations, as well as possible solutions to address these challenges. First, a brief review of existing regulations affecting medical machine learning is provided, showing that properties such as safety, robustness, reliability, privacy, security, transparency, explainability, and nondiscrimination are all demanded already by existing law and regulations - albeit, in many cases, to an uncertain degree. Next, the key technical obstacles to achieving these desirable properties are discussed, as well as important techniques to overcome these obstacles in the medical context. We notice that distribution shift, spurious correlations, model underspecification, uncertainty quantification, and data scarcity represent severe challenges in the medical context. Promising solution approaches include the use of large and representative datasets and federated learning as a means to that end, the careful exploitation of domain knowledge, the use of inherently transparent models, comprehensive out-of-distribution model testing and verification, as well as algorithmic impact assessments.

Eike Petersen, Christian Hoffmann, Philipp Rostalski

Factor graphs have recently gained increasing attention as a unified framework for representing and constructing algorithms for signal processing, estimation, and control. One capability that does not seem to be well explored within the factor graph tool kit is the ability to handle deterministic nonlinear transformations, such as those occurring in nonlinear filtering and smoothing problems, using tabulated message passing rules. In this contribution, we provide general forward (filtering) and backward (smoothing) approximate Gaussian message passing rules for deterministic nonlinear transformation nodes in arbitrary factor graphs fulfilling a Markov property, based on numerical quadrature procedures for the forward pass and a Rauch-Tung-Striebel-type approximation of the backward pass. These message passing rules can be employed for deriving many algorithms for solving nonlinear problems using factor graphs, as is illustrated by the proposition of a nonlinear modified Bryson-Frazier (MBF) smoother based on the presented message passing rules.

Herbert Buchner, Eike Petersen, Marcus Eger et al.

The electromyogram (EMG) is an important tool for assessing the activity of a muscle and thus also a valuable measure for the diagnosis and control of respiratory support. In this article we propose convolutive blind source separation (BSS) as an effective tool to pre-process surface electromyogram (sEMG) data of the human respiratory muscles. Specifically, the problem of discriminating between inspiratory, expiratory and cardiac muscle activity is addressed, which currently poses a major obstacle for the clinical use of sEMG for adaptive ventilation control. It is shown that using the investigated broadband algorithm, a clear separation of these components can be achieved. The algorithm is based on a generic framework for BSS that utilizes multiple statistical signal characteristics. Apart from a four-channel FIR structure, there are no further restrictive assumptions on the demixing system.