Many variables of interest in clinical medicine, like disease severity, are recorded using discrete ordinal categories such as normal/mild/moderate/severe. These labels are used to train and evaluate disease severity prediction models. However, ordinal categories represent a simplification of an underlying continuous severity spectrum. Using continuous scores instead of ordinal categories is more sensitive to detecting small changes in disease severity over time. Here, we present a generalized framework that accurately predicts continuously valued variables using only discrete ordinal labels during model development. We found that for three clinical prediction tasks, models that take the ordinal relationship of the training labels into account outperformed conventional multi-class classification models. Particularly the continuous scores generated by ordinal classification and regression models showed a significantly higher correlation with expert rankings of disease severity and lower mean squared errors compared to the multi-class classification models. Furthermore, the use of MC dropout significantly improved the ability of all evaluated deep learning approaches to predict continuously valued scores that truthfully reflect the underlying continuous target variable. We showed that accurate continuously valued predictions can be generated even if the model development only involves discrete ordinal labels. The novel framework has been validated on three different clinical prediction tasks and has proven to bridge the gap between discrete ordinal labels and the underlying continuously valued variables.
Medical tasks are prone to inter-rater variability due to multiple factors such as image quality, professional experience and training, or guideline clarity. Training deep learning networks with annotations from multiple raters is a common practice that mitigates the model's bias towards a single expert. Reliable models generating calibrated outputs and reflecting the inter-rater disagreement are key to the integration of artificial intelligence in clinical practice. Various methods exist to take into account different expert labels. We focus on comparing three label fusion methods: STAPLE, average of the rater's segmentation, and random sampling of each rater's segmentation during training. Each label fusion method is studied using both the conventional training framework and the recently published SoftSeg framework that limits information loss by treating the segmentation task as a regression. Our results, across 10 data splittings on two public datasets, indicate that SoftSeg models, regardless of the ground truth fusion method, had better calibration and preservation of the inter-rater rater variability compared with their conventional counterparts without impacting the segmentation performance. Conventional models, i.e., trained with a Dice loss, with binary inputs, and sigmoid/softmax final activate, were overconfident and underestimated the uncertainty associated with inter-rater variability. Conversely, fusing labels by averaging with the SoftSeg framework led to underconfident outputs and overestimation of the rater disagreement. In terms of segmentation performance, the best label fusion method was different for the two datasets studied, indicating this parameter might be task-dependent. However, SoftSeg had segmentation performance systematically superior or equal to the conventionally trained models and had the best calibration and preservation of the inter-rater variability.
The integration of artificial intelligence into clinical workflows requires reliable and robust models. Repeatability is a key attribute of model robustness. Repeatable models output predictions with low variation during independent tests carried out under similar conditions. During model development and evaluation, much attention is given to classification performance while model repeatability is rarely assessed, leading to the development of models that are unusable in clinical practice. In this work, we evaluate the repeatability of four model types (binary classification, multi-class classification, ordinal classification, and regression) on images that were acquired from the same patient during the same visit. We study the performance of binary, multi-class, ordinal, and regression models on four medical image classification tasks from public and private datasets: knee osteoarthritis, cervical cancer screening, breast density estimation, and retinopathy of prematurity. Repeatability is measured and compared on ResNet and DenseNet architectures. Moreover, we assess the impact of sampling Monte Carlo dropout predictions at test time on classification performance and repeatability. Leveraging Monte Carlo predictions significantly increased repeatability for all tasks on the binary, multi-class, and ordinal models leading to an average reduction of the 95\% limits of agreement by 16% points and of the disagreement rate by 7% points. The classification accuracy improved in most settings along with the repeatability. Our results suggest that beyond about 20 Monte Carlo iterations, there is no further gain in repeatability. In addition to the higher test-retest agreement, Monte Carlo predictions were better calibrated which leads to output probabilities reflecting more accurately the true likelihood of being correctly classified.
The integration of artificial intelligence into clinical workflows requires reliable and robust models. Among the main features of robustness is repeatability. Much attention is given to classification performance without assessing the model repeatability, leading to the development of models that turn out to be unusable in practice. In this work, we evaluate the repeatability of four model types on images from the same patient that were acquired during the same visit. We study the performance of binary, multi-class, ordinal, and regression models on three medical image analysis tasks: cervical cancer screening, breast density estimation, and retinopathy of prematurity classification. Moreover, we assess the impact of sampling Monte Carlo dropout predictions at test time on classification performance and repeatability. Leveraging Monte Carlo predictions significantly increased repeatability for all tasks on the binary, multi-class, and ordinal models leading to an average reduction of the 95% limits of agreement by 17% points.
Deep learning has the potential to augment many components of the clinical workflow, such as medical image interpretation. However, the translation of these black box algorithms into clinical practice has been marred by the relative lack of transparency compared to conventional machine learning methods, hindering in clinician trust in the systems for critical medical decision-making. Specifically, common deep learning approaches do not have intuitive ways of expressing uncertainty with respect to cases that might require further human review. Furthermore, the possibility of algorithmic bias has caused hesitancy regarding the use of developed algorithms in clinical settings. To these ends, we explore how conformal methods can complement deep learning models by providing both clinically intuitive way (by means of confidence prediction sets) of expressing model uncertainty as well as facilitating model transparency in clinical workflows. In this paper, we conduct a field survey with clinicians to assess clinical use-cases of conformal predictions. Next, we conduct experiments with a mammographic breast density and dermatology photography datasets to demonstrate the utility of conformal predictions in "rule-in" and "rule-out" disease scenarios. Further, we show that conformal predictors can be used to equalize coverage with respect to patient demographics such as race and skin tone. We find that a conformal predictions to be a promising framework with potential to increase clinical usability and transparency for better collaboration between deep learning algorithms and clinicians.
As machine learning (ML) continue to be integrated into healthcare systems that affect clinical decision making, new strategies will need to be incorporated in order to effectively detect and evaluate subgroup disparities to ensure accountability and generalizability in clinical workflows. In this paper, we explore how epistemic uncertainty can be used to evaluate disparity in patient demographics (race) and data acquisition (scanner) subgroups for breast density assessment on a dataset of 108,190 mammograms collected from 33 clinical sites. Our results show that even if aggregate performance is comparable, the choice of uncertainty quantification metric can significantly the subgroup level. We hope this analysis can promote further work on how uncertainty can be leveraged to increase transparency of machine learning applications for clinical deployment.
Medical images are often accompanied by metadata describing the image (vendor, acquisition parameters) and the patient (disease type or severity, demographics, genomics). This metadata is usually disregarded by image segmentation methods. In this work, we adapt a linear conditioning method called FiLM (Feature-wise Linear Modulation) for image segmentation tasks. This FiLM adaptation enables integrating metadata into segmentation models for better performance. We observed an average Dice score increase of 5.1% on spinal cord tumor segmentation when incorporating the tumor type with FiLM. The metadata modulates the segmentation process through low-cost affine transformations applied on feature maps which can be included in any neural network's architecture. Additionally, we assess the relevance of segmentation FiLM layers for tackling common challenges in medical imaging: training with limited or unbalanced number of annotated data, multi-class training with missing segmentations, and model adaptation to multiple tasks. Our results demonstrated the following benefits of FiLM for segmentation: FiLMed U-Net was robust to missing labels and reached higher Dice scores with few labels (up to 16.7%) compared to single-task U-Net. The code is open-source and available at www.ivadomed.org.
Spinal cord tumors lead to neurological morbidity and mortality. Being able to obtain morphometric quantification (size, location, growth rate) of the tumor, edema, and cavity can result in improved monitoring and treatment planning. Such quantification requires the segmentation of these structures into three separate classes. However, manual segmentation of 3-dimensional structures is time-consuming and tedious, motivating the development of automated methods. Here, we tailor a model adapted to the spinal cord tumor segmentation task. Data were obtained from 343 patients using gadolinium-enhanced T1-weighted and T2-weighted MRI scans with cervical, thoracic, and/or lumbar coverage. The dataset includes the three most common intramedullary spinal cord tumor types: astrocytomas, ependymomas, and hemangioblastomas. The proposed approach is a cascaded architecture with U-Net-based models that segments tumors in a two-stage process: locate and label. The model first finds the spinal cord and generates bounding box coordinates. The images are cropped according to this output, leading to a reduced field of view, which mitigates class imbalance. The tumor is then segmented. The segmentation of the tumor, cavity, and edema (as a single class) reached 76.7 $\pm$ 1.5% of Dice score and the segmentation of tumors alone reached 61.8 $\pm$ 4.0% Dice score. The true positive detection rate was above 87% for tumor, edema, and cavity. To the best of our knowledge, this is the first fully automatic deep learning model for spinal cord tumor segmentation. The multiclass segmentation pipeline is available in the Spinal Cord Toolbox (https://spinalcordtoolbox.com/). It can be run with custom data on a regular computer within seconds.