Jayashree Kalpathy–Cramer

Guillermo Lorenzo, Syed Rakin Ahmed, David A. Hormuth et al.

Despite the remarkable advances in cancer diagnosis, treatment, and management that have occurred over the past decade, malignant tumors remain a major public health problem. Further progress in combating cancer may be enabled by personalizing the delivery of therapies according to the predicted response for each individual patient. The design of personalized therapies requires patient-specific information integrated into an appropriate mathematical model of tumor response. A fundamental barrier to realizing this paradigm is the current lack of a rigorous, yet practical, mathematical theory of tumor initiation, development, invasion, and response to therapy. In this review, we begin by providing an overview of different approaches to modeling tumor growth and treatment, including mechanistic as well as data-driven models based on ``big data" and artificial intelligence. Next, we present illustrative examples of mathematical models manifesting their utility and discussing the limitations of stand-alone mechanistic and data-driven models. We further discuss the potential of mechanistic models for not only predicting, but also optimizing response to therapy on a patient-specific basis. We then discuss current efforts and future possibilities to integrate mechanistic and data-driven models. We conclude by proposing five fundamental challenges that must be addressed to fully realize personalized care for cancer patients driven by computational models.

Nishanth Arun, Nathan Gaw, Praveer Singh et al.

Saliency maps have become a widely used method to make deep learning models more interpretable by providing post-hoc explanations of classifiers through identification of the most pertinent areas of the input medical image. They are increasingly being used in medical imaging to provide clinically plausible explanations for the decisions the neural network makes. However, the utility and robustness of these visualization maps has not yet been rigorously examined in the context of medical imaging. We posit that trustworthiness in this context requires 1) localization utility, 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. Using the localization information available in two large public radiology datasets, we quantify the performance of eight commonly used saliency map approaches for the above criteria using area under the precision-recall curves (AUPRC) and structural similarity index (SSIM), comparing their performance to various baseline measures. Using our framework to quantify the trustworthiness of saliency maps, we show that all eight saliency map techniques fail at least one of the criteria and are, in most cases, less trustworthy when compared to the baselines. We suggest that their usage in the high-risk domain of medical imaging warrants additional scrutiny and recommend that detection or segmentation models be used if localization is the desired output of the network. Additionally, to promote reproducibility of our findings, we provide the code we used for all tests performed in this work at this link: https://github.com/QTIM-Lab/Assessing-Saliency-Maps.

Tiago Gonçalves, Dagoberto Pulido-Arias, Julian Willett et al.

The interactions between tumor cells and the tumor microenvironment (TME) dictate therapeutic efficacy of radiation and many systemic therapies in breast cancer. However, to date, there is not a widely available method to reproducibly measure tumor and immune phenotypes for each patient's tumor. Given this unmet clinical need, we applied multiple instance learning (MIL) algorithms to assess activity of ten biologically relevant pathways from the hematoxylin and eosin (H&E) slide of primary breast tumors. We employed different feature extraction approaches and state-of-the-art model architectures. Using binary classification, our models attained area under the receiver operating characteristic (AUROC) scores above 0.70 for nearly all gene expression pathways and on some cases, exceeded 0.80. Attention maps suggest that our trained models recognize biologically relevant spatial patterns of cell sub-populations from H&E. These efforts represent a first step towards developing computational H&E biomarkers that reflect facets of the TME and hold promise for augmenting precision oncology.

Kendall Schmidt, Benjamin Bearce, Ken Chang et al.

The correct interpretation of breast density is important in the assessment of breast cancer risk. AI has been shown capable of accurately predicting breast density, however, due to the differences in imaging characteristics across mammography systems, models built using data from one system do not generalize well to other systems. Though federated learning (FL) has emerged as a way to improve the generalizability of AI without the need to share data, the best way to preserve features from all training data during FL is an active area of research. To explore FL methodology, the breast density classification FL challenge was hosted in partnership with the American College of Radiology, Harvard Medical School's Mass General Brigham, University of Colorado, NVIDIA, and the National Institutes of Health National Cancer Institute. Challenge participants were able to submit docker containers capable of implementing FL on three simulated medical facilities, each containing a unique large mammography dataset. The breast density FL challenge ran from June 15 to September 5, 2022, attracting seven finalists from around the world. The winning FL submission reached a linear kappa score of 0.653 on the challenge test data and 0.413 on an external testing dataset, scoring comparably to a model trained on the same data in a central location.

Andreanne Lemay, Katharina Hoebel, Christopher P. Bridge et al.

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.

Aaron S. Coyner, Praveer Singh, James M. Brown et al.

Background: Artificial intelligence (AI) may demonstrate racial bias when skin or choroidal pigmentation is present in medical images. Recent studies have shown that convolutional neural networks (CNNs) can predict race from images that were not previously thought to contain race-specific features. We evaluate whether grayscale retinal vessel maps (RVMs) of patients screened for retinopathy of prematurity (ROP) contain race-specific features. Methods: 4095 retinal fundus images (RFIs) were collected from 245 Black and White infants. A U-Net generated RVMs from RFIs, which were subsequently thresholded, binarized, or skeletonized. To determine whether RVM differences between Black and White eyes were physiological, CNNs were trained to predict race from color RFIs, raw RVMs, and thresholded, binarized, or skeletonized RVMs. Area under the precision-recall curve (AUC-PR) was evaluated. Findings: CNNs predicted race from RFIs near perfectly (image-level AUC-PR: 0.999, subject-level AUC-PR: 1.000). Raw RVMs were almost as informative as color RFIs (image-level AUC-PR: 0.938, subject-level AUC-PR: 0.995). Ultimately, CNNs were able to detect whether RFIs or RVMs were from Black or White babies, regardless of whether images contained color, vessel segmentation brightness differences were nullified, or vessel segmentation widths were normalized. Interpretation: AI can detect race from grayscale RVMs that were not thought to contain racial information. Two potential explanations for these findings are that: retinal vessels physiologically differ between Black and White babies or the U-Net segments the retinal vasculature differently for various fundus pigmentations. Either way, the implications remain the same: AI algorithms have potential to demonstrate racial bias in practice, even when preliminary attempts to remove such information from the underlying images appear to be successful.

Charles Lu, Ken Chang, Praveer Singh et al.

Despite the intense attention and considerable investment into clinical machine learning research, relatively few applications have been deployed at a large-scale in a real-world clinical environment. While research is important in advancing the state-of-the-art, translation is equally important in bringing these techniques and technologies into a position to ultimately impact healthcare. We believe a lack of appreciation for several considerations are a major cause for this discrepancy between expectation and reality. To better characterize a holistic perspective among researchers and practitioners, we survey several practitioners with commercial experience in developing CML for clinical deployment. Using these insights, we identify several main categories of challenges in order to better design and develop clinical machine learning applications.

Charles Lu, Andreanne Lemay, Katharina Hoebel et al.

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.

Ujjwal Baid, Satyam Ghodasara, Suyash Mohan et al.

The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data. Gliomas are the most common primary malignancies of the central nervous system, with varying degrees of aggressiveness and prognosis. The RSNA-ASNR-MICCAI BraTS 2021 challenge targets the evaluation of computational algorithms assessing the same tumor compartmentalization, as well as the underlying tumor's molecular characterization, in pre-operative baseline mpMRI data from 2,040 patients. Specifically, the two tasks that BraTS 2021 focuses on are: a) the segmentation of the histologically distinct brain tumor sub-regions, and b) the classification of the tumor's O[6]-methylguanine-DNA methyltransferase (MGMT) promoter methylation status. The performance evaluation of all participating algorithms in BraTS 2021 will be conducted through the Sage Bionetworks Synapse platform (Task 1) and Kaggle (Task 2), concluding in distributing to the top ranked participants monetary awards of $60,000 collectively.

Sharut Gupta, Praveer Singh, Ken Chang et al.

Model brittleness is a primary concern when deploying deep learning models in medical settings owing to inter-institution variations, like patient demographics and intra-institution variation, such as multiple scanner types. While simply training on the combined datasets is fraught with data privacy limitations, fine-tuning the model on subsequent institutions after training it on the original institution results in a decrease in performance on the original dataset, a phenomenon called catastrophic forgetting. In this paper, we investigate trade-off between model refinement and retention of previously learned knowledge and subsequently address catastrophic forgetting for the assessment of mammographic breast density. More specifically, we propose a simple yet effective approach, adapting Elastic weight consolidation (EWC) using the global batch normalization (BN) statistics of the original dataset. The results of this study provide guidance for the deployment of clinical deep learning models where continuous learning is needed for domain expansion.

Mehak Aggarwal, Nishanth Arun, Sharut Gupta et al.

While success of Deep Learning (DL) in automated diagnosis can be transformative to the medicinal practice especially for people with little or no access to doctors, its widespread acceptability is severely limited by inherent black-box decision making and unsafe failure modes. While saliency methods attempt to tackle this problem in non-medical contexts, their apriori explanations do not transfer well to medical usecases. With this study we validate a model design element agnostic to both architecture complexity and model task, and show how introducing this element gives an inherently self-explanatory model. We compare our results with state of the art non-trainable saliency maps on RSNA Pneumonia Dataset and demonstrate a much higher localization efficacy using our adopted technique. We also compare, with a fully supervised baseline and provide a reasonable alternative to it's high data labelling overhead. We further investigate the validity of our claims through qualitative evaluation from an expert reader.

Nishanth Thumbavanam Arun, Nathan Gaw, Praveer Singh et al.

Saliency maps have become a widely used method to assess which areas of the input image are most pertinent to the prediction of a trained neural network. However, in the context of medical imaging, there is no study to our knowledge that has examined the efficacy of these techniques and quantified them using overlap with ground truth bounding boxes. In this work, we explored the credibility of the various existing saliency map methods on the RSNA Pneumonia dataset. We found that GradCAM was the most sensitive to model parameter and label randomization, and was highly agnostic to model architecture.

Maarten G. Poirot, Praneeth Vepakomma, Ken Chang et al.

Shortage of labeled data has been holding the surge of deep learning in healthcare back, as sample sizes are often small, patient information cannot be shared openly, and multi-center collaborative studies are a burden to set up. Distributed machine learning methods promise to mitigate these problems. We argue for a split learning based approach and apply this distributed learning method for the first time in the medical field to compare performance against (1) centrally hosted and (2) non collaborative configurations for a range of participants. Two medical deep learning tasks are used to compare split learning to conventional single and multi center approaches: a binary classification problem of a data set of 9000 fundus photos, and multi-label classification problem of a data set of 156,535 chest X-rays. The several distributed learning setups are compared for a range of 1-50 distributed participants. Performance of the split learning configuration remained constant for any number of clients compared to a single center study, showing a marked difference compared to the non collaborative configuration after 2 clients (p < 0.001) for both sets. Our results affirm the benefits of collaborative training of deep neural networks in health care. Our work proves the significant benefit of distributed learning in healthcare, and paves the way for future real-world implementations.

Katharina Hoebel, Ken Chang, Jay Patel et al.

Segmentation tasks in medical imaging are inherently ambiguous: the boundary of a target structure is oftentimes unclear due to image quality and biological factors. As such, predicted segmentations from deep learning algorithms are inherently ambiguous. Additionally, "ground truth" segmentations performed by human annotators are in fact weak labels that further increase the uncertainty of outputs of supervised models developed on these manual labels. To date, most deep learning segmentation studies utilize predicted segmentations without uncertainty quantification. In contrast, we explore the use of Monte Carlo dropout U-Nets for the segmentation with additional quantification of segmentation uncertainty. We assess the utility of three measures of uncertainty (Coefficient of Variation, Mean Pairwise Dice, and Mean Voxelwise Uncertainty) for the segmentation of a less ambiguous target structure (liver) and a more ambiguous one (liver tumors). Furthermore, we assess how the utility of these measures changes with different patch sizes and cost functions. Our results suggest that models trained using larger patches and the weighted categorical cross-entropy as cost function allow the extraction of more meaningful uncertainty measures compared to smaller patches and soft dice loss. Among the three uncertainty measures Mean Pairwise Dice shows the strongest correlation with segmentation quality. Our study serves as a proof-of-concept of how uncertainty measures can be used to assess the quality of a predicted segmentation, potentially serving to flag low quality segmentations from a given model for further human review.

Vivek Sharma, Praneeth Vepakomma, Tristan Swedish et al.

Recently, there has been the development of Split Learning, a framework for distributed computation where model components are split between the client and server (Vepakomma et al., 2018b). As Split Learning scales to include many different model components, there needs to be a method of matching client-side model components with the best server-side model components. A solution to this problem was introduced in the ExpertMatcher (Sharma et al., 2019) framework, which uses autoencoders to match raw data to models. In this work, we propose an extension of ExpertMatcher, where matching can be performed without the need to share the client's raw data representation. The technique is applicable to situations where there are local clients and centralized expert ML models, but the sharing of raw data is constrained.

Vivek Sharma, Praneeth Vepakomma, Tristan Swedish et al.

In this work we introduce ExpertMatcher, a method for automating deep learning model selection using autoencoders. Specifically, we are interested in performing inference on data sources that are distributed across many clients using pretrained expert ML networks on a centralized server. The ExpertMatcher assigns the most relevant model(s) in the central server given the client's data representation. This allows resource-constrained clients in developing countries to utilize the most relevant ML models for their given task without having to evaluate the performance of each ML model. The method is generic and can be beneficial in any setup where there are local clients and numerous centralized expert ML models.

Jin Chao, Ahmad Al Badawi, Balagopal Unnikrishnan et al.

Convolutional neural networks (CNNs) have enabled significant performance leaps in medical image classification tasks. However, translating neural network models for clinical applications remains challenging due to data privacy issues. Fully Homomorphic Encryption (FHE) has the potential to address this challenge as it enables the use of CNNs on encrypted images. However, current HE technology poses immense computational and memory overheads, particularly for high-resolution images such as those seen in the clinical context. We present CaRENets: Compact and Resource-Efficient CNNs for high performance and resource-efficient inference on high-resolution encrypted images in practical applications. At the core, CaRENets comprises a new FHE compact packing scheme that is tightly integrated with CNN functions. CaRENets offers dual advantages of memory efficiency (due to compact packing of images and CNN activations) and inference speed (due to the reduction in the number of ciphertexts created and the associated mathematical operations) over standard interleaved packing schemes. We apply CaRENets to perform homomorphic abnormality detection with 80-bit security level in two clinical conditions - Retinopathy of Prematurity (ROP) and Diabetic Retinopathy (DR). The ROP dataset comprises 96 x 96 grayscale images, while the DR dataset comprises 256 x 256 RGB images. We demonstrate over 45x improvement in memory efficiency and 4-5x speedup in inference over the interleaved packing schemes. As our approach enables memory-efficient low-latency HE inference without imposing additional communication burden, it has implications for practical and secure deep learning inference in clinical imaging.

Yuan Guo, Jennifer Dy, Deniz Erdogmus et al.

Pairwise comparison labels are more informative and less variable than class labels, but generating them poses a challenge: their number grows quadratically in the dataset size. We study a natural experimental design objective, namely, D-optimality, that can be used to identify which $K$ pairwise comparisons to generate. This objective is known to perform well in practice, and is submodular, making the selection approximable via the greedy algorithm. A naïve greedy implementation has $O(N^2d^2K)$ complexity, where $N$ is the dataset size, $d$ is the feature space dimension, and $K$ is the number of generated comparisons. We show that, by exploiting the inherent geometry of the dataset--namely, that it consists of pairwise comparisons--the greedy algorithm's complexity can be reduced to $O(N^2(K+d)+N(dK+d^2) +d^2K).$ We apply the same acceleration also to the so-called lazy greedy algorithm. When combined, the above improvements lead to an execution time of less than 1 hour for a dataset with $10^8$ comparisons; the naïve greedy algorithm on the same dataset would require more than 10 days to terminate.

Bruno Lecouat, Ken Chang, Chuan-Sheng Foo et al.

Supervised deep learning algorithms have enabled significant performance gains in medical image classification tasks. But these methods rely on large labeled datasets that require resource-intensive expert annotation. Semi-supervised generative adversarial network (GAN) approaches offer a means to learn from limited labeled data alongside larger unlabeled datasets, but have not been applied to discern fine-scale, sparse or localized features that define medical abnormalities. To overcome these limitations, we propose a patch-based semi-supervised learning approach and evaluate performance on classification of diabetic retinopathy from funduscopic images. Our semi-supervised approach achieves high AUC with just 10-20 labeled training images, and outperforms the supervised baselines by upto 15% when less than 30% of the training dataset is labeled. Further, our method implicitly enables interpretation of the SSL predictions. As this approach enables good accuracy, resolution and interpretability with lower annotation burden, it sets the pathway for scalable applications of deep learning in clinical imaging.

Szu-Yeu Hu, Andrew Beers, Ken Chang et al.

In this paper, we propose a new pre-training scheme for U-net based image segmentation. We first train the encoding arm as a localization network to predict the center of the target, before extending it into a U-net architecture for segmentation. We apply our proposed method to the problem of segmenting the optic disc from fundus photographs. Our work shows that the features learned by encoding arm can be transferred to the segmentation network to reduce the annotation burden. We propose that an approach could have broad utility for medical image segmentation, and alleviate the burden of delineating complex structures by pre-training on annotations that are much easier to acquire.

Ken Chang, Niranjan Balachandar, Carson K Lam et al.

Deep learning has become a promising approach for automated medical diagnoses. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In such cases, sharing a deep learning model is a more attractive alternative. The best method of performing such a task is unclear, however. In this study, we simulate the dissemination of learning deep learning network models across four institutions using various heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in three independent image collections (retinal fundus photos, mammography, and ImageNet). We find that cyclical weight transfer resulted in a performance (testing accuracy = 77.3%) that was closest to that of centrally hosted patient data (testing accuracy = 78.7%). We also found that there is an improvement in the performance of cyclical weight transfer heuristic with high frequency of weight transfer.

Andrew Beers, Ken Chang, James Brown et al.

Deep learning has quickly become the weapon of choice for brain lesion segmentation. However, few existing algorithms pre-configure any biological context of their chosen segmentation tissues, and instead rely on the neural network's optimizer to develop such associations de novo. We present a novel method for applying deep neural networks to the problem of glioma tissue segmentation that takes into account the structured nature of gliomas - edematous tissue surrounding mutually-exclusive regions of enhancing and non-enhancing tumor. We trained multiple deep neural networks with a 3D U-Net architecture in a tree structure to create segmentations for edema, non-enhancing tumor, and enhancing tumor regions. Specifically, training was configured such that the whole tumor region including edema was predicted first, and its output segmentation was fed as input into separate models to predict enhancing and non-enhancing tumor. Our method was trained and evaluated on the publicly available BraTS dataset, achieving Dice scores of 0.882, 0.732, and 0.730 for whole tumor, enhancing tumor and tumor core respectively.

Cagdas Ulas, Christine Preibisch, Jonathan Sperl et al.

Perfusion-weighted magnetic resonance imaging (MRI) is an imaging technique that allows one to measure tissue perfusion in an organ of interest through the injection of an intravascular paramagnetic contrast agent (CA). Due to a preference for high temporal and spatial resolution in many applications, this modality could significantly benefit from accelerated data acquisitions. In this paper, we specifically address the problem of reconstructing perfusion MR image series from a subset of k-space data. Our proposed approach is motivated by the observation that temporal variations (dynamics) in perfusion imaging often exhibit correlation across different spatial scales. Hence, we propose a model that jointly penalizes the voxel-wise deviations in temporal gradient images obtained based on a baseline, and the patch-wise dissimilarities between the spatio-temporal neighborhoods of entire image sequence. We validate our method on dynamic susceptibility contrast (DSC)-MRI and dynamic contrast-enhanced (DCE)-MRI brain perfusion datasets acquired from 10 tumor patients in total. We provide extensive analysis of reconstruction performance and perfusion parameter estimation in comparison to state-of-the-art reconstruction methods. Experimental results on clinical datasets demonstrate that our reconstruction model can potentially achieve up to 8-fold acceleration by enabling accurate estimation of perfusion parameters while preserving spatial image details and reconstructing the complete perfusion time-intensity curves (TICs).

Allan Hanbury, Henning Müller, Krisztian Balog et al.

Evaluation in empirical computer science is essential to show progress and assess technologies developed. Several research domains such as information retrieval have long relied on systematic evaluation to measure progress: here, the Cranfield paradigm of creating shared test collections, defining search tasks, and collecting ground truth for these tasks has persisted up until now. In recent years, however, several new challenges have emerged that do not fit this paradigm very well: extremely large data sets, confidential data sets as found in the medical domain, and rapidly changing data sets as often encountered in industry. Also, crowdsourcing has changed the way that industry approaches problem-solving with companies now organizing challenges and handing out monetary awards to incentivize people to work on their challenges, particularly in the field of machine learning. This white paper is based on discussions at a workshop on Evaluation-as-a-Service (EaaS). EaaS is the paradigm of not providing data sets to participants and have them work on the data locally, but keeping the data central and allowing access via Application Programming Interfaces (API), Virtual Machines (VM) or other possibilities to ship executables. The objective of this white paper are to summarize and compare the current approaches and consolidate the experiences of these approaches to outline the next steps of EaaS, particularly towards sustainable research infrastructures. This white paper summarizes several existing approaches to EaaS and analyzes their usage scenarios and also the advantages and disadvantages. The many factors influencing EaaS are overviewed, and the environment in terms of motivations for the various stakeholders, from funding agencies to challenge organizers, researchers and participants, to industry interested in supplying real-world problems for which they require solutions.