Mikhail Belyaev

Ekaterina Kondrateva, Polina Druzhinina, Alexandra Dalechina et al.

Magnetic resonance imaging (MRI) data is heterogeneous due to differences in device manufacturers, scanning protocols, and inter-subject variability. A conventional way to mitigate MR image heterogeneity is to apply preprocessing transformations such as anatomy alignment, voxel resampling, signal intensity equalization, image denoising, and localization of regions of interest. Although a preprocessing pipeline standardizes image appearance, its influence on the quality of image segmentation and on other downstream tasks in deep neural networks has never been rigorously studied. We conduct experiments on three publicly available datasets and evaluate the effect of different preprocessing steps in intra- and inter-dataset training scenarios. Our results demonstrate that most popular standardization steps add no value to the network performance; moreover, preprocessing can hamper model performance. We suggest that image intensity normalization approaches do not contribute to model accuracy because of the reduction of signal variance with image standardization. Finally, we show that the contribution of skull-stripping in data preprocessing is almost negligible if measured in terms of estimated tumor volume. We show that the only essential transformation for accurate deep learning analysis is the unification of voxel spacing across the dataset. In contrast, inter-subjects anatomy alignment in the form of non-rigid atlas registration is not necessary and intensity equalization steps (denoising, bias-field correction and histogram matching) do not improve models' performance. The study code is accessible online https://github.com/MedImAIR/brain-mri-processing-pipeline

Alexey Zakharov, Maxim Pisov, Alim Bukharaev et al.

Vertebral body compression fractures are early signs of osteoporosis. Though these fractures are visible on Computed Tomography (CT) images, they are frequently missed by radiologists in clinical settings. Prior research on automatic methods of vertebral fracture classification proves its reliable quality; however, existing methods provide hard-to-interpret outputs and sometimes fail to process cases with severe abnormalities such as highly pathological vertebrae or scoliosis. We propose a new two-step algorithm to localize the vertebral column in 3D CT images and then detect individual vertebrae and quantify fractures in 2D simultaneously. We train neural networks for both steps using a simple 6-keypoints based annotation scheme, which corresponds precisely to the current clinical recommendation. Our algorithm has no exclusion criteria, processes 3D CT in 2 seconds on a single GPU, and provides an interpretable and verifiable output. The method approaches expert-level performance and demonstrates state-of-the-art results in vertebrae 3D localization (the average error is 1 mm), vertebrae 2D detection (precision and recall are 0.99), and fracture identification (ROC AUC at the patient level is up to 0.96). Our anchor-free vertebra detection network shows excellent generalizability on a new domain by achieving ROC AUC 0.95, sensitivity 0.85, specificity 0.9 on a challenging VerSe dataset with many unseen vertebra types.

Anton Vasiliuk, Daria Frolova, Mikhail Belyaev et al.

Detecting out-of-distribution (OOD) samples for trusted medical image segmentation remains a significant challenge. The critical issue here is the lack of a strict definition of abnormal data, which often results in artificial problem settings without measurable clinical impact. In this paper, we redesign the OOD detection problem according to the specifics of volumetric medical imaging and related downstream tasks (e.g., segmentation). We propose using the downstream model's performance as a pseudometric between images to define abnormal samples. This approach enables us to weigh different samples based on their performance impact without an explicit ID/OOD distinction. We incorporate this weighting in a new metric called Expected Performance Drop (EPD). EPD is our core contribution to the new problem design, allowing us to rank methods based on their clinical impact. We demonstrate the effectiveness of EPD-based evaluation in 11 CT and MRI OOD detection challenges.

Anton Vasiliuk, Daria Frolova, Mikhail Belyaev et al.

When applying a Deep Learning model to medical images, it is crucial to estimate the model uncertainty. Voxel-wise uncertainty is a useful visual marker for human experts and could be used to improve the model's voxel-wise output, such as segmentation. Moreover, uncertainty provides a solid foundation for out-of-distribution (OOD) detection, improving the model performance on the image-wise level. However, one of the frequent tasks in medical imaging is the segmentation of distinct, local structures such as tumors or lesions. Here, the structure-wise uncertainty allows more precise operations than image-wise and more semantic-aware than voxel-wise. The way to produce uncertainty for individual structures remains poorly explored. We propose a framework to measure the structure-wise uncertainty and evaluate the impact of OOD data on the model performance. Thus, we identify the best UE method to improve the segmentation quality. The proposed framework is tested on three datasets with the tumor segmentation task: LIDC-IDRI, LiTS, and a private one with multiple brain metastases cases.

Stanislav Shimovolos, Andrey Shushko, Mikhail Belyaev et al.

Deep learning methods provide significant assistance in analyzing coronavirus disease (COVID-19) in chest computed tomography (CT) images, including identification, severity assessment, and segmentation. Although the earlier developed methods address the lack of data and specific annotations, the current goal is to build a robust algorithm for clinical use, having a larger pool of available data. With the larger datasets, the domain shift problem arises, affecting the performance of methods on the unseen data. One of the critical sources of domain shift in CT images is the difference in reconstruction kernels used to generate images from the raw data (sinograms). In this paper, we show a decrease in the COVID-19 segmentation quality of the model trained on the smooth and tested on the sharp reconstruction kernels. Furthermore, we compare several domain adaptation approaches to tackle the problem, such as task-specific augmentation and unsupervised adversarial learning. Finally, we propose the unsupervised adaptation method, called F-Consistency, that outperforms the previous approaches. Our method exploits a set of unlabeled CT image pairs which differ only in reconstruction kernels within every pair. It enforces the similarity of the network hidden representations (feature maps) by minimizing mean squared error (MSE) between paired feature maps. We show our method achieving 0.64 Dice Score on the test dataset with unseen sharp kernels, compared to the 0.56 Dice Score of the baseline model. Moreover, F-Consistency scores 0.80 Dice Score between predictions on the paired images, which almost doubles the baseline score of 0.46 and surpasses the other methods. We also show F-Consistency to better generalize on the unseen kernels and without the specific semantic content, e.g., presence of the COVID-19 lesions.

Daria Frolova, Anton Vasiliuk, Mikhail Belyaev et al.

Deep Learning (DL) models tend to perform poorly when the data comes from a distribution different from the training one. In critical applications such as medical imaging, out-of-distribution (OOD) detection helps to identify such data samples, increasing the model's reliability. Recent works have developed DL-based OOD detection that achieves promising results on 2D medical images. However, scaling most of these approaches on 3D images is computationally intractable. Furthermore, the current 3D solutions struggle to achieve acceptable results in detecting even synthetic OOD samples. Such limited performance might indicate that DL often inefficiently embeds large volumetric images. We argue that using the intensity histogram of the original CT or MRI scan as embedding is descriptive enough to run OOD detection. Therefore, we propose a histogram-based method that requires no DL and achieves almost perfect results in this domain. Our proposal is supported two-fold. We evaluate the performance on the publicly available datasets, where our method scores 1.0 AUROC in most setups. And we score second in the Medical Out-of-Distribution challenge without fine-tuning and exploiting task-specific knowledge. Carefully discussing the limitations, we conclude that our method solves the sample-level OOD detection on 3D medical images in the current setting.

Boris Shirokikh, Anvar Kurmukov, Mariia Donskova et al.

Domain shift presents a significant challenge in applying Deep Learning to the segmentation of 3D medical images from sources like Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). Although numerous Domain Adaptation methods have been developed to address this issue, they are often evaluated under impractical data shift scenarios. Specifically, the medical imaging datasets used are often either private, too small for robust training and evaluation, or limited to single or synthetic tasks. To overcome these limitations, we introduce a M3DA /"mEd@/ benchmark comprising four publicly available, multiclass segmentation datasets. We have designed eight domain pairs featuring diverse and practically relevant distribution shifts. These include inter-modality shifts between MRI and CT and intra-modality shifts among various MRI acquisition parameters, different CT radiation doses, and presence/absence of contrast enhancement in images. Within the proposed benchmark, we evaluate more than ten existing domain adaptation methods. Our results show that none of them can consistently close the performance gap between the domains. For instance, the most effective method reduces the performance gap by about 62% across the tasks. This highlights the need for developing novel domain adaptation algorithms to enhance the robustness and scalability of deep learning models in medical imaging. We made our M3DA benchmark publicly available: https://github.com/BorisShirokikh/M3DA.

Boris Shirokikh, Alexey Shevtsov, Anvar Kurmukov et al.

Target imbalance affects the performance of recent deep learning methods in many medical image segmentation tasks. It is a twofold problem: class imbalance - positive class (lesion) size compared to negative class (non-lesion) size; lesion size imbalance - large lesions overshadows small ones (in the case of multiple lesions per image). While the former was addressed in multiple works, the latter lacks investigation. We propose a loss reweighting approach to increase the ability of the network to detect small lesions. During the learning process, we assign a weight to every image voxel. The assigned weights are inversely proportional to the lesion volume, thus smaller lesions get larger weights. We report the benefit from our method for well-known loss functions, including Dice Loss, Focal Loss, and Asymmetric Similarity Loss. Additionally, we compare our results with other reweighting techniques: Weighted Cross-Entropy and Generalized Dice Loss. Our experiments show that inverse weighting considerably increases the detection quality, while preserves the delineation quality on a state-of-the-art level. We publish a complete experimental pipeline for two publicly available datasets of CT images: LiTS and LUNA16 (https://github.com/neuro-ml/inverse_weighting). We also show results on a private database of MR images for the task of multiple brain metastases delineation.

Aleksei Leonov, Aleksei Zakharov, Sergey Koshelev et al.

Automatic ribs segmentation and numeration can increase computed tomography assessment speed and reduce radiologists mistakes. We introduce a model for multilabel ribs segmentation with hierarchical loss function, which enable to improve multilabel segmentation quality. Also we propose postprocessing technique to further increase labeling quality. Our model achieved new state-of-the-art 98.2% label accuracy on public RibSeg v2 dataset, surpassing previous result by 6.7%.

Boris Shirokikh, Alexandra Dalechina, Alexey Shevtsov et al.

We systematically evaluate a Deep Learning (DL) method in a 3D medical image segmentation task. Our segmentation method is integrated into the radiosurgery treatment process and directly impacts the clinical workflow. With our method, we address the relative drawbacks of manual segmentation: high inter-rater contouring variability and high time consumption of the contouring process. The main extension over the existing evaluations is the careful and detailed analysis that could be further generalized on other medical image segmentation tasks. Firstly, we analyze the changes in the inter-rater detection agreement. We show that the segmentation model reduces the ratio of detection disagreements from 0.162 to 0.085 (p < 0.05). Secondly, we show that the model improves the inter-rater contouring agreement from 0.845 to 0.871 surface Dice Score (p < 0.05). Thirdly, we show that the model accelerates the delineation process in between 1.6 and 2.0 times (p < 0.05). Finally, we design the setup of the clinical experiment to either exclude or estimate the evaluation biases, thus preserve the significance of the results. Besides the clinical evaluation, we also summarize the intuitions and practical ideas for building an efficient DL-based model for 3D medical image segmentation.

Talgat Saparov, Anvar Kurmukov, Boris Shirokikh et al.

Domain shift is one of the most salient challenges in medical computer vision. Due to immense variability in scanners' parameters and imaging protocols, even images obtained from the same person and the same scanner could differ significantly. We address variability in computed tomography (CT) images caused by different convolution kernels used in the reconstruction process, the critical domain shift factor in CT. The choice of a convolution kernel affects pixels' granularity, image smoothness, and noise level. We analyze a dataset of paired CT images, where smooth and sharp images were reconstructed from the same sinograms with different kernels, thus providing identical anatomy but different style. Though identical predictions are desired, we show that the consistency, measured as the average Dice between predictions on pairs, is just 0.54. We propose Filtered Back-Projection Augmentation (FBPAug), a simple and surprisingly efficient approach to augment CT images in sinogram space emulating reconstruction with different kernels. We apply the proposed method in a zero-shot domain adaptation setup and show that the consistency boosts from 0.54 to 0.92 outperforming other augmentation approaches. Neither specific preparation of source domain data nor target domain data is required, so our publicly released FBPAug can be used as a plug-and-play module for zero-shot domain adaptation in any CT-based task.

Daniil Stepanov, Marat Akhin, Mikhail Belyaev

Kotlin is a relatively new programming language from JetBrains: its development started in 2010 with release 1.0 done in early 2016. The Kotlin compiler, while slowly and steadily becoming more and more mature, still crashes from time to time on the more tricky input programs, not least because of the complexity of its features and their interactions. This makes it a great target for fuzzing, even the basic forms of which can find a significant number of Kotlin compiler crashes. There is a problem with fuzzing, however, closely related to the cause of the crashes: generating a random, non-trivial and semantically valid Kotlin program is hard. In this paper, we talk about type-centric compiler fuzzing in the form of type-centric enumeration, an approach inspired by skeletal program enumeration and based on a combination of generative and mutation-based fuzzing, which solves this problem by focusing on program types. After creating the skeleton program, we fill the typed holes with fragments of suitable type, created via generation and enhanced by semantic-aware mutation. We implemented this approach in our Kotlin compiler fuzzing framework called Backend Bug Finder (BBF) and did an extensive evaluation, not only testing the real-world feasibility of our approach, but also comparing it to other compiler fuzzing techniques. The results show our approach to be significantly better compared to other fuzzing approaches at generating semantically valid Kotlin programs, while creating more interesting crash-inducing inputs at the same time. We managed to find more than 50 previously unknown compiler crashes, of which 18 were considered important after their triage by the compiler team.

Boris Shirokikh, Ivan Zakazov, Alexey Chernyavskiy et al.

MRI scans appearance significantly depends on scanning protocols and, consequently, the data-collection institution. These variations between clinical sites result in dramatic drops of CNN segmentation quality on unseen domains. Many of the recently proposed MRI domain adaptation methods operate with the last CNN layers to suppress domain shift. At the same time, the core manifestation of MRI variability is a considerable diversity of image intensities. We hypothesize that these differences can be eliminated by modifying the first layers rather than the last ones. To validate this simple idea, we conducted a set of experiments with brain MRI scans from six domains. Our results demonstrate that 1) domain-shift may deteriorate the quality even for a simple brain extraction segmentation task (surface Dice Score drops from 0.85-0.89 even to 0.09); 2) fine-tuning of the first layers significantly outperforms fine-tuning of the last layers in almost all supervised domain adaptation setups. Moreover, fine-tuning of the first layers is a better strategy than fine-tuning of the whole network, if the amount of annotated data from the new domain is strictly limited.

Mikhail Goncharov, Maxim Pisov, Alexey Shevtsov et al.

The current COVID-19 pandemic overloads healthcare systems, including radiology departments. Though several deep learning approaches were developed to assist in CT analysis, nobody considered study triage directly as a computer science problem. We describe two basic setups: Identification of COVID-19 to prioritize studies of potentially infected patients to isolate them as early as possible; Severity quantification to highlight studies of severe patients and direct them to a hospital or provide emergency medical care. We formalize these tasks as binary classification and estimation of affected lung percentage. Though similar problems were well-studied separately, we show that existing methods provide reasonable quality only for one of these setups. We employ a multitask approach to consolidate both triage approaches and propose a convolutional neural network to combine all available labels within a single model. In contrast with the most popular multitask approaches, we add classification layers to the most spatially detailed upper part of U-Net instead of the bottom, less detailed latent representation. We train our model on approximately 2000 publicly available CT studies and test it with a carefully designed set consisting of 32 COVID-19 studies, 30 cases with bacterial pneumonia, 31 healthy patients, and 30 patients with other lung pathologies to emulate a typical patient flow in an out-patient hospital. The proposed multitask model outperforms the latent-based one and achieves ROC AUC scores ranging from 0.87+-01 (bacterial pneumonia) to 0.97+-01 (healthy controls) for Identification of COVID-19 and 0.97+-01 Spearman Correlation for Severity quantification. We release all the code and create a public leaderboard, where other community members can test their models on our test dataset.

Maxim Pisov, Vladimir Kondratenko, Alexey Zakharov et al.

Vertebral body compression fractures are reliable early signs of osteoporosis. Though these fractures are visible on Computed Tomography (CT) images, they are frequently missed by radiologists in clinical settings. Prior research on automatic methods of vertebral fracture classification proves its reliable quality; however, existing methods provide hard-to-interpret outputs and sometimes fail to process cases with severe abnormalities such as highly pathological vertebrae or scoliosis. We propose a new two-step algorithm to localize the vertebral column in 3D CT images and then to simultaneously detect individual vertebrae and quantify fractures in 2D. We train neural networks for both steps using a simple 6-keypoints based annotation scheme, which corresponds precisely to current medical recommendation. Our algorithm has no exclusion criteria, processes 3D CT in 2 seconds on a single GPU, and provides an intuitive and verifiable output. The method approaches expert-level performance and demonstrates state-of-the-art results in vertebrae 3D localization (the average error is 1 mm), vertebrae 2D detection (precision is 0.99, recall is 1), and fracture identification (ROC AUC at the patient level is 0.93).

Artem Pimkin, Alexander Samoylenko, Natalia Antipina et al.

Recent CT Metal Artifacts Reduction (MAR) methods are often based on image-to-image convolutional neural networks for adjustment of corrupted sinograms or images themselves. In this paper, we are exploring the capabilities of a multi-domain method which consists of both sinogram correction (projection domain step) and restored image correction (image-domain step). Moreover, we propose a formulation of the first step problem as sinogram inpainting which allows us to use methods of this specific field such as partial convolutions. The proposed method allows to achieve state-of-the-art (-75% MSE) improvement in comparison with a classic benchmark - Li-MAR.

Boris Shirokikh, Alexandra Dalechina, Alexey Shevtsov et al.

Stereotactic radiosurgery is a minimally-invasive treatment option for a large number of patients with intracranial tumors. As part of the therapy treatment, accurate delineation of brain tumors is of great importance. However, slice-by-slice manual segmentation on T1c MRI could be time-consuming (especially for multiple metastases) and subjective. In our work, we compared several deep convolutional networks architectures and training procedures and evaluated the best model in a radiation therapy department for three types of brain tumors: meningiomas, schwannomas and multiple brain metastases. The developed semiautomatic segmentation system accelerates the contouring process by 2.2 times on average and increases inter-rater agreement from 92.0% to 96.5%.

Maxim Pisov, Mikhail Goncharov, Nadezhda Kurochkina et al.

Midline shift (MLS) is a well-established factor used for outcome prediction in traumatic brain injury, stroke and brain tumors. The importance of automatic estimation of MLS was recently highlighted by ACR Data Science Institute. In this paper we introduce a novel deep learning based approach for the problem of MLS detection, which exploits task-specific structural knowledge. We evaluate our method on a large dataset containing heterogeneous images with significant MLS and show that its mean error approaches the inter-expert variability. Finally, we show the robustness of our approach by validating it on an external dataset, acquired during routine clinical practice.