Localized Energy-Based Normalization of Medical Images: Application to Chest Radiography

A Simple Normalization Technique Using Window Statistics to Improve the Out-of-Distribution Generalization on Medical Images

Since data scarcity and data heterogeneity are prevailing for medical images, well-trained Convolutional Neural Networks (CNNs) using previous normalization methods may perform poorly when deployed to a new site. However, a reliable model for real-world clinical applications should generalize well both on in-distribution (IND) and out-of-distribution (OOD) data (e.g., the new site data). In this study, we present a novel normalization technique called window normalization (WIN) to improve the model generalization on heterogeneous medical images, which offers a simple yet effective alternative to existing normalization methods. Specifically, WIN perturbs the normalizing statistics with the local statistics computed within a window. This feature-level augmentation technique regularizes the models well and improves their OOD generalization significantly. Leveraging its advantage, we propose a novel self-distillation method called WIN-WIN. WIN-WIN can be easily implemented with two forward passes and a consistency constraint, serving as a simple extension to existing methods. Extensive experimental results on various tasks (6 tasks) and datasets (24 datasets) demonstrate the generality and effectiveness of our methods.

Stroke Lesion Segmentation and Deep Learning: A Comprehensive Review

Stroke is a medical condition that affects around 15 million people annually. Patients and their families can face severe financial and emotional challenges as it can cause motor, speech, cognitive, and emotional impairments. Stroke lesion segmentation identifies the stroke lesion visually while providing useful anatomical information. Though different computer-aided software are available for manual segmentation, state-of-the-art deep learning makes the job much easier. This review paper explores the different deep-learning-based lesion segmentation models and the impact of different pre-processing techniques on their performance. It aims to provide a comprehensive overview of the state-of-the-art models and aims to guide future research and contribute to the development of more robust and effective stroke lesion segmentation models.

A systematic approach to deep learning-based nodule detection in chest radiographs

Lung cancer is a serious disease responsible for millions of deaths every year. Early stages of lung cancer can be manifested in pulmonary lung nodules. To assist radiologists in reducing the number of overseen nodules and to increase the detection accuracy in general, automatic detection algorithms have been proposed. Particularly, deep learning methods are promising. However, obtaining clinically relevant results remains challenging. While a variety of approaches have been proposed for general purpose object detection, these are typically evaluated on benchmark data sets. Achieving competitive performance for specific real-world problems like lung nodule detection typically requires careful analysis of the problem at hand and the selection and tuning of suitable deep learning models. We present a systematic comparison of state-of-the-art object detection algorithms for the task of lung nodule detection. In this regard, we address the critical aspect of class imbalance and and demonstrate a data augmentation approach as well as transfer learning to boost performance. We illustrate how this analysis and a combination of multiple architectures results in state-of-the-art performance for lung nodule detection, which is demonstrated by the proposed model winning the detection track of the Node21 competition. The code for our approach is available at https://github.com/FinnBehrendt/node21-submit.

Contrastive self-supervised learning from 100 million medical images with optional supervision

Purpose

Building accurate and robust artificial intelligence systems for medical image assessment requires the creation of large sets of annotated training examples. However, constructing such datasets is very costly due to the complex nature of annotation tasks, which often require expert knowledge (e.g., a radiologist). To counter this limitation, we propose a method to learn from medical images at scale in a self-supervised way.

Approach

Our approach, based on contrastive learning and online feature clustering, leverages training datasets of over 100,000,000 medical images of various modalities, including radiography, computed tomography (CT), magnetic resonance (MR) imaging, and ultrasonography (US). We propose to use the learned features to guide model training in supervised and hybrid self-supervised/supervised regime on various downstream tasks.

Results

We highlight a number of advantages of this strategy on challenging image assessment problems in radiography, CT, and MR: (1) significant increase in accuracy compared to the state-of-the-art (e.g., area under the curve boost of 3% to 7% for detection of abnormalities from chest radiography scans and hemorrhage detection on brain CT); (2) acceleration of model convergence during training by up to 85% compared with using no pretraining (e.g., 83% when training a model for detection of brain metastases in MR scans); and (3) increase in robustness to various image augmentations, such as intensity variations, rotations or scaling reflective of data variation seen in the field.

Conclusions

The proposed approach enables large gains in accuracy and robustness on challenging image assessment problems. The improvement is significant compared with other state-of-the-art approaches trained on medical or vision images (e.g., ImageNet).

Machine Learning Algorithm Validation: From Essentials to Advanced Applications and Implications for Regulatory Certification and Deployment

The deployment of machine learning (ML) models in the health care domain can increase the speed and accuracy of diagnosis and improve treatment planning and patient care. Translating academic research to applications that are deployable in clinical settings requires the ability to generalize and high reproducibility, which are contingent on a rigorous and sound methodology for the development and evaluation of ML models. This article describes the fundamental concepts and processes for ML model evaluation and highlights common workflows. It concludes with a discussion of the requirements for the deployment of ML models in clinical settings.

Automated Detection and Quantification of COVID-19 Airspace Disease on Chest Radiographs: A Novel Approach Achieving Expert Radiologist-Level Performance Using a Deep Convolutional Neural Network Trained on Digital Reconstructed Radiographs From Computed Tomography-Derived Ground Truth

OBJECTIVES

The aim of this study was to leverage volumetric quantification of airspace disease (AD) derived from a superior modality (computed tomography [CT]) serving as ground truth, projected onto digitally reconstructed radiographs (DRRs) to (1) train a convolutional neural network (CNN) to quantify AD on paired chest radiographs (CXRs) and CTs, and (2) compare the DRR-trained CNN to expert human readers in the CXR evaluation of patients with confirmed COVID-19.

MATERIALS AND METHODS

We retrospectively selected a cohort of 86 COVID-19 patients (with positive reverse transcriptase-polymerase chain reaction test results) from March to May 2020 at a tertiary hospital in the northeastern United States, who underwent chest CT and CXR within 48 hours. The ground-truth volumetric percentage of COVID-19-related AD (POv) was established by manual AD segmentation on CT. The resulting 3-dimensional masks were projected into 2-dimensional anterior-posterior DRR to compute area-based AD percentage (POa). A CNN was trained with DRR images generated from a larger-scale CT dataset of COVID-19 and non-COVID-19 patients, automatically segmenting lungs, AD, and quantifying POa on CXR. The CNN POa results were compared with POa quantified on CXR by 2 expert readers and to the POv ground truth, by computing correlations and mean absolute errors.

RESULTS

Bootstrap mean absolute error and correlations between POa and POv were 11.98% (11.05%-12.47%) and 0.77 (0.70-0.82) for average of expert readers and 9.56% to 9.78% (8.83%-10.22%) and 0.78 to 0.81 (0.73-0.85) for the CNN, respectively.

CONCLUSIONS

Our CNN trained with DRR using CT-derived airspace quantification achieved expert radiologist level of accuracy in the quantification of AD on CXR in patients with positive reverse transcriptase-polymerase chain reaction test results for COVID-19.

Anatomic Point-Based Lung Region with Zone Identification for Radiologist Annotation and Machine Learning for Chest Radiographs

Our objective is to investigate the reliability and usefulness of anatomic point-based lung zone segmentation on chest radiographs (CXRs) as a reference standard framework and to evaluate the accuracy of automated point placement. Two hundred frontal CXRs were presented to two radiologists who identified five anatomic points: two at the lung apices, one at the top of the aortic arch, and two at the costophrenic angles. Of these 1000 anatomic points, 161 (16.1%) were obscured (mostly by pleural effusions). Observer variations were investigated. Eight anatomic zones then were automatically generated from the manually placed anatomic points, and a prototype algorithm was developed using the point-based lung zone segmentation to detect cardiomegaly and levels of diaphragm and pleural effusions. A trained U-Net neural network was used to automatically place these five points within 379 CXRs of an independent database. Intra- and inter-observer variation in mean distance between corresponding anatomic points was larger for obscured points (8.7 mm and 20 mm, respectively) than for visible points (4.3 mm and 7.6 mm, respectively). The computer algorithm using the point-based lung zone segmentation could diagnostically measure the cardiothoracic ratio and diaphragm position or pleural effusion. The mean distance between corresponding points placed by the radiologist and by the neural network was 6.2 mm. The network identified 95% of the radiologist-indicated points with only 3% of network-identified points being false-positives. In conclusion, a reliable anatomic point-based lung segmentation method for CXRs has been developed with expected utility for establishing reference standards for machine learning applications.

Robust classification from noisy labels: Integrating additional knowledge for chest radiography abnormality assessment

Chest radiography is the most common radiographic examination performed in daily clinical practice for the detection of various heart and lung abnormalities. The large amount of data to be read and reported, with more than 100 studies per day for a single radiologist, poses a challenge in consistently maintaining high interpretation accuracy. The introduction of large-scale public datasets has led to a series of novel systems for automated abnormality classification. However, the labels of these datasets were obtained using natural language processed medical reports, yielding a large degree of label noise that can impact the performance. In this study, we propose novel training strategies that handle label noise from such suboptimal data. Prior label probabilities were measured on a subset of training data re-read by 4 board-certified radiologists and were used during training to increase the robustness of the training model to the label noise. Furthermore, we exploit the high comorbidity of abnormalities observed in chest radiography and incorporate this information to further reduce the impact of label noise. Additionally, anatomical knowledge is incorporated by training the system to predict lung and heart segmentation, as well as spatial knowledge labels. To deal with multiple datasets and images derived from various scanners that apply different post-processing techniques, we introduce a novel image normalization strategy. Experiments were performed on an extensive collection of 297,541 chest radiographs from 86,876 patients, leading to a state-of-the-art performance level for 17 abnormalities from 2 datasets. With an average AUC score of 0.880 across all abnormalities, our proposed training strategies can be used to significantly improve performance scores.

COVID-19 on Chest Radiographs: A Multireader Evaluation of an Artificial Intelligence System

Background Chest radiography may play an important role in triage for coronavirus disease 2019 (COVID-19), particularly in low-resource settings. Purpose To evaluate the performance of an artificial intelligence (AI) system for detection of COVID-19 pneumonia on chest radiographs. Materials and Methods An AI system (CAD4COVID-XRay) was trained on 24 678 chest radiographs, including 1540 used only for validation while training. The test set consisted of a set of continuously acquired chest radiographs (n = 454) obtained in patients suspected of having COVID-19 pneumonia between March 4 and April 6, 2020, at one center (223 patients with positive reverse transcription polymerase chain reaction [RT-PCR] results, 231 with negative RT-PCR results). Radiographs were independently analyzed by six readers and by the AI system. Diagnostic performance was analyzed with the receiver operating characteristic curve. Results For the test set, the mean age of patients was 67 years ± 14.4 (standard deviation) (56% male). With RT-PCR test results as the reference standard, the AI system correctly classified chest radiographs as COVID-19 pneumonia with an area under the receiver operating characteristic curve of 0.81. The system significantly outperformed each reader (P < .001 using the McNemar test) at their highest possible sensitivities. At their lowest sensitivities, only one reader significantly outperformed the AI system (P = .04). Conclusion The performance of an artificial intelligence system in the detection of coronavirus disease 2019 on chest radiographs was comparable with that of six independent readers. © RSNA, 2020.

Automated abnormality classification of chest radiographs using deep convolutional neural networks

As one of the most ubiquitous diagnostic imaging tests in medical practice, chest radiography requires timely reporting of potential findings and diagnosis of diseases in the images. Automated, fast, and reliable detection of diseases based on chest radiography is a critical step in radiology workflow. In this work, we developed and evaluated various deep convolutional neural networks (CNN) for differentiating between normal and abnormal frontal chest radiographs, in order to help alert radiologists and clinicians of potential abnormal findings as a means of work list triaging and reporting prioritization. A CNN-based model achieved an AUC of 0.9824 ± 0.0043 (with an accuracy of 94.64 ± 0.45%, a sensitivity of 96.50 ± 0.36% and a specificity of 92.86 ± 0.48%) for normal versus abnormal chest radiograph classification. The CNN model obtained an AUC of 0.9804 ± 0.0032 (with an accuracy of 94.71 ± 0.32%, a sensitivity of 92.20 ± 0.34% and a specificity of 96.34 ± 0.31%) for normal versus lung opacity classification. Classification performance on the external dataset showed that the CNN model is likely to be highly generalizable, with an AUC of 0.9444 ± 0.0029. The CNN model pre-trained on cohorts of adult patients and fine-tuned on pediatric patients achieved an AUC of 0.9851 ± 0.0046 for normal versus pneumonia classification. Pretraining with natural images demonstrates benefit for a moderate-sized training image set of about 8500 images. The remarkable performance in diagnostic accuracy observed in this study shows that deep CNNs can accurately and effectively differentiate normal and abnormal chest radiographs, thereby providing potential benefits to radiology workflow and patient care.

Accuracy of an automated system for tuberculosis detection on chest radiographs in high-risk screening

SETTING:

Tuberculosis (TB) screening programmes can be optimised by reducing the number of chest radiographs (CXRs) requiring interpretation by human experts.

OBJECTIVE:

To evaluate the performance of computerised detection software in triaging CXRs in a high-throughput digital mobile TB screening programme.

DESIGN:

A retrospective evaluation of the software was performed on a database of 38 961 postero-anterior CXRs from unique individuals seen between 2005 and 2010, 87 of whom were diagnosed with TB. The software generated a TB likelihood score for each CXR. This score was compared with a reference standard for notified active pulmonary TB using receiver operating characteristic (ROC) curve and localisation ROC (LROC) curve analyses.

RESULTS:

On ROC curve analysis, software specificity was 55.71% (95%CI 55.21–56.20) and negative predictive value was 99.98% (95%CI 99.95–99.99), at a sensitivity of 95%. The area under the ROC curve was 0.90 (95%CI 0.86–0.93). Results of the LROC curve analysis were similar.

CONCLUSION:

The software could identify more than half of the normal images in a TB screening setting while maintaining high sensitivity, and may therefore be used for triage.

Accurate segmentation of inflammatory and abnormal regions using medical thermal imagery

Methodologies reported in the existing literature for identification of a region of interest (ROI) in medical thermograms suffer from over- and under-extraction of the abnormal and/or inflammatory region, thereby causing inaccurate diagnoses of the spread of an abnormality. We overcome this limitation by exploiting the advantages of a logarithmic transformation. Our algorithm extends the conventional region growing segmentation technique with a modified similarity criteria and a stopping rule. In this method, the ROI is generated by taking common information from two independent regions produced by two different versions of a region-growing algorithm that use different parameters. An automatic multi-seed selection procedure prevents missed segmentations in the proposed approach. We validate our technique by experimentation on various thermal images of the inflammation of affected knees and abnormal breasts. The images were obtained from three databases, namely the Knee joint dataset, the DBT-TU-JU dataset, and the DMR-IR dataset. The superiority of the proposed technique is established by comparison to the performance of state-of-the-art competing methodologies. This study performed temperature emitted inflammatory area segmentation on thermal images of knees and breasts. The proposed segmentation method is of potential value in thermal image processing applications that require expediency and automation.

Deep Learning Guided Partitioned Shape Model for Anterior Visual Pathway Segmentation

Analysis of cranial nerve systems, such as the anterior visual pathway (AVP), from MRI sequences is challenging due to their thin long architecture, structural variations along the path, and low contrast with adjacent anatomic structures. Segmentation of a pathologic AVP (e.g., with low-grade gliomas) poses additional challenges. In this work, we propose a fully automated partitioned shape model segmentation mechanism for AVP steered by multiple MRI sequences and deep learning features. Employing deep learning feature representation, this framework presents a joint partitioned statistical shape model able to deal with healthy and pathological AVP. The deep learning assistance is particularly useful in the poor contrast regions, such as optic tracts and pathological areas. Our main contributions are: 1) a fast and robust shape localization method using conditional space deep learning, 2) a volumetric multiscale curvelet transform-based intensity normalization method for robust statistical model, and 3) optimally partitioned statistical shape and appearance models based on regional shape variations for greater local flexibility. Our method was evaluated on MRI sequences obtained from 165 pediatric subjects. A mean Dice similarity coefficient of 0.779 was obtained for the segmentation of the entire AVP (optic nerve only =0.791 ) using the leave-one-out validation. Results demonstrated that the proposed localized shape and sparse appearance-based learning approach significantly outperforms current state-of-the-art segmentation approaches and is as robust as the manual segmentation.